Chapter 11Video SwitchersCONTENT

Chapter 11
Video Switchers
CONTENTS
11.1
Overview
11.2
Background and Evolution
11.2.1
Small System Switchers
11.2.2
Midsize Systems
11.2.3
Analog Matrix Switcher
11.2.4

Digital Matrix Switcher
11.2.5
Virtual Matrix Switcher
11.3
Small Analog System Switcher Types
11.3.1
Manual
11.3.2
Homing Sequential
11.3.3
Bridging Sequential
11.3.4
Looping-Sequential
11.3.5
Alarming
11.3.6
Synchronous, Non-Synchronous Video
Signal
11.3.7
Switcher Choice
11.4
Matrix Switcher
11.4.1
Analog Technology

11.4.2
Matrix Switcher Control Functions
and Features
11.4.3
Multiple Locations
11.4.4
Digital Switcher
11.5
Virtual Matrix Switcher (VMS)
11.5.1
Evolution
11.5.2
Technology
11.5.3
Remote/Multiple Site Monitoring
11.5.4
Features and Advantages
11.6
Summary
11.1 OVERVIEW
The function of the video switcher, matrix video switcher, and
virtual matrix switcher (VMS) in any multiple-camera security

system is to connect any camera to any monitor and display the
video image in a logical sequence. The
switched camera pictures on the monitor can be recorded on a
VCR or DVR, printed on a video printer, or trans-mitted to a
remote site. In both small and large installa-tions, the switcher
component performs a vital function that simplifies system use
and maximizes the information presented to the security
operator. In small security sys-tems that have several cameras
and one or two monitors, a switcher may not be necessary since
all camera scenes can be displayed on several monitors
simultaneously. For a medium or large installation (16, 32
cameras, or more), the number of monitors in the control
console cannot equal the number of cameras, and a one-to-one,
camera-to-monitor correspondence is not practical. Physical
space may be limited, and one security guard may not be able to
view multiple monitors simultaneously. To view multiple
cameras simultaneously on a single monitor, a combiner or
splitter, quad or multiplexer is used (Chapters 12, 16).
Analog matrix video switchers cope with the ever-increasing
size and complexity of video systems and are used in midsize
and large enterprise systems. The essen-tial function of the

matrix switcher system is to switch any combination of cameras
to any combination of monitors, video recorders, video printers,
or transmission channels.
Matrix switchers are based on micro-processor technolo-gies
that allow tremendous flexibility in routing and pro-cessing the
video signals. These switchers come in various forms including
compact, self-contained units that control 16 or 32 cameras, and
multiple monitors and keyboards. These compact dedicated
switchers include such features as text generation and camera
identification. Larger enter-prise systems having hundreds or
thousands of cameras and hundreds of monitors are usually
based on a modu-lar construction and rack-mounted equipment.
Small and midsize microprocessor controlled switching systems
can form the central control center for an integrated secu-rity
and building management system, combining alarm, access
control, fire, and command-and-control functions.
321
322 CCTV Surveillance
An alternative to the self-contained matrix switcher takes the
form of hardware added to a PC system. Some medium-size
systems (16–64 cameras) use PC boards installed in a standard
PC to effect the video switching. The cables for the cameras,
monitors, and any other equip-ment (multiplexers, quads, etc.)
are connected directly to the PC via RS-232 control cables,
simplifying installation and reducing system cost. A
disadvantage of this config-uration is the requirement to
purchase and maintain a PC as compared with the dedicated
microprocessor-based matrix switcher.
In large systems with hundreds of cameras and moni-tors or

those requiring multiple control consoles at dif-ferent sites, the
approach is to use a PC to control the matrix switcher and other
control command functions. The PC is controlled via the
microprocessor keyboards connected to the PC using RS-232 or
RS-485 protocol. These large systems can connect the images
from video cameras to dozens of video monitors, recorders, or
print-ers automatically via the RS-232 communication links.
They are software-programmable and can simultaneously switch
multiple cameras to multiple output devices using salvo
switching techniques. Systems like these are very pow-erful and
have more functions than can be described here.
For these very large systems, the security operator is confronted
with the difficulty of remembering the camera number or the
site at which it is located, and how to con-trol it. To overcome
this problem, a site plan monitor is provided having maps of the
site programmed into it and overlaid with symbols or icons of
the cameras and mon-itoring locations. With these visual
display units (VDU), the operator can select the camera and
area of interest on the map. This is accomplished with input
from a mouse or with the operator’s finger if the monitor has a
touch screen. This is the ultimate in system control for analog
matrix switcher systems. No knowledge of camera or mon-itor
number is necessary, and operating the system is as simple as
touching the site plan touch screen.
In recent years there has been an evolution of the IP-based VMS
that can eliminate some of the shortcomings of the large
expensive analog matrix switcher. The VMS can digitally
multiplex, switch, record, and transmit the camera signals to the
control console monitoring equip-ment and remote locations via
LAN, WAN, and wireless LANs (WiFi).
11.2 BACKGROUND AND EVOLUTION

Up until the last few years legacy CCTV surveillance sys-tems
have used traditional small switchers, multiplexers, and analog
matrix switchers for interconnecting cameras and routing the
video signal to monitors, VCRs, video printers, and in large
systems to some remote sites via dedicated hardware and cable.
These traditional CCTV solutions rely on analog technology and
wired cabling to
transfer the video images from the analog cameras to the
switchers and onto the monitors and video recorders in the
console room. These analog systems are character-ized by long
coaxial video signal and control cable runs, simple analog
switchers, or large analog matrix switchers to display and
record the camera images. The systems have been acceptable in
applications where monitoring and recording was only required
at a central location and monitoring console. They prove to be
expensive when the requirements are for long distances or when
the cameras and console room cross public property or
inaccessible locations within a facility.
Large systems with many cameras and monitors have
incorporated banks of switchers, multiplexers, and large analog
matrix switchers to route the camera signals to the appropriate
monitoring equipment. The DVR technol-ogy has brought a
significant improvement over the ana-log VCR for recording the
video camera images and the ability to distribute them to remote
locations, to archive them, and to provide rapid retrieval of
video image frames recorded at a particular time.
Within the last few years, traditional video CCTV surveil-lance
technology is converging with PC and networking technology.
This convergence has resulted in the evolu-tion of the VMS,
using the digital signal from an IP camera and routing the

digital signal to the console display or any other remote site via
LAN, WAN, and wireless WiFi digital networks. The result is a
dramatic improvement in the features and functionality that can
be delivered to the security operator and management at an
unprecedented price to performance ratio. The digital video
cameras, digital video transmission, and computer networking
tech-nology is now revolutionizing the analog video security
industry. This new digital technology is entirely computer-
based and often uses existing IP infrastructure instead of
requiring a dedicated video cabling.
11.2.1 Small System Switchers
One-on-one Display. For a small video surveillance sys-tem
with perhaps eight cameras, there can be a one-to-one
correspondence between camera and monitor. This means that
each camera can be displayed on a single individual monitor. In
small systems and when the camera-to-monitor distances are
short (a few hundred feet), the switcher and the switching
controls are one and the same and are located at the console. In
installations having larger dis-tances between the cameras and
monitor, the switcher has two separate units with the switcher
located near the camera sites and the switching controls located
near the console monitor.
Increasing the number of cameras makes it difficult for the
operator to effectively view all the monitors and take
appropriate action when necessary. Increasing the num-ber of
cameras requires that the images from more than one camera be
displayed on one monitor. Displaying four
monitors in a quad configuration or 9, 16, or 32 moni-tors on a
single display reduces the number of monitors required. The
sacrifices are that there is a decrease in resolution and the

additional requirement that the oper-ator views many camera
scenes on a single monitor. See Section 11.3 for small video
system switcher types.
11.2.2 Midsize Systems
One-on-one Display vs. Split Screen vs. Sequencing. A midsize
system having multiple cameras and monitors offers the
designer a choice of displaying all the cameras on the monitor
in a one-on-one presentation, or present-ing multiple camera
images on each monitor. When using one monitor it is
impossible for security personnel to observe all camera
locations simultaneously. If a camera switcher is sequencing
from camera to camera there may be a long time delay before a
particular camera is seen again. This can leave a gap in the
security function.
11.2.3 Analog Matrix Switcher
Analog matrix switchers route multiple analog video sources to
multiple video destinations. They can also route audio signals ,
controls, and other functions from cameras to monitors and
analog and digital recorders. The matrix switcher can route
composite video, S-VHS, HDTV, RGB, and other video formats.
However, a signal type that is input can only be routed to an
output of the same sig-nal type. As an example, a composite
video input can only go to a composite video output. The analog
matrix switch is the workhorse of the industry and the most
common interconnect device to connect cameras to monitors,
etc.
11.2.4 Digital Matrix Switcher
Most cameras, monitors, recorders, and other functional
components of analog video systems are now becoming digital
in design. The video system designer has been await-ing the

arrival of a digital solution for the analog matrix switch. The
digital matrix switch would be a digitized video stream routed
to monitors and recorders or other desti-nations in digital form.
The truth of the matter is that this scenario of a fully digital
matrix switch has not proven effective because the digital
matrix switch just doesn’t doenough. It has also not evolved
because of the rapid evo-lution and acceptance of high-speed
digital transmission over LAN and WAN transmission channels,
and the rapid use of the Internet.
11.2.5 Virtual Matrix Switcher
Many of the video surveillance components of analog video
systems have become digital and the security system
Video Switchers
323
designer, integrator, and end user have been awaiting the arrival
of a digital solution for the analog matrix switch. The digital
matrix switch would digitize a video camera stream routed to
the monitor, recorder, or other destina-tion in digital form. The
rapid evolution of high-speed dig-ital transmission over various
transmission channels and the rapid use of the Internet have
effectively bypassed the necessity for the digital video switch.
The VMS technology has effectively skipped the digital matrix
switch and moved directly from the analog matrix switcher to a
VMS that is integrated into the overall security system. The
VMS provides full analog matrix functionality using a standard
matrix keyboard, but takes advantage of the digital video
streams and connections available on LAN, WAN, wireless
networks (WiFi), and the Internet. Section 11.5 describes the

VMS in more detail.
11.3 SMALL ANALOG SYSTEM SWITCHER TYPES
Small to medium video security systems use five basic switcher
types: manual, homing, bridging, looping, and alarming. By
using one or a combination of these switcher types, cameras at
multiple remote sites can be routed to the security console or
multiple monitoring locations for direct observation, recording,
or printing. Most sequential switchers, whether homing,
bridging, looping, or alarm-ing, have a three-position switch for
each camera input. When one of these switches is in the up
position, it is said to be in the Bypass mode. Any of the camera
switches set in this position will cause the switcher to
automatically skip the corresponding camera in the sequential
switch-ing cycle. The center position of these switches is called
Automatic (Auto) mode. Any camera switch in this posi-tion
will cause the switcher to automatically include the
corresponding camera in the normal switching cycle. The down
position of these camera switches can have several different
functions. Where applicable all camera inputs are automatically
electronically terminated in 75 ohms by the switchers. The
following sections describe the unique features of each
switcher.
11.3.1 Manual
The simplest video switcher is the manual switcher, where the
console operator manually chooses one camera from a number
of cameras and displays the video image on a single video
monitor with front panel pushbutton switches, activated
manually by the operator to connect the individ-ual camera to
the monitor. The manual passive switcher uses a simple switch

for contact closure, whereas the manual active switcher uses an
electronic switch. Manual switchers are available to switch from
4 to 32 video cam-eras. Figure 11-1 shows the two types
available: manual passive and manual active.
MANUAL
CAMERA 1
2
3
4
PASSIVE

INTERLOCKED
MECHANICAL
75
75
75
75
75 ohm
SWITCHES

PUSH–BUTTON SWITCHES
MANUAL
CAMERA 1
2
3
4
ACTIVE
o
o
o

ELECTRONIC
ELECTRONIC
ELECTRONIC
ELECTRONIC
ELECTRONIC
SWITCHES
SWITCH
SWITCH
SWITCH
SWITCH

FIGURE 11-1 Manual passive and manual active switchers
11.3.2 Homing Sequential
The homing sequential switcher allows the continu-ous viewing
of any normally sequenced video camera (Figure 11-2). The
camera signal is connected to a single monitor. This switcher
has a three-position switch for each camera: Automatic,
Homing, and Bypass. In the Automatic position, the switcher
automatically selects and switches the video signal from one
camera after another to the monitor according to the sequence
set by the security operator. The length of time each camera
picture is pre-sented on the monitor (dwell time) can be
changed by the operator. The homing sequential switcher
automati-cally sequences from one camera to the next, assuming
the cameras have not been bypassed. When the specific cam-era
control switch is pressed to the Home position, that camera is
continuously displayed on the single monitor and the switching
sequence stops.
Functionally the three-position front-panel switches on the

homing sequential switcher provide three separate camera
display functions: automatic sequencing, bypass, and homing
(select). When a switch is set to Bypass, that particular camera
is not displayed. When the switch is set to Homing, that camera
picture is presented continuously on the monitor and in essence
overrides the automatic
sequencing function. This permits continuous observation of
any particular camera at the operator’s command. In the
Automatic position, all cameras are sequenced onto the monitor,
one at a time.
11.3.3 Bridging Sequential
The bridging sequential switcher operates like the homing
sequential switcher but has the additional feature that two
monitors can display the video cameras. Figure 11-3 shows the
block diagram for a bridging sequential switcher. Moni-tor 1
always displays the cameras selected for sequential view -ing.
Monitor 2 displays only the camera manually selected for
detailed viewing. For instance, pressing the switch for camera 1
to the down position puts the picture on the second or bridged
monitor for detailed viewing, while the sequence of all cameras
not bypassed continues on the first monitor. Monitor 1 sees the
switched sequence of cameras while monitor 2 sees a selected
camera continuously.
The first monitor (the sequential monitor) functions as a homing
sequential switcher. The bridging monitor displays whatever
camera is manually selected. This allows the operator to
maintain a system overview while viewing in detail the camera
covering a scene of particular interest.

Video Switchers
325
CAMERA 1
2
3
4
o
o
o

ELECTRONIC
75
75
75
75
SWITCH
+12 V

AMPLIFIER
ROTATING
SWITCH
AUTOMATIC
A

A
A
BY
B
B
B
1
SELECT
S
S
S
PASS
P

AUTO
AUTO
DWELL
LEVER
TIME
SWITCH

SELECT = HOMING
MONITOR
FIGURE 11-2 Homing sequential switcher
11.3.4 Looping-Sequential
Homing Sequential. The looping-homing sequential switcher
operates like the homing sequential switcher, with the
additional feature that all camera inputs can be brought out to a
second switcher or other device at another location (Figure 11-
4). The switcher has the ability to drive a second switcher,
monitors, recorders, and trans-mission devices for remote
transmission, thereby providing video images at multiple
locations for display or recording. Unlike other switchers, the
looping-homing sequential switcher camera inputs are not

terminated, thereby allow-ing multiple devices to be connected
to the switcher output. For proper operation, one of these
devices, gen-erally the last device in the line, is terminated in a
75-ohm impedance.
Bridging Sequential. The looping-bridging sequential switcher
operates in the same way as the bridging sequential switcher
except that the looping feature is added. As with the looping
homing sequential switcher, the camera inputs are not
terminated in the switcher. Figure 11-5 shows the block diagram
for looping-bridging sequential systems. A looping switcher
provides the
ability to establish two independently controlled loca-tions.
Each station may select any camera for view-ing without
interfering with the operation of the other station.
Remote Sequential. The use of the manual, homing, bridging,
looping, and alarm versions of sequential switch-ers just
described assumes that the distance between the camera locati on
and the monitor (control console) loca-tion is relatively short.
In many installations this is not the case and the cost becomes
prohibitive to provide separate video coaxial cables from each
camera to the distant monitor location. Remote sequential
switchers overcome this problem. The remote sequential
switcher consists of two parts: a control unit and a switching
unit. They are available in all of the aforementioned versions to
provide complete system design flexibility. Both units are con-
nected by means of multi-conductor cables, fiber-optics, a
multiplexed frequency shift key (FSK), or RS-232 commu-
nications system (Figure 11-6).
The control unit is located near the monitor, and the switcher
unit is located closest to the central location of all the cameras.

The physical separation of the switching and control functions
avoids the use of individual camera coax-ial cables to the
control console. Each switcher requires
CAMERA 1
2
3
4
o
o
75

MONITOR
SWITCH
AUTOMATIC
A
A
A

1
SELECT
S
S
S
PASS
P
P
P
5
10
AUTO
AUTO
AUTO
AUTO

DWELL
TIME
SELECT = HOMING
FIGURE 11-3 Bridging sequential switcher

CAMERA 1
2
3
LOOPING
4 HOMING
SWITCHER
MONITOR
LOCATION 1
1

MONITOR
LOCATION 2
2
FIGURE 11-4 Looping homing sequential switcher
Video Switchers
327
CAMERA 1
2
LOCATION 1

3
MONITOR
MONITOR
4
LOOPING
1
2
BRIDGING
UNTERMINATED
75 ohm
75 ohm

SWITCHER
TERMINATION
TERMINATION
MONITOR
LOCATION
2
2

75 ohm
75 ohm
TERMINATION
TERMINATION
FIGURE 11-5 Looping bridging sequential switcher
only one or two coaxial cables for monitor input. The remote
bridging sequential switcher requires two output coaxial cables.
11.3.5 Alarming

An alarming switcher automatically displays a camera image on
to a monitor and/or starts a recorder each time it is activated by
a camera VMD or other alarm input (Figure 11-7). These
switchers are available in homing, bridging, looping-homing,
and looping-bridging configu-rations. When an alarm input
signal is received, a corre-sponding output signal is generated
and transmitted to a monitor, recorder, or printer.
The homing, bridging, and remote sequential switch-ers can be
provided with an alarm feature. In the event of an external
alarm caused by motion in the video pic-ture and detected by a
VMD, an alarm switch closure caused by any type of sensor
input, simple switch clo-sure, IR source, or pressure transducer,
the alarmed cam-era will automatically override the pre-selected
video on the monitor or be automatically displayed on the sec-
ond monitor. When a bridging type switcher is used, the
automatic homing of the alarmed camera overrides any
manually bridged display on the second monitor. The sequence
of all cameras not bypassed continues on the first monitor.
Simultaneously with this switching, an alarm contact within the
switcher closes to operate a recorder, video printer, or any other
alarm-indicating equipment. Auto-matic alarm-programmed
switchers are especially suit-able for applications where
monitors are occasionally unmanned and recorders used to
record abnormal events. They are also particularly useful during
off hours or over weekends when real-time or TL recorders are
used to mon-itor multiple cameras.
The output monitoring device can be a bell, light, or other
signaling unit, which would notify a security guard to dispatch a
guard to the scene or alert a guard at the scene. Even if there

are multiple monitors affording the opportunity to observe all
locations, the use of alarming switchers puts attention in areas
where guard action is really required. The activation of the
alarm signals a sig-nificant occurrence within the field of view
covered by a particular camera.
LOCATION 1
LOCATION 2
CAMERA 1

2
SWITCHER
75 ohm
CONTROL
TERMINATION

UNIT
CONTROL CABLE
1
2
3
4 5
6
7
8
•

TWISTED PAIRS
•
TWO WIRE MULTIPLEXED
FIGURE 11-6 Remote homing sequential switcher
CAMERA 1

MONITOR
SW1
2
1
SW2
75 ohm
3
TERMINATION
SW3
MONITOR

4
SW4
2
UNTERMINATED
EXTERNAL
ALARM
SWITCHES
ALARM
VIDEO

SWITCHED OUTPUT TO
ACTIVATE VCR OR OTHER
DEVICE WHEN ALARM OCCURS
DVR OR VCR
FIGURE 11-7 Alarming bridging sequential switcher
11.3.6 Synchronous, Non-Synchronous Video Signal
There are two types of video signals that are switched:
synchronous and non-synchronous. Synchronous signals lend
themselves to methods of switching where controlled transition
maintains a degree of signal continuity, and pro-vides a clean,
noise-free video picture during switching. Non-synchronous
signals involve the inherent discontinu-ity of timing pulses that
cannot be corrected by special switching methods, and show up
as noise disturbances in the picture. Picture noise in the video
signal takes the form of streaks, a momentary black screen, or
other picture irregularities. When switching composite video
signals, a break may occur during the synchronizing time and
the synchronization signal may be completely lost. This results
in picture rolling or tearing when the picture from the next
camera is displayed on the monitor. The solution to ensure clean
video camera switching is vertical-interval switching. With this
method, the switching is allowed to occur only during the
vertical interval in the video sig-nal between picture frames
(Figure 11-8) while no picture

Video Switchers
329
information is being transmitted. Since no visible monitor
picture is displayed during the vertical-interval switching time,
switching during this period does not cause pic-ture interruption
or deterioration. This technique permits switching from one
camera to the next with no noise or interruption of intelligence.
To understand vertical-interval switching, refer to Figure 11-9.
The camera video signal is generated in the camera sensor. The
horizontal camera clocking signal scans from left to right and
reads out the video image signal representing the light image on
the sensor. When the clocking signal reaches the right side of
the sensor, it returns to the left side and begins another scan.
During the return time in the analog system, the clocking signal
is addressed down to rows of sensor pixels. After it completes
2621/2 scans (one-half of the full frame), the clocking signal
reaches the bottom of the sensor and returns to the top. The
clocking signal then scans the alternate pixel rows and after
completing the second scan the full sensor has been read out.
The return time from the end of the last hori-zontal scan to the
beginning of the first horizontal scan is referred to as the
vertical blanking interval, since during this
PREVIOUS
SWITCHING PULSE

APPROXIMATELY
VIDEO
50 NANOSECOND
SIGNAL
NEXT VIDEO SIGNAL
VIDEO
DURATION
(LAST

(FIRST HORIZONTAL SCAN)
SIGNAL
HORIZONTAL
SCAN)
(X)
FRONT

PORCH
(Y)
BLACK
LEVEL
HORIZONTAL
BACK PORCH
VERTICAL BLANKING (A–B)
SYNC PULSE

(1.1 MILLISECONDS)
VERTICAL
(Y)
SYNC PULSE
(B)

VERTICAL
BLANKING
INTERVAL
PERIOD

EXPANDED
(A)
HORIZONTAL RETURN LINE
HORIZONTAL
(INACTIVE VIDEO)

ACTIVE SCAN LINE
FIGURE 11-8 Vertical interval switching
(A) TWO VIDEO SIGNALS OUT OF VERTICAL PHASE
(B) TWO VIDEO SIGNALS IN VERTICAL PHASE
PHASE
DIFFERENCE
(C) SYNCHRONIZING GENERATOR
(D) CAMERAS WITH PHASE ADJUST

* VERTICAL SYNC PULSES
* EACH CAMERA HAS
SYNC
*
*
PHASE ADJUST
*
GENERATOR
VIDEO
*
VIDEO
*
SWITCHER
SWITCHER

*
*
SYNCHRONIZED
MONITOR
MONITOR
DVR/VCR
SYNCHRONIZED
DVR/VCR
DISPLAY
DISPLAY
FIGURE 11-9 Sequential switching synchronization

time no video signal is generated. In summary, the picture
information occurs during the left-to-right scanning and the
vertical blanking in-between scans.
In the typical video surveillance application the cameras will
not be synchronized. While they may have waveforms or signals
like Figure 11-9a, the time relationship between cameras is not
synchronized or in phase.
Since the synchronization pulses from each camera occur at
different times, when the switcher switches from one camera
signal to the next, a noticeably scrambled or distorted non-
synchronized image occurs as the monitor tries to adjust to the
synchronization pulses of the new signal. A temporarily
distorted picture might be tolerable in some simple direct-
viewing applications, but in situa-tions where there are multiple
cameras or the information is recorded, the result is
unsatisfactory. Since VCR and DVR use the camera
synchronizing pulses to synchronize the machines, it takes many
frames of video for them to synchronize to the new camera
signal. During this inter-val, noise or other artifacts are
generated each time the
switcher is switched. The out-of-phase signals shown in Figure
11-9a are correctable by at least two methods.
One technique for producing in-phase signals is to install a
synchronizing generator that provides a synchro-nizing signal to
the cameras and ensures that they are all in the same phase,
operating at the same frequency, and syn-chronized (Figure 11-
9b). As an alternative, some cameras can be adjusted so that the
phase is the same for each cam-era. Phasing each camera to be
the same does not produce a clean switchover, however. Even

though the signals may be in phase, if the switching occurs
during the video por-tion of the signal, there are visible
transient effects such as spikes and flashes on the monitor or
recorder image. This problem is eliminated by designing the
switcher to switch during the vertical interval (Figure 11-8),
and hence the name vertical-interval switching.
In operation, the switcher circuitry detects the vertical interval
in the signal and delays the actual switchover from one camera
to the next, to the time when vertical blanking is occurring. By
using this method no transient effects
are visible on the monitor or in the recorded image. The
vertical-interval switching technique may not be important in
simple systems, but is extremely important in medium to large
systems, and in any system using a video recorder.
In summary, the quality of switching, or how smoothly (clear,
noiseless picture) the monitor picture from a cam-era 1 can be
switched to camera 2, and so on, is influ-enced by two related
factors: (1) the type of signals to be switched—synchronous or
non-synchronous; and (2) the switching action itself—the time
within the video signal in which the switchover occurs.
11.3.7 Switcher Choice
The following summary suggests which switcher to use in small
system video applications:
· Passive Switcher. The manual switcher is the simplest and can
switch 4, 8, 16, or 32 cameras depending on model, and display
any one of them on a single mon-itor. It is available in either

passive or active type. In a simple application, any one of the
input cameras can be displayed on a single monitor, one at a
time, through manual switching by the security guard.
· Sequential Switcher. The sequential switcher is used when it is
necessary to switch automatically from cam-era to camera so
that the guard can observe all camera scenes sequentially. As in
the manual active switcher, the electronic circuitry provides
fast, clean switching with no transients on the screen, and is
available with camera dwell times of 1 to 50 or 60 seconds
depending on the adjustment made by the operator.
· Homing Sequential Switcher. The homing sequential switcher
has the additional feature of permitting the operator to stop and
look at one particular camera pic-ture continuously or
sequentially and display all the camera pictures with a dwell
time set by the operator. This system permits the operator to
continuously scan through all the cameras and simultaneously
pick out one camera and view it continuously. In the sequen-tial
mode, the dwell time (length of time any particular camera is
viewed) is independently adjustable for each camera. This
provides the operator with the flexibility to view different
camera scenes for different periods of time. The homing
sequential switcher provides the oper-ator with three options
and adjustments: (1) automatic switching, (2) timing, and (3)
bypass control.
· Bridging Sequential Switcher. The bridging sequential
switcher has two separate outputs for two monitors. One output
is for the programmed sequence of cameras; the second is for
the continuous display of a single cam-era. Unlike the homing
sequential switcher, the bridg-ing sequential switcher provides
this constant viewing of a selected input without giving up the
overview of

Video Switchers
331
all the camera scenes provided by the sequential pro-gram. With
the bridging sequential switcher, if the oper-ator wants to
observe a particular camera continuously, the operator moves
the switch to Select, thereby dis-playing that camera picture on
the monitor continu-ously while simultaneously the other
monitor continues to display the sequentially switched camera
sequence, including the camera that is displayed on the second
monitor continuously.
11.4 MATRIX SWITCHER
11.4.1 Analog Technology
Microprocessors, microcomputers, and massive memory solid-
state RAM and magnetic hard drives have revolution-ized the
video security industry. When a security system has many
cameras and monitors and one or more security control consoles
in multiple locations, it becomes more efficient to use a
configurable microprocessor-controlled video switching and
control system called the matrixswitcher (Figure 11-10).
A matrix switcher is a means for selecting an input source such
as video, audio, or control signals and connect-ing them to one
or more outputs. A video matrix switcher is an electronic device
that accepts and distributes video signals selected from multiple
inputs to multiple outputs.
Many manufacturers produce systems that can switch hundreds
(or thousands) of cameras onto hundreds of monitors and
recorders. These systems are built in mod-ular form with

removable PC boards and rack-mounted modules, permitting the
user to begin with a basic sys-tem and expand when necessary.
The removable modules and plug-in units are divided into
several sub-chassis or modules to provide online serviceability
and to reduce or eliminate system downtime. A disadvantage of
these sys-tems is that expansion is in multiples of 8, 16, and 32,
so that if only one or a few new cameras are planned, only the
addition of these larger multiple of cameras is possible.
These switchers have:
· Keyboard and joystick desktop console
· Rack-mounted card cage chassis housing the multiple sub-
modules for the switching and control functions
· Remote modules located near the cameras for driving the
camera, lens, and pan/tilt hardware, as well as for
communicating the information to the control unit
· Power supply.
The initial design of any analog matrix video switching system
should begin with a detail schematic diagram of the proposed
layout showing camera, control locations, and any other
accessory equipment. In addition, a site plan dia-gram should
show the distances between equipments and cable routes since
many equipments are distance-sensitive.
REMOTE CAMERA LOCATIONS

SECURITY CONSOLE ROOM
REMOTE
EIA 19" RACK
MONITOR
CAMERA

DRIVERS
1
CAMERA 1
PAN
COMMUNICATIONS
RS-232, RS-485
TILT

RS-232, RS-485
1
LENS
VIDEO
VIDEO
1
2
SWITCHING

DVR/VCR
POWER
VIDEO
LENS
N
SUPPLY
PRINTER
1

CAMERA PAN/TILT
AUXILIARY KEYBOARD (S)
SECURITY
KEYBOARD

SUPERVISOR
AT OTHER ROMOTE
N
LOGGING
AUXILIARY
LOCATIONS

CAMERA FUNCTIONS CONTROLLED
FEATURES:
• LENS—IRIS, FOCUS, ZOOM, PRESETS
• ALL SWITCHING FUNCTIONS (HOMING, ALARM,
ETC.)
• PAN/TILT—MANUAL, PRESET
• SALVO/BANK SWITCHING (MULTIPLE CAMERAS
SIMULTANEOUSLY)
• TIME/DATE, CAMERA ID NUMBER

• PRESET PAN, TILT, LENS FOCUS, ZOOM, IRIS
•
ALPHA/NUMERIC ANNOTATION
•
COMMUNICATIONS: RS-232, RS-485
• CABLING: UNSHIELDED TWISTED PAIR (UTP)

FIGURE 11-10 Configurable microprocessor controlled video
switching system
The analog matrix control unit contains the system soft-ware
and microprocessor hardware. In some systems, cus-tomized
switching programs are included in the hardware using
electrically programmable memories (EPROM). These solid-
state memory devices allow storage of switch-ing instructions to
be used at a later time when automatic sequencing is desired.
Systems have alpha-numeric charac-ter generators for camera
name and location information or other pertinent data. Matrix
switchers have text anno-tation card providing each video input
with time, date, a three-digit camera ID number, and a multiple-
line user-programmable alpha-numeric message display.
Medium- to large-size matrix systems use RS-232, RS-422 or
RS-485 transmission protocols for controlling cam-era functions
and other output devices. For systems hav-ing up to about 200
cameras and 40 monitors, a single microprocessor-controlled
keyboard has sufficient process-ing power to operate the system
effectively. One or two slave keyboards may also be added if
there is a requirement for more than one person to control the
system. Gener-ally, these large video control switching
functions are kept separate from any other control functions or
other parts
of the security systems such as alarm, access control, fire, and
safety.

Communication from the console to the remote con-trol camera
module is via RS-232 or RS-485 communi-cation protocol.
Distances between the control console and remote console can
be 1000–5000 feet, with the data signal cable a single twisted-
pair, 22-AWG, shielded wire. Most equipment is housed in 5 to
7-inch-high EIA 19-inch rack-mounted modules, thereby
removing most of the electronics from the desktop area except
for the keyboard. Some systems have the ability to connect
several keyboards to the same control system, thereby
permitting control of the system from several locations.
All basic microprocessor-controlled systems have the capability
for manual, homing, looping, sequential, auto alarming,
bridging, and remote switching functions. A unique feature
called salvo switching allows the opera-tor to switch a selected
bank of cameras into a bank of monitors as a synchronized
group with all of the moni-tors switched together in step. The
unique salvo switching feature allows the operator to view all
scenes in one gen-eral area, such as a single floor in a building,
before switching to the next floor. This feature can significantly
increase the monitoring efficiency of the security guard, since it
automatically switches a logical array of cameras.
These systems can provide the same control over alarm
functions as over the video network functions. The alarms are
constantly monitored by the control console. If one or more of
the alarms is activated, the system automat-ically switches in
the camera nearest the alarm and dis-plays its video scene on
the appropriate monitor. The types of alarm sensors
accommodated include switches, infrared sensors (PIR), and
VMDs. Alarm signals can be monitored via an audible tone alert
or visual indicator. Real-time images can be recorded
automatically by hav-ing the recorder switch from TL to real-
time recording mode. The operator has the ability to bypass or

restore cameras and alarms at will. Individual camera dwell
times and sequencing times can be set by the operator on all
cameras.
In large systems, the camera, monitor, recorder, and other
system functions and hardware are programmed into the PC so
that the system can be customized to suit almost any specific
security application. System pass-words are programmed and
lockout tables used to limit access of unauthorized personnel. In
addition to the oper-ational switching sequences normally
entered from the PC keyboard, complex switching sequences
can be pro-grammed off-line using the PC and then downloaded
to the microprocessor control system. Examples of such com-
plex switching include pan/tilt presets for camera point-ing
position, and lens iris, zoom focal length, and focus settings.
These functions are accomplished via receiving modules located
at the camera sites and the RS-232 communica-tions. This
function is accomplished: (1) by the operator selecting a
specific camera and preset number or (2) auto-matically if the
system is preprogrammed, so that when an alarm occurs at a
location in the scene, the camera auto-matically goes to the
preset condition. Simultaneously, a recorder is activated into
real-time mode and records the activity at the designated preset
camera position.
Figure 11-11 illustrates a complete matrix switching sys-tem
used in a large security application having hundreds of cameras
and dozens of monitors, VCRs, DVRs, and print-ers.
All cameras, lenses, pan/tilt platforms, monitors, recorders, and
printers are controlled, monitored, and switched via the central
matrix switcher. The switcher com-municates control functions
to the hardware via RS-232 or RS-485 protocol or time-
multiplexed signals. Video signals from the cameras are
transmitted from the remote loca-tions via individual coaxial,

two-wire, fiber-optic, or wireless channels. The matrix switcher
has a separate video input connector for each camera and a
separate output connec-tor for each monitor, recorder, or printer
device. To bring the matrix switcher and camera and monitoring
equip-ment online, it must first be “configured” or programmed
Video Switchers
333
FIGURE 11-11 Microcomputer video switching systems
according to manufacturer instructions, the hardware con-nected
to it, and the required functioning of the system. This can take
hours or days to accomplish and requires a detailed plan with
methodical procedures. Figure 11-12 shows a block diagram of
a typical video matrix switcher used in a large security
installation.
11.4.2 Matrix Switcher Control Functions and Features
Matrix switchers are supplied with many different control
functions and features. Some of these user-defined and fixed
controls and features are listed below:
· On-Screen Display: Monitors can display alpha-numeric
characters that can be dynamically changed to show camera

information such as video input number and title.
· Auto or Manual Sequencing: Camera tours can be pro-
grammed for any video output. The security operator may define
a dwell time for any video input to create a custom tour.
· Alarm Switching: Alarm inputs can be routed from any input
or group of inputs to any video output from a graphical user
interface (GUI) or PC.
· System Priority: Keyboard users can be assigned differ -ent
levels of security for the control of camera sites. These
different levels of access can be granted based on need to know.
· Camera Numbers: Camera IP numbers and names may be
assigned to cameras in specific areas around in the facility to
better identify camera locations.
· Monitor Numbers: Monitor numbers may be assigned to
monitors in different console rooms at a facility or facilitie s to
identify monitor locations.
· Salvo Switching: Banks of cameras may be switched to a bank
of monitors with one command.
CAMERA 1
MODULE 1

MODULE 1
MONITOR
PAN/
8 CHANNEL
1 CHANNEL
TILT
8
VIDEO INPUT
VIDEO OUTPUT
RECEIVER

DRIVER
9
MODULE 2
MODULE 2
MONITOR
2
8 CHANNEL
1 CHANNEL

CAMERA 8
16
VIDEO INPUT
VIDEO OUTPUT
PAN/

57
MODULE 8
MODULE 8
3
8 CHANNEL
1 CHANNEL

64
VIDEO INPUT
VIDEO OUTPUT
VIDEO
CAMERA 64
VIDEO BUS
CASSETTE
PAN/
RECORDER

TILT
CHARACTER DISPLAY
RECEIVER
VIDEO

SWITCHER
DATA
DRIVER
PRINTER
COMMANDS

CENTRAL
KEYBOARD 1
SIGNAL
REMOTE*
MANCHESTER
PROCESSING
DISTRIBUTION
KEYBOARD 2

COMMANDS
CONVERTER
MODULE
UNIT
COMMUNICATIONS VIA:
POWER
DIGITAL
KEYBOARD 8

MULTIPLEXED
*COAXIAL
SUPPLY
SIGNALS
2–4 WIRE

ALARM
UTP
FIBER OPTICS
INTERFACE

MATRIX SWITCHER
WIRELESS
FIGURE 11-12 Video matrix switcher block diagram
· Partitioning: Password-protected user accounts can be set up
with specific access to cameras, sequence tables, multiplexer
tables, and salvo tables.
· Camera and platform pan, tilt, zoom (PTZ) Control

· Hardware support for RS-232, RS-422, and RS-485.
Some important capabilities and restrictions a video matrix
switcher system should have are:
· Operator should have passwords that allow access to the
system.
· The system should have the capability to limit the num-ber of
system controllers (keyboards, etc.) that a given operator can
log onto.
· The system should have the capability to limit the number of
cameras that can be selected by any given operator.
· The system should have the ability to limit the cameras that
can be selected or controlled from any operator control location.
· The system should have the ability to limit the monitors that
can be viewed from any operator control location.
· The system should have the ability to limit the cameras that
can be shown on any particular monitor.
When given access to the system, the operator should be able to
form the following basic functions:
· Switch video signals to the monitors
· Operate the camera functions such as pan, tilt, zoom, and
focus
· Activate preprogrammed group presets (set groups of cameras
to previously selected positions)

· Activate previously established camera tour sequences
· Acknowledge and reset alarms
· Activate auxiliary contacts
· Access camera-specific features by camera menu.
Selected operators should have the ability to program automated
sequences as described below:
· Group presets: Ability to set up camera preset positions,
including camera to monitor selections.
· Tour sequences: Preprogrammed camera display seq-uences in
both forward and backward direction. Each step of the sequence
consists of the camera num-ber, dwell time, camera position
preset, and auxiliary controlled state.
· Group tour sequences: Multiple camera group presets may be
linked together with a dwell time.
The matrix switcher can control many other video equipments
such as multiplexers, VCRs, DVRs, quads, motion detectors,
and video transmission systems using the RS-232 or other
control signals. These RS-232 ports connected to the matrix
switcher controller generate the commands and appropriate
protocols to operate different functions generated from the keys
on the keyboard.
11.4.3 Multiple Locations
Video security systems are often required for large build-ings

with many floors with separate guard consoles located away
from the main building site or in widely sepa-rated sites.
In large systems with 200 cameras, 40 monitors, or requiring
more than two control consoles at different sites, the general
approach is to use a PC to control the matrix switcher and other
control command functions. The PC is controlled via the
microprocessor keyboard connected to the PC using RS-232 or
RS-485 protocol. Systems like these are very powerful and have
more functions than can be described here. For these large
systems, the operator is confronted with the difficulty of
remembering the num-ber of the camera, the site at which it is
located, and how to control it. To overcome this problem, a site
plan moni-tor is provided that has maps of the site programmed
into it and overlaid with symbols or icons of the cameras and
monitoring locations. With these VDUs the operator sim-ply
selects the area of interest on the map and then selects the
camera to be used. This can be accomplished with input from a
mouse or with a finger if the monitor has a touch screen. This is
the ultimate in system control for analog matrix switcher
systems. No knowledge of camera, monitor, or monitor numbers
is necessary, and operating the system is as simple as touching
the screen.
11.4.4 Digital Switcher
Most of the functional components in legacy analog video
systems are now becoming digitally networked. The video
security industry has been awaiting the arrival of a digi -tal
solution for the analog matrix switch. A digital matrix switch
would digitize a video signal and route the video stream to the
monitor, recorder, or other destination in digital form. The truth
of the matter is that this scenarioof a fully digital matrix switch
has not proven effective because it just doesn’t do enough. It

has also not evolved because of the rapid evolution of high-
speed digital trans-mission over various transmission channels
and the rapid use of the Internet. The technology is effectively
skipping the digital matrix switch. Switching systems are
moving
Video Switchers
335
directly from the analog matrix switch to the VMS that is
integrated into the overall security system.
11.5 VIRTUAL MATRIX SWITCHER (VMS)
11.5.1 Evolution
The VMS provides full analog matrix functionality using a
standard matrix keyboard, but takes advantage of the digital
video streams and connections available on LAN, WAN, WiFi,
and the Internet. The VMS lays a foundation to integrate and
enable the combination of three essential security technologies:
the DVR, the multiplexer, and the matrix switch.
Matrix switching has evolved from: (1) first-generation video
system using a matrix switch, multiplexer and switches, (2)
second-generation matrix switch with DVRs connected to the
intranet or Internet network, (3) local matrix switching
connected to an Ethernet, LAN/TCP/IP switching network, (4)
to a true network-based system using a VMS and all Web-based
cameras connected to an Ethernet, LAN/TCP/IP switching
network with remote access from any location. These four
switching systems are shown in Figure 11-13.
Until recently, traditional analog video matrix systems have

been the dominant method for routing video signals (Figure 11-
14). At the heart of these systems is an analog cross-point
matrix switcher that allows any camera input to be viewed on
any monitor output. The switchers are usu-ally connected to text
generators used to annotate time, date, camera ID, and name
information on the displayed video signal. These digital matrix
switchers are used with a keyboard and GUI and other devices
to control and provide full-featured surveillance functionality.
Legacy analog video systems have some disadvantages in that
the video signals are susceptible to external interfer-ence from
EMI or RFI noise sources. Coax cables carrying video signals
can only be run over distances up to 1000 feet without using
optical-fiber or unshielded twisted-pair (UTP) wiring.
Digital networks, on the other hand, deliver signifi-cant
advantages over analog transmission methods. These include
improved signal integrity over long distances and compatibilit y
with off-the-shelf IT hardware. These net-works allow video
surveillance, access control, alarm, and other functions to be
successfully routed through LAN, WAN and Internet networks.
The network routing of video over these channels is
functionally equivalent to the role of the analog cross-point
matrix switcher in the legacy matrix system, but is instead
distributed through-out the network structure in digital form. In
this digital domain, the network replaces the centralized
hardware switcher and coaxial cables in the matrix system. Only
the keyboards, controller, and text overlays are left intact to
preserve the user experience of the legacy matrix system
1st GENERATION SYSTEM

ANALOG CAMERAS
2nd GENERATION SYSTEM
REMOTE PC
ANALOG CAMERAS
CLIENT
SOFTWARE
FIXED
MONITOR
MONITOR

MULTIPLEXER
FIXED
MONITOR MONITOR
FIXED
MULTIPLEXER
FIXED
MATRIX SWITCH
P/T/Z
VCR
CONTROL
PANEL
KEYBOARD PRINTER
3rd GENERATION SYSTEM
ANALOG CAMERAS

REMOTE PC
CLIENT
FIXED
SOFTWARE
MONITOR
MONITOR
MULTIPLEXER
FIXED
P/T/Z
MATRIX SWITCH

NETWORK CAMERAS
DVR
FIXED
ROUTER/
PRINTER
SERVER
INTERNET/
FIXED
CONTROL
INTRANET
PANEL
P/T/Z

ETHERNET LAN/TCP/IP
NETWORK BASED SYSTEM
SUPPORTS IP CAMERAS
SUPPORTS ANALOG CAMERAS
WEB-BASED REMOTE VIDEO ACCESS
MATRIX SWITCH
P/T/Z
DVR
CONTROL
PANEL
INTERNET/

PRINTER
INTRANET
KEYBOARD
MATRIX, ETC.
REPLACED BY
PC WORKSTATION AND
APPLICATION SOFTWARE
4th GENERATION SYSTEM
ANALOG CAMERAS
REMOTE PC
CLIENT

FIXED
SOFTWARE
FIXED
P/T/Z
NETWORK CAMERAS
FIXED
ROUTER/
SERVER
INTERNET/
FIXED

INTRANET
P/T/Z
ETHERNET LAN/TCP/IP
NETWORK BASED SYSTEM
SUPPORTS IP CAMERAS
SUPPORTS ANALOG CAMERAS
WEB-BASED REMOTE VIDEO ACCESS
FIGURE 11-13 Evolution of the matrix switcher to the virtual
matrix switch
Video Switchers
337
1
2

P/T/Z CONTROL
GENERATORS
ALARM EVENT
MONITOR

CROSS-MATRIX
CAMERA N
INPUTS
SYSTEM CONTROLLER
EVENT RESPONSE

GENERATOR
KEYBOARD:
CONTROL
P/T/Z
FIGURE 11-14 Traditional analog video cross-matrix switch
while still delivering the powerful functionality of a full-
featured matrix switch. For all intents and purposes, the
network represents a cross-point matrix and is in fact a VMS—a
virtual video cross-point matrix.
11.5.2 Technology
The first step in realizing the virtual video matrix is to digitize
the video signal for transmission over the net-work using a
video IP encoder for each analog camera. Figure 11-15 shows
the virtual video matrix in which the video signal has been
digitized for transmission over the network using a video
encoder for each camera.
Internet Protocol cameras already have these encoders built into
them specifically to communicate over these networks. The best
encoders are designed to supply high-efficiency digital MPEG
video streams. Connections for the video and PTZ control
signals from each cam-era are made to the encoder using
standard coaxial serial data wiring. These encoders also have
inputs to support alarm sensor contacts and outputs to control
relays or other alarm annunciation devices. Two-way audio is

also available as an option.
The next step is to connect the encoder to the nearest network
via a Cat-5 or Cat-3 cable. Once video signals are present on the
network, there are a number of important security applications
that are possible.
Several advantages of VMS technology are realized in any
midsize or enterprise security system that is already using
computer hardware. There is no need to purchase and install the
analog matrix switcher. The requirement and expense for
installing coaxial cables or other new wiring throughout a
facility is eliminated. The VMS sys-tem allows the user to
leverage the computer, monitor, and network that already exists
at the facility. Additionally, the hardware is generic so that the
end user maintains flexibility and cost control over any new
critical hardware decisions.
All analog matrix switch systems have costly and cumber-some
scaling limitations. As an example, to add one more monitor to
a 32 monitor system requires the addition of shelves of matrix
switching equipment, since the systems are based on multiples
of 8, 16, and 32 cameras and mon-itors. This is not true of an
integrated software-based VMS system. Only additional user-
licenses and encoders or IP cameras in the exact increment
desired from as small as one to any number of cameras or
monitors are needed.
REMOTE SITE 1

REMOTE SITE 2
REMOTE SITE 3
VIDEO: IP CAMERA,
VIDEO
VIDEO
ANALOG CAMERA
ACCESS CONTROL
ACCESS CONTROL
AND SERVER
COMMUNICATION
COMMUNICATION
OTHER

COMPRESSION: MJPEG
OTHER
MPEG-4
ROUTER/SERVER
ROUTER/SERVER
ROUTER/SERVER
•

NETWORK REPRESENTS
CROSSPOINT MATRIX WITH
ALL DIGITIZED TRANSMISSION
WAN
INTRANET
•
CABLE: COAXIAL
INTERNET
FIBER OPTIC
LAN

WIFI
UTP
CAT 5
VIRTUAL MATRIX SWITCHER COMPUTER/CONTROLLER
EXISTING IT
COMPUTER
TEXT ANNOTATION
NETWORK
KEYBOARD
CAMERA CONTROLLER
P/ T/Z PRESETS

FIGURE 11-15 Virtual video matrix for network transmission
and control
11.5.3 Remote/Multiple Site Monitoring
Enterprise-level systems require customized installations of
cable and hardware entailing significant costs and wiring
needed to bring analog signals back to the control con-sole, not
to mention the distance limitations on these cables. VMS
technology eliminates these costly and time-consuming
demands. If the user needs to move and relo-cate to a new
facility, the cost and miles of wasted coaxial cable represents a
major consideration. VMS technology provides the flexibility to
meet these demands with min-imum cost and time. Upgrading
from an analog to VMS does not extend the availability of video
information. How-ever, when either the analog or a digital
matrix system is integrated into a networked video system, the
VMS system provides wide area connectivity, and video data
becomes available anywhere. The VMS technology allows
organiza-tions to fully leverage their security investment. With
the level of access and functionality provided by the VMS sys -
tem: (1) human resources now have a visitor monitoring system,
(2) operations have the ability to monitor traf-fic in the lobby or
loading dock areas or elsewhere, (3) retail operations can
prevent overloading at cash register lines and can monitor
cashiers, and (4) marketing per-
sonnel can remotely monitor the level of interest shown at
product displays in stores. Many other applications can be cited.

11.5.4 Features and Advantages
Analog video systems require a dedicated wiring and cabling for
each camera. Digital systems using VMS tech-nology require
only Cat-3, Cat-5 cables or digital wireless transmission. The
VMS with built-in DVR capability can record video images
without any degradation loss, sup-port multiple playback, re-
recording, and transmission, and can distribute the video images
to multiple locations. The VMS represents a centralized control
and record-ing ability allowing local monitoring and remote
multiple site viewing. The video camera generates a digital
signal using digital signal processing (DSP) in the camera and
produces a digital signal at the output. It transmits the dig-ital
video signal over the LAN, WAN, or wireless network while
retaining complete integrity and image quality. The VMS
likewise distributes and records the digital signal so that it
remains high quality during switching, reproduc-tion, and
transmission.
The digital video system with VMS produces evidence that has
high integrity. When producing evidence from a standalone
DVR there is no way to verify the actual source of the images as
cameras can be switched on the back of the DVR unit. Using IP
cameras and the VMS, the images are kept under the MAC
address of the specific camera. This is a clear one-to-one
identification of the source of images.
Since the VMS uses off-the-shelf servers, workstations, and
computers, the system can always be upgraded to the latest
hardware for the best price/performance. This is also true of the
IP cameras and other software and hardware that support the
system. The VMS system can integrate existing analog cameras,

infrared (IR) cameras, covert cameras, and of course the IP
camera. The digital technology permits object recognition and
tracking, face
Video Switchers
339
recognition, license plate recognition, direction detection,
people/car counting, etc.
11.6 SUMMARY
The heart of a good security system is a highly functional video
switching control system. In small systems, the switch-ers will
take the form of simple passive, homing, sequential or alarming
switchers. In medium- to large-size systems, the switchers will
take the form of an analog cross-point matrix switcher or a
VMS. During the design phase of any security system,
management and security personnel must decide what
information needs to be displayed, acted upon by the security
operator, recorded, and printed, and choose the switching
system suitable to accomplish the task.

Chapter 12
Quads and Multiplexers
CONTENTS
12.1
Overview
12.2
Background
12.3
Quad Split-Screen Displays
12.3.1
Quad-4 Image

12.3.2
Multi-Image 9, 16, 32
12.4
Multiplexer Technology
12.4.1
Image Rate vs. Number of Cameras
12.4.2
Encoder/Decoder
12.5
Hardware Implementation
12.5.1
Simplex
12.5.2
Duplex/Full Duplex
12.5.3
Triplex
12.6
Recording and Playback
12.6.1
Analog and Digital Recording

12.6.2
Video Playback
12.7
Video Motion Detection
12.8
Alarm Response
12.9
Integrated Multiplexer and DVR
12.10
Remote Distributed Multiplexing
12.11
Summary
12.1 OVERVIEW
Sequential switchers display the images from multiple cameras
on one monitor sequentially, one at a time, with a dwell time
between the display of each camera image. A disadvantage of
sequential switching and recording is that when a single video
camera is displayed on the monitor all the other cameras are not
being viewed. This can result ina great loss of intelligence from
the cameras not being displayed. Each camera image is
displayed for a dwell time set by the operator adjusted from a
few seconds to many seconds. With the use of sequential
switchers, many activ-ities on many cameras can be missed
since not all camera scenes are being displayed simultaneously.

A quad or video multiplexer displays all of the images from
many cameras onto a single split-screen monitor
simultaneously. These devices generate a video signal thatcan
record all the images at a much higher refresh rate than is
possible with a sequential switcher. The use of video
multiplexers eliminates the normal video time gaps created by
conventional sequential switchers.
There are basically three generic types of multiplex-ers:
simplex, duplex/full duplex, and triplex. The simplex
multiplexer can display multiple images—4, 9, 16, and 32—on
the same multi-screen monitor. The duplex mul-tiplexer
displays multiple images on a display but can also provide the
necessary encoding and decoding signals to simultaneously
record images on a VCR or a DVR. A triplexmultiplexer can
simultaneously display multiple live images on a display, record
camera images on a recorder, and display playback images from
a recorder.
Most multiplexers offer some form of basic video motion
detection (VMD). This might be listed in a variety of ways in
the literature but it essentially amounts to detecting movement
in the field of view of the camera by electroni-cally discerning
changes in the light level within the image.
In addition to displaying motion, multiplexers can respond to
alarm inputs from external sensors (door switch, infrared

detectors, glass break, microwave, etc.). Manufac-turers are
quick to point out, however, that the multiplexer’s primary
purpose is to furnish efficient video multiplexing and multi -
screen display. Alarm handling and motion detec-tion are
secondary functions, and the video multiplexer sys-tem should
not be the only alarm device on site.
12.2 BACKGROUND
Video multiplexing is an example of time division multi-
plexing. The video multiplexer constructs a sequence of pictures
captured from each of a number of cameras, in
341
CAMERA
1
2

3
8 CHANNEL
4
MULTIPLEXER
MONITOR
5
6
7
8
1
8
1

9
16 CHANNEL MULTIPLEXER
DVR/VCR
FIGURE 12-1 Video multiplexing system diagram
turn, and displays the video images in a split-screen format on
one monitor (Figure 12-1).
The initial electronic image splitters became available in the
form of a four-way splitter or quad. Subsequently the 9, 16, and
32 camera image splitters—multiplexers— became available.
These larger units had the ability to take synchronize d or
unsynchronized cameras and display them on a single monitor
simultaneously in a synchronized and stable format.
Early multiplexers were basically video switchers that could
mark each camera with a unique ID number in the vertical
interval. This required the cameras to be gen-locked or v-phased
(vertical sync) so the VCR would see a contin-uously composite
sync signal and so that it would not lose servo-lock on the

switched incoming video signals. To play-back the camera
images, the VCR switched the correct cam-era onto its output
only during its active period on the tape and switched to a gray
solid background picture for the rest of the time. This caused
severe image flicker but produced a viewable single camera
image display and was effective. Later generations of this
multiplexer design saved the active camera image until a new
picture was displayed eliminat-ing the gray background,
providing a better playback result. The primary benefit of this
technology was that this device
guaranteed a continuous composite sync to the VCR regard-less
of the video signal quality. A secondary benefit was that a non-
gen-locked or any other camera could be used with this
multiplexer. Present-day cameras have quality gen-locking
systems, and/or stable line-locked vertical interval sync, and
DVRs are used so this is no longer an issue.
The quad or multiplexer can also output these pictures as a
single continuous video signal with all the necessary encoding
and decoding for recording on a VCR or DVR or network. The
multiplexer adds the digital camera ID coding to the signal so
that the individual camera fields belonging to each camera can
be identified and recovered by the recording equipment on
replay. For display moni-toring purposes, the same sequential
scan process is used and each camera’s picture used is
electronically reduced in size and displayed in a pre-determined
position on the screen. Each camera is assigned a different
position so as to produce the familiar mosaic or cameo of
reduced size camera images on a single display monitor.
Most current multiplexers have the ability to display 4, 9, 16, or
32 pictures simultaneously on the screen. Usually the image
update rate is the same as the output multiplex rate, but some

manufacturers have refresh rates up to real-time capability. This
feature is useful since the screen is for
viewing only the multiplexed output being a full screen high-
resolution image. The multi-screen feature to display an
alarmed camera in a cameo format is very useful in playback of
all recorded images, and later single camera selection when an
alarm or some other activity needs to be viewed.
12.3 QUAD SPLIT-SCREEN DISPLAYS
The quad display is the simplest form of this multiplexing
technique where the signals from four cameras are pro-cessed to
appear in four quadrants of a single monitor display.
12.3.1 Quad-4 Image
The quad splitter permits viewing four live video cam-eras
simultaneously or selecting one camera full screen or
sequencing through all or selected cameras (Figure 12-2).
The quad splitter can display the images in quad or full screen
format while recording to a VCR or DVR in quad format. On
playback from the recorder the image from the quad multiplexer
can be zoomed up 2X and a freeze frame image can be displayed
for detailed analysis. The resolution of the quad ranges from
720 × 480 pixels up

343
to 1024 × 512 pixels for high-resolution systems. The units have
the ability to annotate the video image with time, date, camera
ID, and title in both the live monitor display and recorded
image display. The quads provide 30 fields per second, real -
time refresh rate.
Figure 12-3 shows diagrammatically the different scene formats
that the quad system can display. In the single camera select
mode the full screen images from camera 1, camera 2, camera 3,
and camera 4 outputs can be selected. In the quad mode, four
camera scenes are displayed and each individual picture on the
monitor is a full camera scene reduced in size (compressed).
Many quad systems can “freeze” a displayed image on the
monitor for detailed examination. This permits the security
operator to view a single scene in more detail over a period of
time until it is released by the operator. In this mode, a
recording can be made of the full screen or quad pictures.
Other options include alarm mode, video loss indica-tion, and
security lock. In the alarm mode, the system brings the alarmed
camera to full screen on the moni-tor alerting the operator of an
alarm activity, all while the recorder records in quad format.
Another feature available in some quad multiplexers is called
picture in a picture (PIP) in which a reduced image from one
cam-era is embedded into the full screen image of another
camera.

1
2
3
4
SCENE 2
SCENE 3
SCENE 1
SCENE 4
1
2
1
2
3
4

3
4
QUAD
SELECT
SEQUENCE
MENU
1
2

3
4
· 2
· 4
MONITOR DISPLAYS QUAD PICTURES OR ANY
INDIVIDUAL SCENES IN THREE MODES:
QUAD—FOUR COMPRESSED PICTURES SELECT— ONE
FULL PICTURE SEQUENCE THROUGH 4 SCENES
FIGURE 12-2 Quad splitter display block diagram
QUAD MODE—COMPRESSED SCENES
SEQUENCE MODE—FULL PICTURES
SCENE 1
SCENE 2

SCENE 1
SCENE 2
SCENE 3
SCENE 4
B
SCENE 4
A
SCENE 3
B

A
t = T1
t = T2
t = T3
t = T4
SELECT MODE—FULL PICTURES
SCENE 1
SCENE 2
SCENE 3

SCENE 4
B
OR
A
OR
OR
FIGURE 12-3 Quad combiner system
12.3.2 Multi-Image 9, 16, 32

Multiplexers are available to display images from 9 to 32
cameras on the same monitor display and are available with
similar features to those found in the quad system. Figure 12-4
shows two 9 and 16 camera examples of these systems and the
monitor images. Table 12-1 lists features of some of the quad
split-screen equipment available.
12.4 MULTIPLEXER TECHNOLOGY
A quad or multiplexer is an electronic device that time-
multiplexes video pictures from many cameras onto one video
display or video recorder. This means that the mul-tiplexer
displays one field or one frame from one camera, and then
immediately following that picture it displays the field or frame
from the next camera. It repeats the same procedure for all
subsequent cameras, and then starts all over again. These
images of multiple cameras are displayed on one monitor
simultaneously. Using this technique the full resolution of each
camera is maintained but the dwell time between displayed or
recorded image—2–3 second dead time switching time from a
sequential switcher—is reduced instead to milliseconds. When
recording the camera sig-nal to a VCR or DVR, the multiplexer
switches its input circuitry to each of the connected cameras, in
turn. To syn-chronize the cameras during recording, a series of
digital codes are embedded into the multiplexer output signal.
Part of this code identifies the camera channel number so that
the channels may be electronically recognized by the
multiplexer during playback. During playback,
another part of the code carries alarm status informa-tion so that
external alarm events are also recorded on the tape.
Time division multiplexing combines several camera video
input signals into one video output signal to dis-play all the

camera images on the monitor simultane-ously. Single images
are digitally captured from each of the video input channels,
and then lined up (queued) sequentially to form a continuous
video signal of time-sliced camera images. Included with each
captured image of video can be status information such as
alarms, cam-era titles, and time/date. Captured images are
controlled by an internal library that the multiplexer
automatically modifies to respond to alarms, motion detection,
or video loss (Figure 12-5).
To generate a multi-picture mosaic display, the multi-plexer
switches its input circuitry to each of the connected cameras, in
turn. The multiplexer has a video frame store (and electronic
image memory) used to capture a single picture from each
camera. As each image is captured, its size is electronically
reduced by a predetermined factor and the resulting cameo
picture is written into part of the frame store. This results in a
small image from the selected camera channel being frozen in
one area of the display screen. The same process is provided for
each of the cam-eras to similarly reduce the size and to position
the image in a particular location of the monitor display area.
As the multiplexer scans repeatedly around the channels, each
image is continuously refreshed and updated with new images
from the designated camera. This results in the familiar mosaic
of small camera images.
345

9 CAMERA
16 CAMERA
COMBINER
COMBINER
CAMERA 1

CAMERA 9
CAMERA 1
CAMERA 16
MODES:
MODES:

•
SELECT (FULL)
•
SELECT (FULL)
•
QUAD
1
9

•
QUAD
1
16
•
NINE
•
NINE

1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
9 10 111213 1415 16
QUAD
NINE

QUAD
NINE
SIXTEEN
FULL SCENE
1
2
3
FULL SCENE
1
2
3
4
FROM EACH

FROM EACH
5
6
7
8
CAMERA
4
5
6
CAMERA

15
16
(1/3 × 1/3 = 1/9)
(1/4 × 1/4 = 1/16)

MONITOR
MONITOR
FIGURE 12-4 Multi-image 9, 16, split-screen display

EQUIPMENT TYPE
FULL
SCREEN
4 CHANNEL—MONOCHROME
STANDARD RESOLUTION
4 CHANNEL—MONOCHROME
HIGH RESOLUTION
4 CHANNEL—COLOR

STANDARD RESOLUTION
4 CHANNEL—COLOR
HIGH RESOLUTION
STANDARD FEATURES ON MOST QUADS:
GRAY SCALE—256
DIGITAL ZOOM—2x
ADJUSTABLE SEQUENCE: 1–120 sec REMOTE CONTROL:
RS232, 422, 485 ALARM-DRY CONTACTS, RS232, 422, 485
4-ALARM INPUTS
OPTIONS: TIME/DATE ANNOTATION CAMERA ID
ANNOTATION
SCREEN DISPLAY MODE
RESOLUTION–FULL SCREEN
GRAY

QUAD
(H ×V)
COLORS
SEQUENTIAL
LEVELS
(4 CAMERA)
NTSC
CCIR/PAL
512 × 512

648 × 512
64
1024 × 512
720 × 576
256
720 × 480

64
16 M
1024 × 512
256
16 M
SCREEN FREEZE CAPABILITY
VIDEO LOSS ALARM
LOOP THROUGH TO DVR/VCR FOR RECORDING AND
PLAYBACK
SETUP MENU, ENGLISH, OTHER
NTSC 525, PAL 625 TV LINES
Table 12-1 Quad Split-Screen Equipment Parameters and
Features

CAMERA REPEATS EVERY 4 FIELDS
1
2
3
4
1
2
3
4
1
CAMERA 1
CAMERA 2
CAMERA 3
CAMERA 4
CAMERA 1
CAMERA 2
CAMERA 3
CAMERA 4
1
2
3
4
5
6
7
8
9
FIELDS
FOR THE MULTIPLEXER WITH 4 CAMERAS EACH

CAMERA IMAGE REPEATS EVERY 4 FIELDS = 4 × 1/30 sec
= 0.133 sec
FIGURE 12-5 Multiplexed signal from video stream
Most multiplexers can display the video cameras in four
different configurations: (1) quad, 4-way, (2) 9-way,
· 10-way, and (4) 16-way, and of course full screen for any
camera. Many can also display the cameras in different size
configurations. Figure 12-6 illustrates some of these split-
screen presentations.
In a standard sequential switcher the camera images are
displayed at a 30 frame per second rate. They are displayed
sequentially on the monitor at a rate determined by the number
of cameras in the system and the pre-assigned dwell times for
each camera. In the multiplexer switching system the number of
images displayed per second is based on the total number of
camera inputs.
If there is only one input the multiplexer displays at a
30 fps rate whereas with four camera inputs it display at
a 7.5 images per camera rate. With a larger number of cameras,
say 16 camera inputs, the final display rate would only be
approximately two images per second per camera, producing a
very jerky display (Figure 12-7).

Multiplexers now feature RS-422 and RS-485 and over-the-coax
digital PTZ control to eliminate the need to provide additional
controlling units for camera platform pointing and lens control.
Other features include motion detection, electronic digital
zoom, adjustable image sizes, and RS-232 interfaces to other
equipments.
The most common type of camera identification is the
annotation of digital information into the vertical inter-val time
of the video signal. This is accomplished by dividing a line of
video into, say, eight different sections. Each section is defined
as a one or zero by either the
1 2
3 4
4 IMAGES 2 × 2
1 2
10
10 IMAGES 2×8, 1×2

1 2 3
4 5 6
7 8 9
9 IMAGES 3 × 3
1
16
16 IMAGES 4 × 4
FIGURE 12-6 Multiplexer multi-screen displays
347
FIGURE 12-7 Multiplex camera
CAMERA FIELD SEQUENCE
sequencing technique
A1

A2
A3
A4
A5
A6
A7
A8
CAMERA A VIDEO FIELDS
B1
B2
B3

B4
B5
B6
B7
B8
CAMERA B VIDEO FIELDS
C1
C2
C3
C4
C5

C6
C7
C8
CAMERA C VIDEO FIELDS
D1
D2
D3
D4
D5
D6
D7

D8
CAMERA D VIDEO FIELDS
MULTIPLEXED CAMERA FIELD SEQUENCE
A1
B2
C3
D4

A5
B6
C7
D8
MULTIPLEXED VIDEO FIELDS STREAMED TO DVR OR
VCR
presence or absence of black video or white video. By doing
this with eight sections it can be interpreted as one byte of
digital data that can be converted to a number from 0 to 255.
12.4.1 Image Rate vs. Number of Cameras
A factor to be considered is that multiplexers are basically fast
video switchers. When many cameras are connected and the
time lapse (TL) recording is too slow, the time of recording a
single image from a particular camera may be too long to catch
any event. The multiplexer system basically takes the number of
camera inputs and dividing that by the recorded pictures per

second. To calculate the refreshed or update rate:
Update rate =
Number of Cameras
(12-1)
Recorded Pictures/second
If there are 16 cameras in the system and the recording time is
168 hours in TL mode, then it will take 17.4 sec-onds to record
a new image from any one camera input. This obviously would
have little use in any application since someone could walk by
and never be recorded. Reducing the recording time to 24 hours
in TL, there would be a new image every 3.2 seconds. This
would be more accept-able but not very applicable in high-
traffic areas. Going to a 24-hour virtual real-time (pictures per
second), there
would be a new image every 0.8 seconds. This would be useful
in most applications.
12.4.2 Encoder/Decoder

Multiplexers require encoders and decoders to identify each of
the incoming video camera signals for processing. All current
encoder/decoder designs use analog to digital (A/D) converters
to convert the standard video signal into a digital format for use
with common digital logic devices. After the signal is
processed, it is later converted from digital to analog (D/A) for
output to be displayed back onto the analog video monitor or
recorder.
12.5 HARDWARE IMPLEMENTATION
There are basically three different generic types of mul-
tiplexers: (1) simplex, (2) duplex/full duplex, and (3) triplex
(Figure 12-8).
The simplex multiplexer can display multiple images: 4, 9, 16,
and 32—on the same multi-screen monitor. The duplex
multiplexer displays multiple images on a display but can also
provide the necessary encoding and decod-ing signals to
simultaneously record images on a VCR or a DVR. A triplex
multiplexer can simultaneously display mul-tiple live images on
a display, record camera images on a recorder, and display
playback images from a recorder.
CAMERA 1
CAMERA 1
CAMERA 1

DUPLEX AND
SIMPLEX
CAMERA 2
CAMERA 2
TRIPLEX
CAMERA 2
FULL DUPLEX
MULTIPLEXER
MULTIPLEXER
CAMERA 3
MULTIPLEXER
CAMERA 3
CAMERA 3

Chapter 11Video SwitchersCONTENT

MONITOR
DVR/VCR
DISPLAYS SINGLE OR MULTIPLE IMAGES ON MONITOR

CANNOT RECORD AND SHOW MULTISCREEN DISPLAY
AT THE SAME TIME
DISPLAYS THE MULTISCREEN
AND RECORDS ON THE DVR/VCR
SOME CAN PLAYBACK FROM ONE
RECORDER WHILE RECORDING ON
ANOTHER AND GIVE UP THE MULTI-
SCREEN VIEWING IN THIS MODE
MONITOR DVR/VCR
ALL THE FEATURES OF THE FULL DUPLEX. IN ADDITION
THE MULTI-SCREEN OUTPUT CAN BE SUBSTITUTED FOR
A THIRD DVR/VCR RECORDER
NOTE: 1. THE FULL DUPLEX CAN RECORD THE
MULTIPLEXED OUTPUT TO ONE DVR/VCR, PLAYBACK
FROM ANOTHER AND VEIW THE MULTISCREEN AT THE
SAME TIME.
2. MONITORS DISPLAY EITHER FULL SCREEN OR
MULTISCREEN.
FIGURE 12-8 Generic multiplexer types: simplex, duplex, and

triplex
12.5.1 Simplex
The simplex multiplexer is the lowest-cost multiplexer type, has
the least number of features, and is easy to install and set up.
They are generally used in small systems when there is no
security operator active at the console. The simplex multiplexer
does not have the ability to record and show a multi-screen
display and record simultaneously. The simplex multiplexer unit
can either display or record the video information with the
initial setup of the mul-tiplexer determining the choice. They
are available for monochrome or color camera systems and with
options for VMD and alarm handling.
12.5.2 Duplex/Full Duplex
A duplex multiplexer is designed to display
either:
(1) a live camera view, (2) a live multi-screen display, or (3)
previously recorded images. This multiplexer has the ability to
display the multi-screen camera images and record the
multiplexed video and control data to the VCR or DVR. Some
duplex multiplexers can playback from one recorder while
recording on another but the multi-screen viewing is forfeited.
A full duplex multiplexer has the ability to: (1) record the
multiplexed output to one recorder, (2) playback
from another, and (3) view the multi-screen at the same time.

Duplex and full duplex multiplexers are available for
monochrome or color camera systems and with options for VMD
and alarm handling.
12.5.3 Triplex
A triplex multiplexer allows viewing of live and recorded
images on one monitor simultaneously, eliminating the need for
a separate playback monitoring station. The triplex multiplexer
has all the features of the full duplex but the multi-screen
output can be used for a third recorder, or to display live video
which is the more com-mon application (Figure 12-8).
Triplex multiplexers are available with two monitor outputs.
Output #1 produces a full screen or multi-screen digital image
display that can be frozen on the screen or zoomed in or out.
Output #2 displays a full screen, live.
Triplex multiplexers are available for monochrome or color
camera systems in 10 and 16 camera models with options for
VMD and alarm handling. They have on-screen menu prompts
to simplify installation and setup. Table 12-2 lists features of
some of the multiplexer equip-ment available.
349
EQUIPMENT TYPE

MULTISCREEN
RECORD TO
RECORD TO
RESOLUTON
DVR/VCR,
NUMBER OF CAMERAS *
DISPLAY
DVR/VCR
NTSC
PAL
DISPLAY

SIMPLEX**:
720 × 512
720 × 512
16 CHANNELS
32 CHANNELS

DUPLEX:
4, 8, 16
†
16 CHANNELS
†
720 × 480
720 × 576

32 CHANNELS
4, 8, 16, 32
FULL DUPLEX:
4, 8, 16
720 × 572
720 × 572
16 CHANNELS
32 CHANNELS
4, 8, 16, 32

TRIPLEX‡:
4, 8, 16
720 × 512
720 × 512
16 CHANNELS
4, 8, 16, 32
32 CHANNELS

· 4 CHANNEL AND 10 CHANNEL ALSO AVAILABLE
· MOST SIMPLEX DO NOT HAVE CAPABILITY TO
DISPLAY MULTI-SCREENS
· DUPLEX—SOME CAN PLAYBACK FROM ONE DVR/VCR
WHILE RECORDING ON ANOTHER BUT GIVE UP MULTI-
SCREEN VIEWING DURING THIS TIME.
· TRIPLEX—ALL FEATURES OF DUPLEX BUT MULTI-
SCREEN OUTPUTS CAN BE SUBSTITUTED FOR A THIRD
DVR/VCR.
Table 12-2 Multiplexer Equipment Parameters and Features
FEATURES IN MOST:
ALARM INPUT: EACH CAMERA ALARM OUTPUT

ANNUNCIATION DIGITAL ZOOM: 2x DATE/TIME,
CAMERA ID
VIDEO LOSS INDICATION
ON-SCREEN MENU
12.6 RECORDING AND PLAYBACK
12.6.1 Analog and Digital Recording
When recording the camera signal to a VCR or a DVR, the
multiplexer switches its input circuitry to each of the connected
cameras, in turn. The video frame-store in this mode is used to
capture a single full-screen field image from each camera. The
separate video fields captured from the cameras are re-
synchronized by the frame-store for recording onto the video
recorder. Using this method, it is possible to record at an
average rate of more than 30 cameras per second. Since the
video frame-store inher-ently time-base corrects the sync and
synchronizes the camera signals as part of the camera capture
process, the cameras need not be externally synchroni zed.
To synchronize the camera signals during recording, a series of
digital codes are embedded into the multiplexer output signal.
Part of this code identifies the camera channel number so that
the channels may be electronically recog-nized by the
multiplexer during playback. During playback, another part of
the code carries alarm status information so that external alarm
events are also recorded.
12.6.2 Video Playback
When the multiplex recording is played back through the

multiplexer to the monitor, the multiplexer first extracts the
digitally coded data element and uses this to identify the camera
ID number information. When a valid channel
number is identified, the multiplex captures the associated video
images in the frame-store. The simplest playback mode is where
a single camera channel is requested for playback. In this mode
the multiplexer captures the cor-responding video images from
this camera and displays them as full screen images, and
updates each time it iden-tifies another image with the same
camera ID number. The embedded digital data packets are
decoded, and all the associated status information, titles, time
and date of recording, etc. are re-constructed and displayed with
on-screen text during playback.
During playback, the user can select one of several screen
formats, the cameras to be displayed, and the cam-era positions
in the multi-screen display on the unit. The playback speed is
selected on the recorder, not on the multiplexer unit.
For multi-picture playback mode the frame-store is used in a
similar way as that used for live multiplexer viewing. Many
channels are reviewed off the recorder alongside one another on
the screen. In this mode the same size reduction and image
positioning processes go on in the multiplexer as was described
for the live multi-picture mode. In this case, however, there is
only one input sig-nal, i.e. that coming from the video recorder
playback, but there are many camera channels within the signal.
The multiplexer recognizes each new camera channel number, it
size-reduces the captured image, and places it on the screen in
the predetermined location and with a size cor-responding to
that camera. This results in a multi-picture display very similar
to the live one showing recorded infor-mation rather than live
video. If the operator sees an

image in the cameo images of interest, the multiplexer can be
switched to the full screen mode for that camera, to examine the
scene more closely.
12.7 VIDEO MOTION DETECTION
Most multiplexers offer some form of basic motion sens-ing or
VMD. This might be listed in a variety of ways in the literature
but it essentially amounts to detecting move-ment in the field of
view of the camera by electronically discerning changes in the
light level within the image.
In operation the multiplexer digital signal processing (DSP)
electronics determines if something has changed in the video
image of any camera. If nothing has changed the multiplexer
records fewer pictures per second from that camera, thereby
increasing the images per second recorded for other cameras
that have motion or activity and their scenes. One caution
regarding VMD: some out-door environments have complex
detection requirements. In those cases, use a non-video motion
detector sensor intended specifically for such situations.
Motion sensing can be considered effectively as an alarm that is
flagged internally to the multiplexer. This feature is particularly
useful in the recording mode since it can
allow the frame/field recording rate of the recorder to be altered
such that images showing movement are recorded at a faster rate
than the static ones.
The multiplexer can optimize the display and recording by
displaying and recording only video camera images in which

activity is occurring.
12.8 ALARM RESPONSE
Multiplexers offer VMD including built-in zone selection with
sensitivity settings and alarm linking per camera. Motion
detection is used to adjust the rate at which camera images are
recorded and can also act as an intrusion alarm sensor to trigger
on alarm input. The VMD can be used to simply optimize
recording or as an alarm condition. Motion detection can be
used as an alarm condition only if movement is detected where
no movement is expected or permitted.
Non-video alarms are signal inputs from external sensors that
can be acted upon by alarm monitoring hardware or a security
operator at a console (Figure 12-9).
Common alarm sensors take the form of door con-tacts, PIR,
glass break, microwave motion sensors, trip-wire, photo-
electric, magnetic, seismic, etc. All are examples of
ALARM SENSORS:
DOOR CONTACT
PIR MOTION SENSOR
CAMERA 1

MICROWAVE MOTION
2
PHOTO-ELECTRIC
GLASS BREAK
3
SMOKE/FIRE
4
MULTIPLEXER
MONITOR

8
1
1
9
DVR/VCR
TO ALARM
ANNUNCIATOR
FIGURE 12-9 Alarm signals trigger multiplexer
external devices that can output signals to the multiplexer when

an intruder enters a monitored area, and be used for alarm
annunciation. All these devices can be used by most
multiplexers as an input to bring up the picture of the camera
that is located in the alarm sensor area and to annunciate an
alarm via sound or light indicator. They are also used to
command the VCR or DVR recorder to change from TL to real-
time recording mode for that camera, and record at a faster
speed. In normal use, the recorder is in TL mode to make
economic use of the storage media. When an alarm event
occurs, the recorder speed is increased to real-time. State-of-
the-art multiplexers can cope with making these changes from
TL to real-time and acting on alarm inputs. An input from an
external alarm by a contract closure to the recorder or by a
serial RS-232 port command will be multiplexed, though, and
will cause the recorder to change speed. The multiplexer also
makes it possible to select logical groups of cameras and to
salvo or bank switching of those cameras. Salvo switching
accomplishes the switching of several or many cameras in a
related zone simultaneously when an alarm input occurs. As an
example, in a 20 camera installation the normal record-ing set
up may provide the TL recording for all 20 cameras. There may
be PIR, other motion sensors, and/or switch sensors in the area.
When an alarm event is triggered via one of these sensors, the
multiplexer causes the images from cameras in the area of the
sensors to be recorded in real-time. Ideally the video system
should take automatic
351
action as much as possible, and not require the operator to
intervene.

12.9 INTEGRATED MULTIPLEXER AND DVR
The VCR has been replaced by the DVR in many video security
systems. Consequently, many DVRs are now incorporating the
video multiplexer into the DVR unit. The combined video
multiplexer–digital recorder simpli-fies and reduces errors
during the hardware setup pro-cedure and simplifies the design,
operation, and cost of the system. Figure 12-10 shows a full-
featured DVR– multiplexer combination. Table 12-2 lists some
of the mul-tiplexer equipment available.
12.10 REMOTE DISTRIBUTED MULTIPLEXING
Digital technology is finding its way into the use of
multiplexers in LAN, WAN, etc. as a superior technique for
distributing, controlling, and recording video signals especially
at remote distances (Figure 12-11).
Some multiplexers designed for larger physical security
installations lend themselves to distributed multiplexing. This
permits groups of cameras that are located in
DVR/MULTIPLEXER

CAMERA INPUTS
TRIPLEX
DIGITAL
HARD DRIVE
COMMUNICATION
VIDEO RECORDER
DISPLAY
(1–16 TYP)
MULTIPLEXER
(80, 160, 320 GB)
DRIVERS/PORTS
ELECTRONICS

SYSTEM FEATURES: • COMBINED DVR/TRIPLEX
MULTIPLEXER.
· NUMBER OF CAMERA CHANNELS: 4, 9, 16
TRIPLEX OPERATION: SIMULTANEOUSLY VIEW LIVE
AND PREVIOUSLY RECORDED VIDEO IMAGES WHILE
CONTINUING TO RECORD AT THE SAME TIME, USING
EITHER ONE OR TWO MONITORS.
· LIVE IMAGE RESOLUTION: 720 × 480 NTSC.
· RECORD RESOLUTION: 720 × 224 NTSC.
NETWORKABLE VIA ETHERNET (TCP/IP)
· MOTION DETECTION WITH CONFIGURABLE
SENSITIVITY LEVELS.
· DIAL UP MODEM.
· REMOTE ALARM NOTIFICATION FROM MOTION

DETECTION.
· PRE-ALARM VIDEO.
· EXPORT VIDEO VIA USB PORT.
FIGURE 12-10 DVR-multiplexer system
CAMERAS
CALL
MAIN
1
MONITOR
MONITOR
KEYBOARD
2

1
9
3
MODEM
4
IEEE-1394 (FIREWIRE)
USB
ALARM INPUT/

OUTPUT DEVICES
POTS
DOMES
ROUTER
MODE
LAN
16

16 CHANNEL TRIPLEX
INTERNET
MULTIPLEXER/DVR
HARD DRIVE
CD READ/WRITE
MONITORING STATIONS
FIGURE 12-11 Remote distributed multiplexing
physically distant locations to be connected to a slave mul -

tiplexer. Several of these remote multiplexers are con-trolled
from a single master multiplexer at the central location. The
master unit communicates over the Internet and provides the
video processing, recording, and display process signals. It
commands each slave unit to deliver the required camera
channels from the slave to the master, where they can be
combined by one or several multiplexer recorder systems.
Multiplexing digitally compresses the images of each video
frame and transmits them over the digital network to a DVR and
onto digital monitors. The video images are compressed before
transmission and later decompressed to display them on the
monitor.
12.11 SUMMARY
The multiplexer and integrated multiplexer–recorder (DVR)
have become an important part of the video
surveillance hardware. It is a powerful tool capable of
combining many video images onto one multi-screen dis-play,
thereby reducing the number of monitors required to view the
cameras in the system. It can call up cam-era images showing
motion in the scene. It also provides the capability to prioritize
incoming alarm signals from external sensors with the VMD
alarms. In an analog net-work, the multiplexer can send the
video images and cam-era identification signals to a VCR or
DVR for proper synchronization on recording and playback to
the mul-tiplexer. In a digital network, it can transmit the com-
pressed images over digital networks to monitors and recorders
to remote locations for remote site monitor-ing. The simplex,
duplex/full duplex, and triplex types are available to provide a
multiplexer solution to most applications.

Chapter 13
Video Motion Detectors
CONTENTS
13.1 Overview
13.2 Background

13.3 Functional Operation
13.3.1 Surveillance
13.3.2 Detection Probability
13.3.3 Motion Assessment
13.3.4 Scene Lighting
13.3.5 Training Function
13.4 Analog Video Motion Detector (AVMD)
13.4.1 Technology
13.5 Digital Video Motion Detector (DVMD)
13.5.1 Mode of Operation
13.5.2 Technology
13.5.2.1 Programming the digital VMD
13.5.2.2 DVMD Setup Procedures
13.5.2.3 Sensitivity Settings
13.5.2.4 Motion Detection Sensitivity
13.5.3 Hardware
13.5.3.1 Normal Mode
13.5.3.2 Trace Mode

13.5.3.3 DVMD Graphic Site Display
Maps
13.5.4 Features
13.6 Guidelines, Pros and Cons
13.7 Summary
13.1 OVERVIEW
The method by which current security systems trigger secu-rity
alarms can be divided into two classes. At one end of the
spectrum there are systems that sense physical move-ment, such
as simple contact switches and PIR sensors. While all these
systems can be quite varied in the technol-ogy they use, the
systems have one thing in common: they can only recognize
movement. On the other hand, there

are visual detection systems ranging from guards posted at
specific locations to camera systems with analog video motion
detectors (AVMD) or digital video motion detec-tors (DVMD).
The DVMDs use monitors, real-time and/or TL VCRs or DVRs
to discern between allowable activities, breach of security or
provide identification of individu-als, and give instructions to a
guard on what a response should be.
Any video security system should include the following four
ingredients: (1) surveillance, (2) detection, (3) assess-ment, and
(4) response. The VMD can be a part of the system hardware to
provide the surveillance, detection, and assessment, and provide
accurate detailed and con-cise information to the guard force,
allowing the force to respond optimally. As a free by-product,
the VMD also makes available a training tool to practice and
perfect the guard response philosophy. To achieve high
detection probabilities in any moderate to large security system,
the integrated video system must operate with an automated
VMD detection system.
The recent availability of affordable DSP techniques has forever
changed the security scenario and eliminated the shortcomings
of the simple motion detectors and first generation AVMD
detectors. In simple terms, advanced DSP technology has
brought intelligence into the world of DVMD. DVMD systems
combine visual video presentation of the motion detection with
recording technology. Intelli-gent VMD systems go a step
further by using sophisticated DSP algorithms so that motion
detectors learn or adjust to a changing or new scene, virtually
eliminating false alarms that were prevalent in the analog and
simple first generation DVMD technologies. Intelligent DVMDs
can be programmed to overlook small changes in the scene such
as rain, dust, moving tree branches that often render traditional
VMDs unusable.

The useful security information displayed on a video monitor
often comes from motion within the scene—a
353
moving person, vehicle, object, or some activity involving
motion. Irrespective of the number of security monitors, it is
important to have an alarming device to alert the guard to
motion or activity in a scene. Medium to large video
installations generate many camera scenes that must ulti-mately
be displayed on monitors, but it is difficult for a security guard
to watch multiple monitors over long peri-ods of time. The
video multiplexer goes a long way in reducing the number of
monitors the guard must view and at the same time increases the
operator’s ability to react to real threats, but it is the VMD that
electronically analyzes and monitors camera images to detect
changes (motion) that are judged to warrant an alarm. The VMD
provides an electronic alternative to a guard sitting and staring
at the monitors, and can notify the guard immediately of situa-
tions requiring attention. VMD systems operate to detect
changes in a specified area within the camera FOV. They do this
by comparing the light levels of camera pixels from one video
frame to the next, looking for changes considered significant. In
the simpler, lower-cost AVMD systems, large areas in the
incoming frame are compared with those of a previous reference
frame. This type of sys-tem works reasonably well indoors,
where there are few changes in the scene and where lighting is
constant. Ana-log systems are, however, susceptible to false
alarms caused by lighting changes, debris passing through the
camera FOV, small animals, ripples on bodies of water, or
camera vibration. They are therefore not recommended for most
outdoor applications and instead the DVMD is used. The

microprocessor DSP-based DVMD can analyze thousands of
picture zones and operate with low false-alarm rates even under
severe light-level changes. Most DVMDs, with the exception of
those using the latest intelligent image processors and learning
algorithms, are not suitable for PTZ applications.
Environment plays a major factor in choosing the DVMD for
outdoor applications. The DVMD can toler-ate some camera
vibration, but the camera should be mounted as securely as
possible. The DVMD can also toler-ate light-level changes as
might occur when a cloud passes in front of the sun, without
causing a false alarm. Some DVMD systems can subtract out or
ignore inherent scene motions such as waving flags, leaves, or
trees, so that they will not be a source of false alarms. Some
have the ability to selectively sensitize and desensitize certain
portions of the scene in order to prevent false alarms. They
desensi-tize parts of the scene where inherent motion and no
real activity is expected, such as leaves rustling on trees. This
reduces the chance of false alarms.
After a target has been detected and classified, the DVMD
tracks that object within the site as the target moves from
camera to camera. Systems are now available that can display
images from remote locations showing targets in motion. The
system can detect, classify, locate, and track objects within the
FOV of the camera. The operator has a mapped display of the
site, highlighted with icons of the
various types of targets (cars, personnel, gates, etc.), and can
see an icon of the moving car or other target on the digitized
site map. The path which vehicles take is synthe-sized in the
monitor display of the FOVs of several cameras that the car had
passed and traversed. Actual video scenes are available by

clicking on the icon. These type systems are finding use in
environments such as airports, seaports, and large i nstallations.
13.2 BACKGROUND
In its most general sense, a motion detector is an ana-log device
that responds to movement recognized as a specific type and
rate of change within a defined moni-tored area of coverage.
The original motion detectors were designed to detect motion or
movement in a stable back-ground by means of PIR technology
using pyroelectric detectors. These PIRs sensed gross changes
in movement but provided very little intelligence as to the cause
of the movement.
A video camera provided with appropriate VMD pro-cessing
electronics can make the camera operate as an alarm sensor. The
VMD processing electronics memorizes the instantaneous video
picture, and then if some part of the picture changes by a
prescribed amount, the system generates an alarm signal to alert
a guard or activate a video recorder. The AVMD or DVMD is
connected into the video system as shown in Figure 13-1. The
figure shows an individual entering a room and the successive
video frames showing the person walking through the facility.
The VMD will detect the motion of the person, highlighting the
per-son on the monitor screen, and/or also producing a visual or
audible annunciation to the security officer.
Two VMD processing electronic types have been developed: the
first-generation analog and the second-generation digital. The
DVMD provides significantly more capability and reliability but
costs more. Surveillance of any scene is achieved by the use of
conventional video cameras and lenses positioned throughout
the area of interest at locations that permit recognizing an
intruder or movement within the camera FOV. Cameras should
be positioned so they can view all activity and targets of inter -

est. Figure 13-2 illustrates the VMD’s place in the video
surveillance system.
In the 1980s, several DVMD systems became available. These
were large, complex, and expensive units with elec-tronic
memory and logic that dissected a video image into zones. Each
zone represented an area in which motion could be monitored.
By dividing the video image into hun-dreds of zones, the target
could be localized in the scene and defined in size and motion,
than it could in the orig-inal AVMD system. The light level of
each zone likewise could be analyzed providing further
intelligence about the scene. These systems were only
affordable by large com-mercial institutions and government
facilities. It was only
355
TO MONITOR,
VIDEO
ANALOG
CAMERA
OR

RECORDER
LENS
OR PRINTER
DIGITAL
VMD
CAMERA/LENS

FIELD OF VIEW(FOV)
G
G
G
SUCCESSIVE VIDEO FRAMES SHOW PERSON WALKING
THROUGH FACILITY
G
G

G
G
G
T = 0
T = 1
T = 2
T = 3
T=4 SEC
FIGURE 13-1 Video motion detector (VMD) in the video
security system
SCENE

LENS
VIDEO MOTION
MONITOR,
CAMERA
RECORDER
DETECTOR (VMD)
OR PRINTER

DECISION ELECTRONICS
VIDEO
VIDEO
THRESHOLD CRITERIA:
VISUAL
SCENE LEVEL

FROM
TARGET SIZE
CAMERA
AUDIBLE
TARGET SHAPE

TARGET SPEED
SWITCHED
NUMBER OF TARGETS
ALARM
SIGNAL
OUTPUT

PRESET MOTION THRESHOLD
(OPERATOR CONTROL)
FIGURE 13-2 Video motion detection system and detection
parameters
into the mid-1990s that digital electronic costs were suffi-
ciently reduced to make the present DVMD practical in security
applications.
The evolution of the AVMD to the DVMD provided a
significant step forward in identifying the source of an intrusion
or movement in a video scene by providing more intelligence to

the security operator. Early analog systems were limited to
monochrome video cameras since color cameras were not in
widespread use during the 1980s. The modest electronics in the
AVMD limited their use to indoor applications as they could not
deal with all the uncontrolled lighting, weather, and stray
motion interfer-ences in an outside environment. The
introduction of the CCD camera in the 1980s and low cost color
cameras in the 1990s initiated the advent of a totally new
technology in VMD. This, however, was not sufficient to make
the AVMD a reliable product for indoor applications and
especially not for outdoor applications. In the mid- to late
1990s, however, the introduction of the DVMD in conjunction
with the CCD camera with DSP improved motion detec-tion
significantly. Digital circuitry and availability of inex-pensive
solid-state memory brought about the widespread use of DVMD.
The DVMD has the ability to dissect the video image and
analyze the scene on a pixel-by-pixel basis, thereby allowing
sophisticated analysis of the motion in the scene. These new
DVMDs have proven to be very reliable for alarm management,
and provide automatic intrusion detection and automatic
recording of intrusion events. They are used in open areas and
relieve console guards of the tedious monitoring of empty
hallways, rooms, parking lots, and parking garage levels that
have no activity. The improvements in reliability through
sensitivity adjustments and digital analysis of movement on a
pixel level has given credibility to the idea that video
surveillance systems can and should perform automatic motion
detection without individual camera scenes requiring active
monitoring by a console guard.
Advanced programming for “specific act recognition” is just
beginning to emerge from development. This is motion
detection that recognizes unique and complicated motions
associated with undesirable act phenomena. Recognized acts can
be the typical movement of shoplifters, acts of physical
violence, or phenomena such as fire. Video smoke detection

software programs are currently being marketed. Also emerging
is the coupling of motion detection with alpha-numeric
character and biometric recognition. This takes the form, for
example, of spotting license plates or vehicle signage and
processing the numbers or characters remotely. Some facial
systems provide recognition of spe-cific faces in a crowd and
are finding their way to market.
13.3 FUNCTIONAL OPERATION
Before any AVMD or DVMD can be applied to a partic-ular
application, its location—indoor or outdoor—must
be considered. In an indoor application, the light-level changes
are usually predictable or at least not very signif-icant.
Successful VMD operation depends on recognizing light-level
changes in specific parts of the scene (caused by an intrusion or
disturbance) in contrast to overall scene light-level changes
caused by changing lighting conditions. These two phenomena
must be differentiated to avoid undue false alarms. In indoor
lighting applications where the light level is controlled by the
user, a simpler AVMD system can be used.
13.3.1 Surveillance
Video surveillance is accomplished via the use of cameras and
lenses located and positioned for maximum intelli-gence
gathering of a viewing area. The cameras can act synergistically
with other alarms as remote eyes to present a visual image of an
area as well as the source for an alarm input.

Monitoring a large area such as a parking lot using VMDs
presents multiple possibilities including: (1) a wide-angle lens,
(2) multiple cameras, and (3) dual-lens, split-screen. When a
wide-angle lens is used, the alarm source (intruder) appears
small on the monitor screen and a guard does not detect the
intruder, especially if the intruder takes cover quickly. The
VMD can detect the intruder and register an alarm. With
multiple cameras, the parking lot FOV is divided among the
cameras, each viewing a section of the overall area. Each
camera must use a separate VMD. With the split-screen
technique one lens can be wide-angle, the other a medium or
narrow-angle lens.
If the system includes pan/tilt equipment the guard must pan,
tilt, and zoom the camera/lens to locate the alarm source. This
is not a simple task, and in the time required for the guard to
perform it, the intruder may be gone. In more sophisticated
systems, in order to speed reaction time, the location of the
motion in the image is used to point the pan/tilt platform in the
direction of the motion.
13.3.2 Detection Probability
The protection of outdoor areas presents the most difficult
problem in facility security. All sensing devices are plagued by
false alarms due to the unpredictable nature of natu-ral
phenomena and intentional artificial alarms. Seismic sensors
produce false alarms due to vibrations caused by wind, vehicles,
and other objects. Microwave sensors pro-duce false alarms due
to moving animals, blowing papers, or leaves. An effective
outdoor security system is best aug-mented using video cameras
viewing the actual scenes to filter out and recognize false
alarms. Although an alarm denotes that a certain area has been
disturbed, without a

visual image little information is provided as to the nature of
the alarm or the precise location at which it occurred. Without a
video image, security personnel must be sent out to investigate
and determine the nature of an alarm. Since outdoor monitored
areas are often large, in many cases by the time a security guard
responds to the alarm, the intruder is gone or the activity has
ceased.
A guard monitoring a medium to large video secu-rity system
must view many monitors that display either:
· sequenced scenes, (2) several monitors—one for each camera,
or (3) monitors with split-screens. To assure a high probability
of detection, the camera lens magnifica-tion must be such that
an intruder is displayed on the monitor magnified enough so
that the guard can easily see him and attract his attention. Using
multiple cameras is often the best solution to provide the
necessary coverage to detect the intruder.
For a guard’s response to an intrusion to be effective, the guard
must first know that he is responding to a real intrusion, its
location, and nature. The VMD function is to display only
intrusion alarms on the video monitor with-out any human
intervention. The guard then assesses the alarm by viewing the
monitor.
The VMD system must give timely information as to the exact
location and nature of the activity and must:
· respond to small changes (motion) in the camera/lens FOV, (2)
activate an alarm output on the monitor to alert the guard that
an intrusion has occurred, and (3) dis-play the alarmed scene on
the monitor. It should also be accompanied by an audible and/or
video alarm and acti-vate a VCR or DVR and video printer. For
larger digital infrastructures it should be able to provide
transmission of the video image over a network. The displayed

scene should show the location within the scene that has been
activated and give immediate information to security per-sonnel
as to the precise location, movement, and nature of the alarm. If
an intruder is hiding, a flashing pattern on the monitor should
show the path of the intruder from entry of the scene to the
point to where he is hiding.
Intrusion detection probability is controlled by the placement of
cameras and is a system design parameter. The ideal motion
detection system would give a 100% probability of detection of
intrusions, zero false-alarm rate, zero nuisance alarms, and zero
equipment failure. With proper camera placement and reliable
equipment, target-detection probabilities can be 95–99%. Alarm
assessment takes place in the time it takes for the operator to
view the scene and identify the cause. When a VMD is used, the
security operator does not have to identify the camera or locate
the movement on the screen, since the cause of the alarm is
indicated by the brightened flashing map on the monitor. If it is
an intruder, the guard responds accordingly, knowing where the
intruder is and who he will be confronting. If it is not an alarm,
the guard can press an alarm reset button and go on to the next
alarm.
357
Video motion detectors are valuable not only because they can
cue a video response but also because they are an independent
source of vital information. There may be particular situations
where a specific activity within an area covered by the camera
would be difficult to detect with other conventional forms of
alarms. It is often important to know not only that an intrusion
occurred in a certain space or area but also the path the intruder
took. VMDs with enhanced mapping display capability can
provide this information.

13.3.3 Motion Assessment
Assessment is the ability of the console operator to identify and
evaluate the cause of the alarm. This judgment call is one of the
most important decisions for two reasons:
· if a real intrusion occurs the guard’s assessment must be rapid
and accurate and depend on a visual judgment,
· if the alarm is not a valid intrusion, the guard must be able to
make that decision rapidly and accurately— which again
requires visual observation of the cause of the alarm—and then
cancel it.
In some DVMD systems a RAM module stores the alarmed
locations in a separate RAM alarm map mem-ory (AMM). Upon
alarm, the contents of the AMM are displayed on the alarmed
video monitor scene as a flash-ing, highlighted array of alarm
points. This feature is a key to quick, accurate assessment of all
alarms. The AMM enables the operator to determine instantly
the exact loca-tion where the disturbance or intrusion has
occurred and provides a quick, precise evaluation of the alarm
to provide the appropriate response. To clear the alarm
condition after a response has been made, the operator presses
an alarm reset switch and the monitor returns to the normal
blank condition. This accurate, rapid assessment optimizes the
use of the response force. If a second or additional alarm occurs
prior to resetting, the alarm scenes are dis-played with their
alarm maps in sequence on the master monitor, at a selectable
rate.
When a large number of cameras are alarmed simulta-neously,
an assessment problem can occur. By the time the guard views

the last camera, the intruder most likely has left the scene and
only the map remains. The DVMD effectively controls the
situation by providing a video out-put to record all alarmed
camera images. This is done automatically while the guard
watches the monitor. The video frames (scenes) are sent to the
VCR or DVR at a rate of 30 fps. The pictures are recorded—one
from each camera—in sequence and continue until the operator
resets the equipment. When a guard realizes a multiple-intrusion
attempt is in progress, the guard can playback the recorded
video images into the monitor and replay the intrusion with the
alarm map to determine the cause of the alarm in the scene.
Using this technique the alarm assessment capability is
extremely high. The guard need
not leave the console during an alarm condition unless i t is
necessary to initiate a direct response to a real intrusion. The
guard can observe the progress of the intruder into the area by
observing the monitor as the intrusion map is generated.
optimally. This important training improves the plan, the guard
response time and method, and overall security.
13.4 ANALOG VIDEO MOTION DETECTOR (AVMD)
13.3.4 Scene Lighting
Since the VMD makes its decision based on the scene the
camera is viewing, it is important that lighting at the camera
site is adequate. The VMD equipment must be able to
compensate for variations in average scene lighting occurring

during daylight hours as well as when auxiliary artificial
lighting is provided during nighttime operation. VMD systems
operate with scenes illuminated by visible or infrared lighting.
In outdoor applications, the environment is not as con-trollable:
significant light-level changes are caused by sun-light, cloud
variations, lightning, and many different types of objects
passing through the camera/lens FOV. Many DVMD systems
operate well under most outdoor condi-tions but they lose some
of their capability under adverse environmental conditions of
heavy snow or rain, and alter-native systems using other sensors
should be relied upon. The DVMD used in an outdoor
environment has a signifi-cantly higher potential for false
alarms due to these unpre-dictable lighting changes and moving
clutter. The DVMD must have outdoor algorithms that correctly
account for these rapid changes in overall scene brightness and
illumi-nation, as well as area changes in illumination caused by
rapidly moving phenomena. If there is movement in the scene it
must be detected while the movement is still in the scene.
Therefore, if updates of the scene occur at too slow a rate, an
object at a distance may elude detection.
To determine whether a target is of interest or a false alarm, the
equipment must be able to distinguish its size, speed, and shape.
In outdoor applications a DVMD is the only solution.
13.3.5 Training Function
In the intrusion scenario, when an alarm occurs the con-sole
operator is called upon for the first time to evaluate the alarm
on a previously blank CCTV monitor. The mon-itor displays the
intruder and the exact location within the scene by some
flashing indicator superimposed on his exact location.

Management uses AVMD, DVMD, and video recorders to test a
security plan and guard response, and evaluate guard and
overall system performance. A system using the motion detector
permits security personnel to train before an actual event, and
when an intrusion does occur, the sys-tem can immediately
recall the decisions to form an instant plan of action. This
directs the efforts of the response force
For several decades, the AVMD has attempted to identify
motion and activity of interest in a video scene. It has enjoyed
some degree of success for indoor applications but has not been
successful in outdoor environments. With the recent
introduction of the DVMD in conjunction with DVRs and
digital multiplexers, VMD has now become an important, even
essential, tool for video monitoring.
The AVMD system is simple: it monitors any change in the
video signal that comes from the camera and produces an output
indicating that there was an alarm. Unfortu-nately, many other
changes in light levels are not caused by targets of interest but
rather from background changes. The particular causes for these
false alarms are:
· An overall change of the scene lighting caused by sud-den
light changes or fluctuations in overall lighting, and turning
lights on and off
· Flashing a light across a scene causing an immediate contrast
change
· Open flames, flashing neon signs, cigarette lighters
· The sun passing behind a cloud
· Flying debris: flying paper boxes, etc. through the cam-era

FOV
· Environmental dust, a rainstorm, or snowstorm
· Animals, birds passing through the camera FOV
· Continuous motion from water fountains, revolving doors,
escalators, ripples on water, or wave motion.
For all these reasons, the AVMD is not a viable solution for
detecting motion, real target, or activity in a video system, and
does not find widespread use except in small systems.
13.4.1 Technology
The AVMDs have been available for many years and pro-vide a
low-cost video device to detect simple motion in a video scene.
They operate reliably only in indoor, well-controlled
environmental and lighting conditions and should not be used
for outdoor applications. Figure 13-3 shows a block diagram of
the AVMD.
The simplest AVMD uses analog subtraction. The refer-ence
frame and the frame in which motion has occurred are
subtracted and an alarm declared depending on the amount of
signal difference between frames. This analog system, while
acceptable for most indoor applications, is prone to false alarms
and is not suitable for outdoor appli-cations. A digital DVMD
should be used in all outdoor applications.
359

SECURITY
OPERATOR
CONTROLS
RESET
VIDEO OUTPUT
VIDEO
MONITOR

SENSITIVITY
OUTPUT
RECORDER
INTERFACE
PRINTER
CAMERA
WINDOW
ALARM OUTPUT

VIDEO
SIGNAL
AUDIBLE ALARM
SIGNAL
SIZE,
SHAPE,
LEVEL
VISUAL ALARM

CONDITIONING
LOCATION,
COMPARATOR
LOGGING PRINTER
GENERATOR
REFERENCE:

AUTOMATIC
ADJUSTMENT OF
SLOW CHANGES
IN LIGHT LEVELS
FIGURE 13-3 Analog video motion detector (AVMD) block
diagram

Two generic detection options available in many VMDs are: (1)
detection of motion or activity, (2) detection of the presence or
absence of an object. These systems can be configured so that
these two different type windows operate independently and be
can be combined within the same camera FOV. Motion windows
are designed to detect movement of objects or personnel into
and through their detection zones. They also detect anything
that moves into the window and stays there even though the
object stops moving. They can have a programmable time-out
feature so that an object can enter the detection window and
stay there for a given length of time without causing an alarm.
This ensures that the DVMD does not indefinitely remain in an
alarm mode. The motion windows look for significant changes
in image contrast or pattern in the detection zone. They detect
only significant changes in most objects that are bright or dark
but are much smaller then those expected from some debris, and
will not trigger a false alarm.
In the object presence or absence mode of operation, the system
monitor displays the movement of objects that are expected to
remain stationary during the surveillance while ignoring
surrounding movement. If particular assets are to be protected
and can be defined in space, the VMD defines a tight window
around the object to instruct the system to signal an alarm if the
object moves while ignor-ing anyone passing through the FOV.
When using either of the two modes the individual windows are
augmented by background scene monitoring functions so that
the overall scene illumination levels are monitored to detect and
compensate for sudden light level changes.
All AVMDs have an adjustable detection-of-motion zone
(DMZ), which is a selected portion of the monitor screen. Any
movement (change of light level) in the scene within the DMZ
automatically triggers any one of four alarms:

· an internal audible alarm, (2) a front-panel signal light, (3) an
AC or DC outlet that can activate an AC- or DC-operated
signaling device, or (4) an isolated terminal relay contact to
activate a video recorder, printer, bell, or other security device.
On most AVMD equipment, the size, shape, and loca-tion of the
active area in the entire scene is adjusted with front-panel
controls. The DMZ size and configuration chosen depends on
the requirements of the surveillance application. Figure 13-4
illustrates some examples of DMZ shapes available, including
split-screen, square, rectangle, L-, C-, and U-shaped.
The areas of sensitivity are chosen to surround a location in the
scene where motion is expected. The DMZ enables the operator
to select (sensitize) specific portions of the camera scene area,
while the entire scene is always dis-played. An alarm occurs
only if there is motion in the DMZ itself. Depending on the
equipment, DMZ is represented on the video monitor screen by
a brightness-enhanced window (or a brightness-enhanced
frame), adjustable via the front-panel controls. After initial
setup, the brightened window (or frame) may be switched off so
that the scene looks normal to the operator. The active DMZ on
the screen can be set up to cover an area anywhere from 5 to
90% of the viewed picture width and height. The AVMD system
sensitivity is usually set to respond to a 25% change
MONITOR SCREEN DISPLAY

SPLIT SCREEN SQUARE RECTANGULAR
L SHAPED C SHAPED U SHAPED
VMD SENSITIVE TO ALARMS IN CROSSHATCHED
AREAS ONLY
FIGURE 13-4 Detection of motion zones (DMZ) in analog
video motion detectors (AVMD)
in video signal level, in 1% of the picture area occurring within
a time period of several frames.
The AVMD operates by analyzing the analog video sig-nal from
the camera and determining whether the scene has changed. The
system “memorizes” the value of a stan-dard reference scene
depicted within the DMZ and com-pares it with a value in the
current real-time scene. If the two values are the same within

the active DMZ, electronic circuitry declares that there has been
no motion and no alarm is declared. On the other hand, if there
has been a scene change caused by someone intruding into the
scene, an object moving, or some other light-level disturbance,
providing the change is larger than a prescribed amount,
typically 10–25%, then electronic circuitry decides that a
change has occurred, there has been motion in the alarmed area,
and an alarm signal is produced. This alarm signal is used to
produce an audible or visual alarm, turn on or activate a video
printer. The AVMD operates inde-pendently of the video
monitor or any other recording equipment, and in no way
interferes with it.
13.5 DIGITAL VIDEO MOTION DETECTOR (DVMD)
While analog VMDs have been in use for security appli-cations
for many years, they have only been moderately
successful in indoor applications where lighting has been well
controlled. In outdoor applications, a far more com-plex digital
electronic system is needed to provide reli-able VMD
capability. The DVMD must take into account the many
variations of lighting, type of target movement, and electrical
background disturbances caused by exter-nal sources and noise
in the system. In the past, these sophisticated expensive systems
have been used in large government facilities and nuclear power
plants. With lower cost derived from high density memory and
more power-ful computers, the DVMD is now in more
widespread use in commercial installations.
The DVMD allows the user to divide the monitor’s video scene
into small detection areas called windows, and in some cases

even smaller size areas going down to the pixel level. The
flexibility of these windows allows the user to specify
particular areas or zones of interest. Each window or zone has
its own set of programming levels for sensitivity and alarm
triggering level. Only the windows are activated or processed
for alarm events: all the other parts of the scene which either
are not of interest or may contain false alarm producing motion
are not. Using this technique, doors to a building may be
monitored while headlights from an adjacent car parked or other
bright lights in the scene are ignored. Since average light-level
changes in the scene occur, the system automatically adjusts to
both increasing and decreasing illumination by monitoring and
updating reference levels for each video input. The entire scene
is also continually monitored for light and illumina-tion
changes and full image scene changes such as those caused by a
lightning or clouds drifting in front of the sun. The scene
changes would not trigger an alarm but rather reset the
references for each window, and the VMD would continue
monitoring the detection zones for motion or inactivity. The
sensitivity of each window is monitored and controlled by the
user.
The more sophisticated and expensive DVMD systems use
elemental detection zones, in which the scene is divided into a
large number of zones (hundreds to thousands) and converted
into a digital signal. The processor analyzes these individual
zones and makes a decision whether or not an alarm is present.
With these microprocessor-based systems, many parameters are
ana-lyzed, thereby forming a more reliable basis for an alarm
signal decision. Light-level changes in these DVMD sys-tems
are compared with the previously stored values ratio-
metrically—that is, on a percentage basis. Ratio-metric
thresholding causes the system to cancel out any gross change

in the scene lighting, so that an alarm decision is made strictly
on an incremental basis, for a small portion of the total picture
area.
The digital electronics in the DVMD subdivides the cam-era
scene into many small elemental zones—as many as 10,000—
and makes a zone-by-zone comparison (subtrac-tion) of the non-
moving or steady scene with the motion scene. It goes into an
alarm mode when a threshold is detected in any one or a
multiple of these zones. By con-verting the signal from analog
to digital and dividing it
361
into many zones, a much more sensitive device results. This
technique allows discrimination between real targets and false
alarms and other scene lighting variations, and provides a more
reliable system for outdoor use.
The user-selected zones are positioned over specific areas
where motion is expected. These zones may cover assets to be
protected, entry or exit points, parking lot slots, perimeter
areas, and perimeter fence lines. Each zone may be set with a
different sensitivity appropriate to the per-centage change
required to trigger the alarm in that zone. The larger the
percentage required to cause an alarm, the less sensitive the
system is to contrast changes and the less likely it is to produce
false alarms. The DVMD is much more sensitive than the large
area detection AVMD.
13.5.1 Mode of Operation

The DVMD processing unit converts an analog video signal into
a digital code and performs DSP to make it sensitive to specific
types of motion in the camera scene (Figure 13-5). For each
camera a specific detection pattern or area is selected, or
already programmed into the electronics memory. The detection
pattern is part or all of the camera image scene within which
specific sample points are desig-nated. Depending on the
manufacturer, the sample points vary in number and location. At
a designated rate, the sam-ple or reference image from a
specific camera is converted from the analog to the digital
format, and the digital val-ues are stored in temporary memory
in the VMD unit. This reference or base image is updated at
variable rates
OPERATOR CONTROLS:
SCENE LEVEL

DIGITIZED
CELL
VIDEO OUTPUT
ANALOG
VIDEO

DIGITAL
PREPROCESSOR
PRINTER
VIDEO IN
CONDITIONING
PROCESSOR
CONVERSION

VIDEO AND ALARM/CELL
ANNOTATION (TO MONITOR)
SECURITY OPERATOR
INPUT
FIGURE 13-5 Digital video motion detector (DVMD) block
diagram
362
CCTV Surveillance
to compensate for small changes in the scene that do not
target is viewed from a distance it appears to have a small
constitute alarm events.
size on the monitor image. As the target moves closer to
At programmable rates at a later time, the camera
the camera it increases its apparent size thereby causing
images are converted into a digital format and electroni -
the confusion in target identification. Motion detectors
cally compared with the stored reference image. If there
generally have a more positive identification of a target

has been movement in the scene or any variation in a
if the target is moving perpendicularly or at an angle to
significant number of sample points over some range, an
the camera, rather than toward or away from the cam-
alarm is triggered. If some harmless objects such as a
era. If cameras can be mounted to have this relation-
small animal or bird or debris pass through the scene no
ship to the target, a positive identification can usually
alarm will occur. If, however, there is movement within the
be made. The most significant new parameters added to
scene—such as a person entering a window or opening or
digital VMD processors to improve the capability for out-
closing a door—the VMD will be triggered. The number
door operation have been: (1) improved multi-directional
of sample points and the amount of change within the
detection, (2) 3-dimensional perspective analysis, and (3)
areas to produce an alarm output depend on the particu-
automatic adjustment to changing environmental condi-
lar manufacturer, model, and operator control settings.
tions. Improved multi-directional detection provides the
Depending on the design, a VMD can process 1, 10, 16,
ability to determine whether the object is moving directly
32, or 64 cameras and sample them serially: that is, camera
toward or away from the camera, especially when the tar-

1, then camera 2, and so on, and then back to camera 1.
get is at a distance. The ability to automatically adjust to
Some systems sample and process multiple cameras simul-
changing environmental conditions removes the technical
taneously, then analyze and respond to multiple alarms.
difficulty to manually readjust the system sensitivity set-
When a VMD detects an alarm event its output can be used
ting to match daily weather variations. Systems not having
for multiple functions. It can display the alarmed camera
this ability are difficult to calibrate and require constant
on a monitor, alert a guard with a visible or audible sig-
recalibration.
nal, record the alarm on a video recorder, send the alarm
signal to a remote site, or activate a TL VCR or DVR with
an alarm input to change its recording mode from TL to
13.5.2 Technology
real-time.
In contrast to the AVMD that detects the change in light
When a video image is converted to data in a digital for -

level in one or a small number of scene locations (zones),
mat, the image information becomes the stored digital
the DVMD electronically analyzes hundreds or thousands
value. This digital value changes as the video image (the
of zones in the video signal and provides information such
source of the data) changes. Complex algorithms analyze
as the location in the picture where a motion or intrusion
has occurred. Its output drives various audible and visible
the changing digital values to recognize patterns. This is
alarm signals, a graphic monitor map showing the motion
considered as video content analysis. These algorithms are
path in the image, and a record of the intrusion using
a software function and are programmed into electronic
a recorder or video printer. In normal operation when
chips that can be installed in cameras, standalone mod-
there is no motion or change in a scene, the VMD takes
ules, DVRs, and dedicated computer processors. DVMD

the video signal from the camera, stores the video frame
is also available as software for installation in off-the-shelf
(containing no motion), continually updates and memo-
computers.
rizes the subsequent frames, and compares them to the
Algorithms have been designed to decrease the number
previous frame to see if there is a difference in the new
of monitors that must be viewed. This is accomplished
frame. If there is no motion there is no alarm. If there
by scene averaging and filtering techniques to eliminate
is a difference of measurable and defined value, then an
items that do not fit the model of the motion or activity
alarm is declared and an output produced.
and do not represent a threat to the site. Once the system
Caution must still be taken for outdoor applications,
detects an object, it applies various tests in an attempt to
however, in which there are rapid changes in sunlight,
classify the object, taking into account such characteristics
clouds, shadows, distance of objects, rain, snow, movement
as size, shape, true height to width ratio, and location. If
of trees or shrubbery, camera movement in winds, automo-
the object or activity fits one of the criteria for a target, it
bile lights, ripples on the water, and other small moving
is marked and a more accurate determination is made to

objects. This can represent a fairly impressive range of
identify personnel and activities.
problems that must still be considered in outdoor appli -
Digital VMD technology has the ability to monitor every
cations. To address some of these problems, DVMD sys-
pixel of every image individually and/or as a group. The
tems have additional automatic adjustments (algorithms)
light level of each pixel can be memorized in storage and
to process the visual signal data to exclude some of these
compared to subsequent images to determine if there is a
problematic false alarms. One problem, in particular, is
light-level change and how much the change is. By apply-
to determine the size of a target in the scene. When a
ing this technology over the entire image, the light-level
changes in each pixel can be examined and a determi-nation
made whether it fits the criterion of an alarm. Algorithms are
designed to identify objects of specific size, shape, movement,
etc. on a pixel-by-pixel basis. Flying debris and other false
alarms can be filtered out by size, object direction and speed,
color, and type of motion and pattern.
Determining the size of an object in the FOV is difficult since
the object appears as a different size depending on its distance
from the camera. If the object is close to the camera it is large
and as it moves away from the camera it becomes smaller and

smaller. For this reason, parameters such as shape and
movement are also required to deter-mine the identity of the
object. Object direction can be determined easily since the
object activates many pixels and by keeping track of the left-to-
right or up-and-down motion it is easily accomplished.
In some cases the color of the object may be useful, and this is
easily determined in the color camera by mon-itoring the color
of each pixel in the moving object. This can be important if a
person with particular color cloth-ing has been identified as the
target. The parameter of color is used to continue tracking that
person. Likewise, in outdoor applications if an automobile is
identified with a particular color, the color might be the most
important criterion for tracking the vehicle. Environmental
condi-tions producing dust, fog, rain, snow, and sleet produce
some ambiguity in target detection. These disturbances
generally reduce the range over which VMD is effective.
Combining object motion and pattern recognition can provide
additional information in determining the identity of a person
and the behavior of the target. Algorithms have been devised to
identify the movement of a person walk-ing. They have been
able to tell the difference between a person walking, a walking
dog, a crawling man, and oth-ers. There are also various
motions that an intruder or criminal makes as compared to our
normal movement, and these abnormal motions can be saved
and put into storage and can help to identify a person exhibiting
such movements in the video image.
An object’s speed is used by setting criteria for how fast the
object of interest is able to move, and if the object is moving
faster or slower than a predetermined speed it is registered as a
false alarm. The VMD can have a library that stores information
about the unique movement and pattern of particular objects
such as paper leaves, ripples on water, birds.

The DVMD has the ability to remove constant motion from the
scene which often takes the form of rain storms, snow, sleet,
hail, water fountains, waves on water, etc. Algo-rithms stored in
memory are used to filter out these con-stant motion
disturbances. If there is an object within suchconstant motion
moving at a different speed the system is able to identify this
target.
The DVMD digitizes the frames from each camera into a large
number of zones corresponding to exact locations on
363
the monitor screen. The number of digitized zones varies from
hundreds to many thousands. The system assigns an absolute
gray-scale value (light level) to each zone and stores the
digitized gray-scale value and location in RAM. This procedure
is carried out for each video camera chan-nel. The DVMD can
digitize the picture into 16–256 gray-scale levels, thereby
storing (memorizing) the image scene very accurately. After
this reference scene has been mem-orized in RAM, the DVMD
digitizes subsequent camera frames and compares them to the
stored values, zone by zone. If the stored levels at any location
differ by one or two gray scale levels—between the stored
frame and the live frame—an alarm condition exists.
Most DVMDs in use today use standard menu screens to
monitor and respond to alarms, using either simple keyboards or
a mouse device for programming, adjust-ment, and normal
operation. Most current systems do not require a personal
computer (PC) for operation, but all provide an RS-232
interface for computer integra-tion or remote programming and
reporting. The RS-232 approach and menu-driven screens for
operation and con-trol of the digital VMD systems provide a

friendly interface to the user.
Self-contained DVMDs are based on proprietary signal
processing algorithms and easily integrate into existing multi -
camera video systems. Most camera inputs are digi-tally
sampled with a resolution of 768 by 480 pixels and eight bits
(256 levels) of grayscale. All images are sampled and displayed
at 30 fps (60 fields per second). Each cam-era is associated with
a dedicated event when an alarm output occurs, and can be
connected to a video recorder or audible or visual anunciator
whenever any window in any camera has been alarmed.
Additionally, a video loss output signals an alarm if the camera
loses power or no video signal is present, and remains active
until the video signal is restored or the time-out feature resets.
Many DVMD systems have two monitor outputs although only
one monitor is required for viewing. Many users prefer a dual -
monitor approach. One monitor is used to view live sequencing
from camera to camera or a specific camera view. The second
monitor is used in dig-ital mode to view motion detection
windows triggered by an alarm. When alarms occur from
multiple cameras, the operator can sequence through the
alarming cameras at a user-defined rate or go to the quad or 9 or
16 split image display with the alarming cameras in that mode.
In any case, the images from the alarm cameras are highlighted
graphically on the display. The minimum hold time for each
alarm is user-defined, usually from several seconds to 5–10
minutes. The user can also select freeze times for any of the
alarmed images ranging from seconds to minutes. In the freeze
frame mode the video display is locked into a full screen. When
an event occurs in that camera after the freeze frame time has
elapsed, the video continues in full motion allowing the guard
to continue monitoring the cameras.

In the playback of recorded images from the VCR or DVR the
output can be displayed on either or both mon-itors. This allows
one monitor to be left in the normal display mode monitoring
potential alarms while the other plays back the recorded images
for review.
13.5.2.1 Programming the digital VMD
The DVMD includes an RS-232 interface to allow the user a
choice of using either the front-panel controls or a mouse for
system setup. Either way the window placement, size, or
sensitivity are simply defined. Each camera can be programmed
to include numbers and titles defining the specific camera,
which is later displayed whenever that camera is displayed.
These titles may be positioned anywhere within the full screen
window so as not to obscure any important areas in the image.
The system utilizes pull-down programming menus to control
split-screen sequence rates, the camera ID information, and any
other titles. Menus are available to adjust the sensi -tivity and
scene area balance of the pixel level for alarm functioning.
Some systems can provide not only intruder detection but also
lost object detection. Even in the presence of multiple moving
objects in the same window, intelligent DVMD systems can
accommodate a rapidly changing illu-mination condition
commonly found in outdoor scenes, as well as sudden
illumination changes from man-made and natural sources.
13.5.2.2 DVMD Setup Procedures
The DVMD system uses graphic symbols for motion sen-sitivity
settings, simplifying the motion detection setup. In addition to a
flashing cursor on screen, text prompts appear as shown in

Figure 13-6.
Cameras can have motion detection in particular areas in the
scene completely disabled. This should not be confused with
enabling or disabling individual zones or pixels in areas of
interest. Disabled zones that may con-tain unimportant or
incidental movement include the following:
· Trees that can sway in the wind
· Pedestrians and vehicular motion that is not important
· Reflections from glass, bodies of water, or other highly
polished surfaces, which can be sources of apparent motion.
The different alarm zones can be designated on the monitor in
different colors for identification purposes. Examples are:
Choice Color of Flashing Cursor
No action Gray/white
Enable zones Black/white
Disable zones Clear/white
ACTIVE ZONE SETUP
ALARM AREA OF ACTIVITY

DISABLE PROBLEM ZONES
PEDESTRIAN AND VEHICLE
MOTION THAT IS NOT
IMPORTANT
MOVING TREES, BUSHES
CLOUDS IN SKY
RIPPLES ON WATER
REFLECTIONS FROM GLASS
OTHER
MOTION
SENSITIVITY
GRAPH
FIGURE 13-6 On-screen digital video motion detector
(DVMD) graphic display
13.5.2.3 Sensitivity Settings

A bar graph is often used to illustrate the alarm sensitivity
setting for the camera. The bar graph displays the sensi -tivity
setting as a red line. A black line moves from the bottom to the
top of the bar to indicate a change in motion or activity in the
scene. When the black line reaches the red line above, a motion
alarm is activated (Figure 13-7). The user selects a number or
sensitivity button between 1 through 10 to change the
sensitivity. In practice, watchingthe scene from a camera and
watching the motion helps todetermine the appropriate
sensitivity setting for the cam-era. This procedure is performed
for each camera during the initial setup phase of the system.
13.5.2.4 Motion Detection Sensitivity
Motion detection sensitivity for each camera can be set to levels
from 1 through 10. The setting is made on a camera-by-camera
basis, and applies to all enabled zones in any particular camera
scene. Each of the zones distinguishes among 256 grayscale
levels averaged over each zone’s area. A sensitivity of 1 is the
least sensitive to motion and a set-ting of 10 is the most
sensitive to motion. These settings are made using a bar graph
similar to that used in the sen-sitivity settings above. Some
recommendations for setup are listed below:
· If motion detection activates without an apparent cause,
reduce the sensitivity.
365
· When setting sensitivity, select the highest setting that does
not result in frequent false motion detection.

· The higher the sensitivity, the more likely the incidental
movement to be detected as motion.
· When setting high sensitivity, such as 8–10, sources of false
motion like reflections and windblown trees should be absent,
otherwise alarms will occur.
The DVMD used as a sensor activates alarm inputs, essen-tially
creating a motion-based alarm sensor input. The system in this
scenario does not distinguish between an input from an external
alarm sensor (switch, PIR, glass break detector, etc.), or when
activated internally to the VMD system.
13.5.3 Hardware
Some DVMDs monitor up to 32 separate video cameras by
sampling, time-sharing each camera sequentially. Each camera
can have a separately adjustable sensitized alarm-ing area,
thereby optimizing each camera to the scene it views. Likewise,
the number of sensitive zones in each cam-era is chosen
independently to match the scene require-ment. If one camera
views a large area scene looking for small intrusions, the
operator can make the alarming zone small for this first
channel. If another camera views a small area scene looking for
large intrusions, the operator can make the alarming zone large
for this channel, and so on. Equipment setup procedures differ
from manufacturer to
ACTIVE ZONE SETUP
AREA AROUND THE HOUSE ALARMED

GRAPH CHANGE INDICATES DETECTION
MOTION
SENSITIVITY
INDICATOR
PEDESTRIAN
MOTION
DETECTED
IN ONE CELL
FIGURE 13-7 Bar graph sensitivity display
manufacturer, but there are some common parameters and
controls that must be determined and set when ini-tially
installing the DVMD system. Typical setup controls include:

· Channel Mode Control. A switch selects the mode for each
video camera channel. In the down position— INHIBIT—the
channel is disabled and no alarms are registered. In the middle
position—NORMAL—the cameras are ready for motion
detection and alarming. In the up position—SET—the console
operator can man-ually select any camera on the alarm monitor.
When released from the SET position the switch returns to the
NORMAL mode.
· Alarm Area Control. The alarm area control lets the operator
manually adjust the position and size of the alarmed area zone.
These adjustments can desensitize areas of the camera’s FOV
where normal movement would cause an unnecessary alarm. For
example, in an outdoor scene where a flag is constantly waving,
the desensitized area would appear on the monitor but
movement within that area would not cause an alarm.
· Refresh Control. The refresh rate refers to the time interval
during which the reference frame memorized in RAM is stored,
before it is again updated. Systems use refresh rates varying
from 1/30 second up to several seconds. The operator selects the
refresh rate, which is normally a function of the number of
cameras and the kinds of alarms expected in the scenes.
· Ranging Control. Most systems allow adjustment of the
electronic analog dynamic range of the analog-to-digital (A/D)
converter. The function of the A/D converter is to change the
camera’s analog electronic video signal to digital values. To
provide the best scene resolution for each camera, the operator
adjusts the range of white to black level in the digitized video
signal.
· Masking Control. The masking control allows the oper-ator to
enter scene areas on the monitor screen for which no alarming
will occur. It is entered by inserting rectangular, square, or

other masked areas. In some sys-tems the operator enters the
masking with a light pen. The light pen permits irregular shapes
to be desensi-tized merely by drawing around the object in the
CCTV monitor scene.
In many VMD systems the detection zones may be of any shape
and be divided into separate areas to accommodate unique
detection requirements. Zones can be individu-ally turned on or
off to accommodate entrance, hallway, parking area, or other
locations. Two examples of zones being turned on or off
individually are the following: (1) a zone encompassing a gate
or doorway can be turned off during shift changes while other
zones in the same scene can remain active to alarm and alert an
operator of unauthorized intrusions, (2) a zone encompassing a
file cabinet can be left off during normal working hours and
turned on overnight. The systems can have independent
16-step zone sensitivity, signal integration (retention), plus
multilevel digital filtering to maximize motion alarm detection
and minimize false alarms. Periodic automatic rebalancing
minimizes the effect of slow light changes, such as those
occurring between daylight and nighttime conditions.
In operation, a cell is activated by the changes in the video
content of successive picture fields. A higher retention setting
delays the automatic rebalancing to opti-mize detection of slow
changes or slow-moving objects. Both the video change
(sensitivity) and the rebalancing time (retention) assigned to a
zone can be adjusted to opti-mize detection and minimize false
alarms for that zone. Any activated cell in a zone alerts
(activates) that zone and channel.
Systems have integral video switchers with dual video outputs
and RS-232 port to allow the DVMD to function as a standalone

system. An audio output is available to warn the operator of an
alert, and a relay closure can start a recorder for recording
alerted channels. The RS-232 ports provide both a control input
and an alarm output. They permit remote system control via a
separate control keyboard, a data terminal, or a computer.
Either of the two on-screen alert presentation modes may be
selected to highlight the intruder’s path through the facility.
They are normal or trace.
13.5.3.1 Normal Mode
In the normal mode, a bright dot is displayed in the pic-ture on
the alarm monitor at the center of each activated cell. With
manual reset, this dot remains lighted until the channel is reset.
With automatic reset, each dot disap-pears 16 seconds after the
cell was first activated. Thus an intruder moving into a zone
will cause a series of dots to appear as he first activates cells
and leaves a trail of dots through the zone or to the point in the
zone where he stopped or hid.
13.5.3.2 Trace Mode
In the trace mode, a bright dot is displayed in the picture on the
alarm monitor at the center of each activated cell as in the
normal mode. In addition, each illuminated dot emits a quick
burst of flashes 8 seconds after it is activated. With manual
reset, this results in a continuous moving trail of flashes at 8-
second intervals along the path of intrusion. With automatic
reset, a single burst of flashes occurs before each cell is
automatically reset. These flashes can assist the operator in
determining the size, direction, and location of an intrusion.
Larger monitoring sites require more cameras and mon-itors and

a more comprehensive DVMD digital system. A high-speed
microprocessor analyzes detected motion for size, position, and
rate of movement to discriminate against undesired targets and
to verify a valid intrusion
before the system signals an alarm. Verified intrusions ini -tiate
audio and visual alarm signals. Video from alarmed cameras are
connected to outputs for an alarm monitor, a recorder to monitor
and record the track and position of intruders. Independent
output relays provide control of external devices. A built-in
sequential switcher provides normal system viewing of all
cameras by separate video output.
For ease of operation, some systems have user-defined detection
of active areas initiated using a light pen. Zones can be
individually deactivated while observing the pic-ture to
eliminate detection of areas where insignificant or acceptable
motion could cause some false alarms. The systems have the
ability to perform target discrimina-tion. Each camera module is
programmable to optimize target discrimination based on a
combination of antic-ipated characteristics, such as size, rate of
movement, and indoor/outdoor scenes. In order to see the
intrusion track and position display, zones where motion has
been detected are highlighted on the video displays.
The system microprocessor analyzes the cell data and removes
background clutter and identifies any changes in the cells as
targets to be tracked. The target’s motion, speed, direction, and
distance traveled are analyzed to see if they match the
characteristics of a human intruder. When a human intruder is
identified, on-screen graphics highlight his position and an
alarm is signaled.
Special setup graphics define the camera zones to be monitored.

Target discrimination is based on target size, contrast, speed,
and direction. Target tracking is used to verify detection before
declaring an alarm, resulting in a low false-alarm rate. The
operator sets up sharply defined detection zones configurable
for each camera, which may be tailored to reflect the optical
differences between near and distant areas and act as distance
compensation.
Digital video motion detectors are available in sizes suitable for
small to large video surveillance systems (Figure 13-8). A
family of products available is suitable for a single channel or
four channels all provided with DSP electronics and
microprocessors to analyze the entire video scene up to 30 fps
for precision video detection of motion. At each update the
system measures the pre-cise change in each pixel’s gray-scale
level, i.e. the change in light intensity. These DVMD units are
small in size, easy to install, and have simple pushbutton access
for on-screen menu programming. They have access codes and
password protection to protect against unwanted changes in
programming by unauthorized personnel. The motion detection
criteria include duration of motion and sensi-tivity. There are
99 levels of sensitivity permitting use in a variety of lighting
situations. The 4, 9, and 16 channel units have built-in
sequential switchers and provide alarm and video output from
alarmed cameras. Alarm outputs can trigger TL VCR and DVR
recorders, matrix switchers, quads, video printers, or video
transmission devices.
367
One system has the ability to cascade up to 16 of the single
channel units via a single host RS-232 serial port (Figure 13-
8d). Figure 13-9 shows a block diagram of the multiple camera
VMD system. One DVMD digitizes the scene by creating up to

16,000 individual zone locations per scene in up to 16 camera
scenes. With this high resolving power, the system can detect an
intruder occu-pying as little as 0.01% of the area. The DVMD
system operates normally with a blank monitor. When a camera
receives or detects motion, an audible alert is sounded and the
disturbed scene appears on the monitor. The DVRs are activated
for recording the intrusion scene or for reviewing the alarmed
scene at a later time. When the DVMD displays the picture on
the monitor, the guard sees the intruder in the scene even
though he occupies only a small portion. The guard will also
know where the intruder is, even if he is hidden from camera
view, since the system displays the intruder’s path on the
monitor. This display is accomplished by displaying bright
flashes on the monitor at all locations the intruder has passed
through. The guard now knows not only which scene was
intruded upon but also the exact location of the intruder in that
scene at that instant. He can therefore concentrate immediately
on what decision to make and what action to take.
There is no industry standardization for the design and
specifications of AVMD or DVMD systems. The fea-tures of
some representative VMD systems and specific attributes are
described in the following sections and Table 13-1.
13.5.3.3 DVMD Graphic Site Display Maps
An auxiliary display useful with VMD systems is an illumi-
nated graphic display consisting of an overlay that is a plan
view diagram of the entire monitored site. The map over-lay
shows the location of each camera and alarm sensor, and flashes
on the display when an intrusion occurs. To ensure that no
intrusion is missed, particularly if there are simultaneous
intrusions or motions in the scenes, video recorders are used.

The recorder records the video scene, the intruder, his track
through the scene, as well as a graphic alarm map if available.
In the event of multiple video alarms in a single recording
system, the recorder is set to record one alarm scene for a
predetermined time interval and then switch to the next alarm
scene. If a non-video sensor detects an alarm, the system acti-
vates the appropriate camera(s) and the recorder. The displayed
information enables the console operator to assess the situation
rapidly and accurately and report any diversionary tactics.
Present DVMD equipments are able to detect 20 times the
number of intrusions as those detected by a guard looking at the
video monitor without the benefit of the DVMD. This DVMD
system is not easily mesmerized!
(B)
(A)

(C) (D)
FIGURE 13-8 Single and cascaded-single digital video motion
detector (DVMD)
SECURITY OPERATOR
CONTROLS
CAMERA
1
VIDEO OUTPUT

FIGURE 13-9 Multiple camera digital VMD block diagram
The DVMD analyzer detects the alarm condition by storing the
scene in solid-state RAM. In one system, the storage process
takes approximately 33 milliseconds and consist of sampling the
picture scene (up to 16,384 dis-
crete locations) that are spaced throughout the scene. At each
location the brightness is measured (one of 256 dif-ferent gray-

scale levels). The address (pixel location in the scene and
camera) is stored with the brightness number.
DISPLAYED VMD *
INFORMATION
ACTIVE AREA
MASKED AREA
MOTION ALARM
LOCATION OF ALARM:
SIZE OF MOTION AREA
(H × V) PIXELS)
MOVEMENT OF ALARMED † AREA-TRACKING
SETUP PARAMETERS
SENSITIVITY **
SIZE OF ACTIVE AREA (ZONES)
(NUMBER OF H × V PIXELS)
NUMBER OF ACTIVE ZONES
SHAPE OF ACTIVE ZONE(S)

DISABLED ZONES (ZONE MASK)
(SIZE, SHAPE, NUMBER)
PROBABILITY OF DETECTION ** ALARM LEVEL
CONTRAST
369
FEATURES
ON-SCREEN SETUP MENU
VIDEO LOSS DETECTION
NTSC/CCIR/PAL FORMATS
CONTROL P/T/Z
ALARM INPUTS
ALARM OUTPUTS
PASS THROUGH VIDEO
· ON-SCREEN DISPLAY: VARIES WIDELY DEPENDING ON
SPECIFIC EQUIPMENT
** WHEN SET UP OPTIMALLY:

TYPICAL PPROBABILITY OF DETECTION—BETTER THAN
96%
TYPICAL NUISANCE ALERM RATE—LESS THAN 2%/DAY
TESTS BASED ON INDUSTRY STANDARDS
· AVAILABLE ON SOME MODELS
Table 13-1 Digital Video Motion Detector (DVMD)
Features
This occurs for all zones in the scene. After the bright-ness and
location information are stored, a comparison process is
initiated that compares the present live picture from the camera
(which the camera generates 30 times a second) to the stored
picture. Whenever there is a bright-ness discrepancy in any
zone, the address of that particular zone location is also stored
with its brightness value. Zone locations where these
differences are caused by electrical noise or ambient scene
motion such as blowing leaves, trees, or flags are processed out
and are not considered as alarms. All scene areas where
detection is not desired are removed or masked out.
When a sufficient number of zones change, an alarm is
processed. The comparison process occurs across the entire
scene 30 times a second. The alarm condition is established by
counting the number of locations with dif-ferent values; if a
preset threshold count is reached (any number, but generally 1
in 8 counts), the system then alarms. The count is cleared each
time a new storage process takes place. The memory is
refreshed on a pre-set basis and ranges from 1/15th of a second
to many seconds. Memory refresh prevents normal changes,
such as scene lighting, moving clouds, or electronic drifts in the

camera from being interpreted as alarm conditions. The camera
viewing the intrusion scene is automatically switched to the
monitor (any standard video monitor) and the scene displayed.
The monitor is usually blank prior to an alarm, since there is no
reason to display the scene if no activity is occurring. Table 13-
2 summarizes the parameters of several commercially available
digital VMD systems.
13.5.4 Features
VMD technology is not standardized, and therefore selecting the
appropriate VMD approach requires under-standing the VMD
features available and requirements of the application. Basic
motion detection typically recog-nizes any type of motion in the
camera FOV. A single output then activates automatic call up to
the monitor screens for the surveillance personnel and initiates
auto-matic VCR or DVR recording. With the advent of LANs,
WANs and the Internet, the video call up is no longer limited to
cabled CCTV systems, but can be transmitted over these
communications channels, or even wireless. Advanced VMD
products enhance the concepts of basic VMD through the use of
elaborate algorithms that search out detailed movement
patterns, and only activate a sys-tem response under very
specific conditions. These activity criteria include:
· Intruder Identification: Identifying unauthorized humans in
specified areas of the video FOV.
· Environmental Compensation: Recognizing and ignor-ing
wind-blown debris, animals, background traffic, etc.
· Counting: Recognizing a quantity of a particular object or
number of persons moving through an area.

· Direction: Ignoring objects moving in one direction, while
alarming for objects moving in unauthorized directions (no
identification).
· Item Recognition: Activating when specific user-selected
items are removed from, placed in, or passed through the FOV.
VMD TYPE
CAMERAS
MONITORED
SINGLE CHANNEL
1
SIXTEEN CHANNEL
16
USER SETUP
*
TARGET **

SENSITIVITY
CONTROLS
PARAMETERS
SENSITIVITY
MINIMUM AGE †
OBJECT SIZE
MINIMUM MOVE
OBJECT DIRECTION
(#OF CELLS TO
OBJECT COLOR
CAUSE ALARM)
OBJECT MOTION
TARGET SIZE

MINIMUM VELOCITY
AND PATTERN
RESOLUTION ‡ (PIXEL LEVEL)
720 × 486
260,000
720 × 486
260,000
INPUT/OUTPUT
SIZE (inch)
SIGNALS
VIDEO

SMALL
ALARM INPUT/OUTPUT
1.5×3.5×5
DRY CONTACT
RS232, 422, 485
5 inch
CONTROL P/T/Z
RACK
OBJECT SPEED
MAXIMUM VELOCITY (PER CHANNEL)
MOUNT
· EITHER DONE VIA FRONT PANEL CONTROLS OR
THROUGH SOFTWARE AND COMMUNICATION PORT
· TYPICAL PROBABILITY OF DETECTION–BETTER THAN
96% TYPICAL NUISANCE ALARM RATE–LESS THAN
2%/DAY STANDARD TESTS BASED ON INDUSTRY

STANDARDS
· NUMBER OF FRAMES A TARGET MUST BE TRACKED
BEFORE IT GENERATES AN ALARM. RANGES BETWEEN
1–300 FRAMES
· EACH ZONE IS COMPRISED OF A “BLOCK” OF PIXELS
DEFINING THE ACTIVE OR MASKED ZONE
Table 13-2 Digital Video Motion Detector (DVMD) System
Parameters
· Subject Tracking: Highlighting and following a specific
person or item as it moves about the FOV or from the FOV of
one camera to another.
· Multiple Subject Tracking: Highlighting and follow -ing
multiple persons or items simultaneously as they move about
the FOV or from the FOV of one camera to another.
13.6 GUIDELINES, PROS AND CONS
Some basic questions to be answered where VMD is required:
· Detection: Is there anything there?
· Classification: What is it—a car, person, bird, boat, van?
· Location: Where is it?
· Identification: Is it an unauthorized person?
· Is the person in the correct location at the site?

The security director and managers of a facility and the design
professional who understand the VMD hardware options should
begin a project by asking several important questions:
· What can move in the video image?
· What do we want to know about its movement?
The first objective is to identify what can move. This deter -
mines the surveillance areas to be covered by the cameras and
begins to define the VMD product required. The answer to what
can move includes items of interest and any moving background
items that may distract the sys-tem. The items of interest can be
items that are typically in motion, and therefore either passed
through the FOV or stopped in the FOV, these items require
identification or must be followed by the surveillance cameras.
Some of these moving targets include:
· Vehicles moving through entrances or a prescribed traf-fic
route
· Routine entry and exit by unauthorized personnel
· Baggage left unattended
· Personal property that is carried by the public
· Suspect individuals
· Employee work methods or handling of assets.
Other items or activity that should be of concern include:
· Intruders or unauthorized personnel in an area or perimeter

· Leaks or mechanical failures
· Smoke, fire, or flame
· Violent or erratic behavior
· Counter-flow directional movement.
After the type of movement is understood, the next crite-ria
affecting design and selection should be: What action should be
taken when the motion of interest occurs? Does the alarm of
interest require immediate response? If the incident requires
immediate response, active surveillance personnel must receive
the image and understand what they are seeing. They must also
have instructions as to what action to take for each type of
alarm. If the primary purpose of the video is for documentation
or prosecution or litigation, changes in the FOV to
accommodate the movement should be minimal, and more
cameras should be implemented to confirm the events. In order
to min-imize controversy and to allow acceptance in court, the
graphic enhancement of the VMD, the storage methods for the
video, and the signal compression methods must be closely
scrutinized.
What should be the response and what action, if any, is
warranted on the part of the officer based on the infor-mation
presented? Video-based motion detection systems are providing
many of the answers and solutions to this question.
13.7 SUMMARY
The primary function of the VMD is to allow the security force
to make optimum decisions about an intrusion or unlawful

activity in a minimum amount of time. Profes-sional intruders
and thieves use devious and sophisticated techniques, making
the guard’s response more complex. The intrusion scenario
works to the advantage of crimi-nals because they can spend
time planning it, as well as anticipating the guard’s action under
duress.
The DVMD is a sensitive and valuable video security tool since
it provides security personnel the visual information taken at
the intrusion location when there is motion in the camera FOV.
The intrusion scenario can be displayed on a monitor(s),
recorded on a VCR or DVR, printed out on a hard-copy video
printer, or transmitted to a remote site over a network.
The use of a DVMD significantly increases the security level
and reduces the human error in any security system. The choice
of the optimum VMD for a specific application requires that the
security designer understands the equip-ment capabilities and
limitations and match them to the problem. Of highest
importance is whether the VMD can properly react to the
changing lighting conditions in the video scene and generate
meaningful alarm information and reject false alarms.
371
The present state of the art indicates that AVMDs can operate
acceptably only in well-controlled indoor environ-ments, while
DVMDs can operate in all indoor environ-ments and do well in
most outdoor environments. Because of the variety of
approaches and differences in DVMD equipment, characteristics
of systems manufactured by leaders in the field must be
considered on their own mer-its. Analyzing the systems
described exposes the designer to some of the features available
and permits asking the manufacturer sensible questions to

determine suitability for the problem to be solved. Some helpful
comments and hints follow:
· AVMDs or DVMDs are suitable for indoor applications.
· DVMDs should be used for all outdoor applications.
· The VMD should be able to switch video to a VCR or DVR
and produce a hard-copy video printout.
· Once the VMD system is set up, most of the decision-making
should be automatic.
· Following initial setup, alarm declaration should be automatic,
using a menu-driven program.
An important axiom to remember is that the applicationshould
define the system rather than the system defining
theapplication.
Chapter 14
Dome Cameras

CONTENTS
14.1
Overview
14.2
Speed-Dome Background
14.3
Fixed Dome
14.3.1
Technology
14.3.2
Housing
14.3.3
Hardware
14.4
Speed Dome
14.4.1
Technology
14.4.2
Housing
14.4.3
Hardware
14.5
Dome Mounting Hardware

14.5.1
Fixed Dome
14.5.2
Moveable Speed Dome
14.6
Cabling-Video Signal and Controls
14.7
Special Features
14.8
Special Applications
14.8.1
Outdoor Building Mounts
14.8.2
Pole Mounts
14.9
Summary
14.1 OVERVIEW
Fixed and Speed-Domes. The fixed dome camera has found
widespread use in the video security industry. It has a
monochrome or color camera and a fixed focal length lens. The
camera is often mounted on a simple manually adjustable pan
and tilt mount and the entire assembly mounted on a wall or
ceiling.
The pan/tilt speed-dome has become one of the most popular
scanning video camera surveillance system in the industry. The
primary reason for their popularity is in the large amount of

visual intelligence they can provide to the security operator in
such a small physical package. The speed dome can be mounted
almost anywhere: ceil-ing, wall, building exterior and pole.
High resolution light
sensitive color cameras and compact zoom lenses with auto-
focus mounted in ultra-fast pan/tilt module make them very
effective in most environments including retail stores, casinos,
commercial and government office build-ings, warehouses,
airports, highways, etc.
One technique used to combine the conventional sepa-rate
camera, lens, housing, and pan/tilt video surveillance assembly
is to integrate them all in a plastic dome. The dome housing is
more discrete than most other conven-tional housings. The
dome camera consists of a round or hemispherical clear or
tinted dome in which a camera, lens, and a manual or motorized
pan/tilt mechanism are housed. The ceiling-mounted, below-the-
ceiling, and wall-mounted hemispherical dome looks totally
different from the rectangular housing and other shaped
housings, and blends in well with many architectural décors.
Since the hemispherical dome is circularly symmetrical, it can
be in a fixed position and the camera pointed in any direction to

view the scene. A pan/tilt module in the dome can rotate and tilt
the camera and lens while inside the confines of the dome. This
differs from cameras mounted inside rect-angular housings
where the entire housing assembly and the camera move as one
unit, and the pointing direction is known to the observer below.
If the dome is tinted then the person down at floor level viewing
the dome cannot see the camera and lens, and it is possible to
point the camera in any direction without the observer knowing
it is there, or seeing it move. This capability can act as an
additional security deterrent because the observer does not
know when he or she is under surveillance. Domes are less
obtrusive and generally accepted in any environ-ment. Bullet
cameras (commonly called bullet or lipstick cameras) are
smaller and less noticeable, but they are visu-ally directional
and the viewing and pointing direction is visible.
The moveable speed-dome camera contains an inverted pan/tilt
mechanism suspended inside the dome with an
373
integral zoom lens and video camera module. The dome
enclosure containing the camera/lens and pan/tilt mech-anism
eliminates the precipitation, wind loading, dust, and dirt
problem. The dome pan/tilt design is adaptable for use in
outdoor applications on poles in parking lots, on building
parapets, and under building eaves and passage-ways.
Indoor and outdoor, fixed and movable camera dome systems
are available in many sizes ranging from 5 to 15 inches in
diameter depending on the model. Fixed domes with miniature

cameras and fixed lenses can be small and discreet and can have
a manual pan/tilt adjusted during installation. Speed-dome
systems use high-resolution color and monochrome CCD
cameras with auto-focus and digital zoom, and zoom lenses.
The dome systems include camera pointing presets for pan,
elevation, zoom lens focal length, and other param-eters.
Another feature some dome systems have is privacy zone
blanking that allows specific sections of the camera scene to be
masked so that the operator cannot view scenes at pre-
programmed camera pointing angles and zoom lens ranges. This
prevents viewing the windows of private homes, hotels, or other
buildings in the vicinity of the camera, as well as secured and
classified areas. The zoom lenses and electronic zoom in the
dome cameras can pro-vide powerful zoom capability with
magnifications up to 200 times using electronic and optical
magnification. The systems have sensitive CCD cameras that
provide excellent color viewing during daytime operation and
more sen-sitive monochrome viewing during nighttime
operation. Dome cameras can be equipped with VMD and can
send an alarm signal to the operator if there is movement in the
image when it is viewing a fixed display.
14.2 SPEED-DOME BACKGROUND
There are essentially two types of camera systems that allow the
operator to pan, tilt, and zoom the video image onto the
monitor. The first type of system has been in use for many
years, and uses a fixed camera and zoom lens mounted on a
motorized pan/tilt mechanism. The electronics required for
communications with the camera and platform motors and
switches are installed in a sepa-rate enclosure. This type of
pan/tilt platform is assembled from separate components and
different manufacturers and has several shortcomings:

· The pan/tilt system is bulky and heavy.
· The camera pan/tilt pointing motion is slow—usually less than
10 /sec.
· The camera motion is usually restricted by the inter-connected
cables, reducing the panning range below 360 .
· The cost for this type system is usually more than the newer
speed-dome technology that uses an integrated camera, lens, and
pan/tilt all in one dome assembly.
The new high speed dome systems employ newer more
sophisticated technology having performance character-istics
far superior to the older pan/tilt camera platform system. These
speed-dome cameras are small in size: 5−7 inches in diameter
and contain all the required con-trol and communication
electronics located inside the unit. The dome module weighs far
less than the older pan/tilt platforms so that they can be
mounted almost any-where. The panning speed is typically
300−360 /sec and there are no interconnecting cables so that the
cameras can be continuously panned without reversing
direction. There are various manufacturers that can provide
prod-ucts that have these basic functions.
Prior to the integrated dome with PTZ, pan/tilt plat-forms were
assembled by ordering a housing, a pan/tilt mechanism, a
camera, a lens, and wiring them up before installation. In the
early 1980s, Sensormatic Inc. made a marketing decision to go
into the video speed-dome market. With some of the initial
concept coming from a company they had acquired, a large
dome system using slip-rings to allow continuous 360 rotation
and using a mirror—to reflect the incoming image onto the lens
and produce a lower profile dome system—was built. The sys-

tem also integrated the receiver driver portion of the PTZ
control electronics into the dome assembly. To improve
accuracy of pan and tilt and increase the speed sub-stantially,
stepper motors replaced the AC motors. This also made possible
the incorporation of dome pan/tilt pointing presets into the
system for defining targets, pat-terns and boundaries. The entire
dome was assembled and tested in 1985, and represented one of
the first fully integrated “speed” domes. The first system had a
clear viewing bubble 22 inches in diameter and was 8 inches
deep. The system used a monochrome vidicon tube cam-era and
weighed approximately 40 lbs. In 1988 a second-generation
speed-dome using a color CCD camera imager was produced.
The bubble was reduced to 12 inches in diameter and 5 inches
deep and weighed 26 lbs. To get the smallest size for the system
the CCD sensor and lens were located remotely from the camera
body using a high-flex cable. This produced a very short
camera-lens assembly. In 1992 the speed-dome received a
complete mechani-cal redesign and used a close-loop DC servo
electronic pan/tilt design providing the ability to point to any
target in less than one second. The first application for these
sys-tems was in casinos and interfaced with American Dynam-
ics matrix switchers.
A second pioneering company in the speed-dome field was
Diamond Electronics, producing a high-velocity rate-
proportional digital tracking system. It had a slip-ring design to
permit continuous 360 rotation at speeds up to 80 /sec and tilt
speeds up to 25 /sec. Dynamic brak-ing featured immediate
precise stops with ±0 5 accu-racy when de-accelerating from
any speed. The system had gold and chrome tinted dome capsule
enclosures providing one way mirror capsules providing for
discrete

surveillance. These dome capsules were optically corrected for
high-performance monochrome and color camera sys-tems. The
drive electronics and camera electronics were all contained
within the dome package.
The current and latest generation speed-domes are available in
sizes from 4.5 to 10 inches in diameter and have variabl e high-
speed pan/tilt stepper or servo motor drives with continuous 360
rotation obtained with metal or optical slip-rings. Camera-
pointing features include Pre-sets, Patterns, and Boundaries.
The cameras include high-resolution daytime and nighttime
capability using color for daytime and switching to monochrome
for higher sensitiv-ity during nighttime operation. Camera
features include VMD and alarm activation on motion. These
camera-lens modules are equipped with motorized zoom lenses
with optical and digital zoom, auto-focus, and iris control.
Zoom lenses have 20:1 zoom ratios to obtain telephoto and
wide-angle viewing. The camera pan/tilt module is mounted
within a rugged, clear, or smoked hemispheri-cal optical grade
acrylic plastic dome designed for quick installation, mounting,
and servicing in a ceiling, a wall, outdoor on a building parapet,
a parking lot pole, or on a highway.
14.3 FIXED DOME
The fixed dome camera assembly has become a very attractive
enclosure for providing surveillance in almost any environment.
The nature of the round dome with a smoked or tinted dome
makes it unobtrusive and does not allow the observer to
determine in which direction the camera is viewing. There are
many manufacturers pro-ducing fixed video domes with fixed
camera or manu-ally adjustable pan/tilt mounts. The cameras
provided are monochrome or color and the lenses have FOVs
from 90 wide-angle to 30 narrow-angle providing an

inexpensive, attractive integrated camera for most indoor
applications. Some models have variable focal length (vari-
focal) lenses to make it easier to obtain just the right camera
FOV. For outdoor applications, larger domes with larger and
longer focal length lenses are available to provide sufficient
mag-nification for the longer distances. These outdoor domes
are available with thermostatically controlled heaters and fans
and are sealed against moisture and the environment. All these
fixed dome cameras are available with infrared LED to provide
operation in nighttime at distance up to 20 feet without any
auxiliary lighting.
14.3.1 Technology
The fixed dome cameras use monochrome and color CCD or
CMOS cameras with lenses to view narrow-angle, medium- to
wide-angle FOVs under most lighting condi-tions. Typical
sensitivities are 1–2 lux for color cameras and
Dome Cameras
375
0.1 lux for monochrome. Resolution is typically 480 TV lines
for color and 570 TV lines for monochrome. When there is not
enough or no lighting an infrared LED cam-era is used. Many
dome manufacturers mount the camera so that it can be
manually adjusted in the horizontal and vertical (pan/tilt)
directions. These fixed domes are small and lightweight and are
easily mounted onto a drop ceil-ing, hard ceiling, or a wall.
Dome cameras are available with standard analog signal outputs
for use with coaxial cable, unshielded twisted pair (UTP) or to
interface with other transmission means. There are IP network
dome cameras that can be connected directly to a LAN, WAN,
or the Internet.

14.3.2 Housing
Most indoor fixed dome housings are manufactured using ABS
or polycarbonate plastic. The lower dome bubble through which
the camera lens views is manufactured from optically clear
acrylic plastic. Most systems are provided with a clear, tinted,
or smoked plastic bubble. Special variations include bronze,
chrome, and gold. The clear bubble essentially transmits all of
the light and is used when maximum light throughput is
required. The smoked dome loses about 30% of the light (70%
transmission), the bronze approximately 50% (50%
transmission) of the light, and the gold approximately 75% of
the light (25% transmission).
Outdoor housings are available with UV-protected ABS or
vinyl, or painted aluminum or steel. For harsh or extreme
environments or where corrosive atmospheres or severe
vandalism is present, dome housing materials are fabricated
from polycarbonate plastic, machined or cast aluminum, or
stainless steel.
14.3.3 Hardware
There are many manufacturers of fixed dome camera sys-tems.
Figure 14-1 shows examples of indoor and outdoor fixed dome
cameras. The size of these domes varies from 4 to 6 inches in
diameter and weigh from 1 to 2 lbs. They are available with
clear or smoked viewing domes.
14.4 SPEED DOME

The majority of conventional camera/lens pan/tilt plat-forms in
housings consist of components obtained from several different
manufacturers all assembled by the sys-tems integrator and
made to operate as a complete system. This is a practical
solution for installations in which the parameters and
characteristics of the fixed or movable dome might be
unacceptable. The dome camera inte-grates the camera/lens,
pan/tilt, housing, and mounting
FIGURE 14-1 Fixed dome camera
systems
(A) FIXED COLOR DAY/NIGHT (B) FIXED COLOR
INTERNET (IP)
ANALOG CAMERA MPEG/JPEG CAMERA
system in a single module from a single manufacturer. The
integral design results in a smaller, lighter weight module
having a high scanning speed and wide angular coverage.

Speed-dome systems can scan at a rate of 300 /sec and are
capable of panning 360 continuously using slip rings. With 360
continuous horizontal scanning the lens/camera module does not
have to come back 360 in order to follow a moving target. All
the components can be housed in a 5−7-inch diameter ceiling-
mounted dome. Through advanced engineering and compact
packaging, these fast scan rates were obtained: the moving parts
are small in size, and have low masses and moments of inertia.
The obvious advantage of a fast system is that if an incident
occurs anywhere within the dynamic FOV of the pan/tilt system,
the camera/lens can be pointed in any direction in the shortest
possible time while the lens zooms and focuses on the target.
Microprocessor-based dome systems with camera-pointing
preset capabilities can take advan-tage of these fast pan/tilt
designs.
14.4.1 Technology
The speed-dome assembly contains a high-speed pan/tilt
assembly, high resolution day (color) or night (monochrome)
CCD camera with a compact 20:1 zoom ratio lens with
continuous full-time auto-focus function. One system has a wide
dynamic range feature that can pro-vide detailed images when
the camera is viewing images that have bright light and low
light level image areas.
Camera, Lens. Most speed-dome systems use high sen-sitivity
color cameras that can be: (1) operated in color,
· operated in monochrome, or (3) switchable from color to
monochrome automatically. The CCD cameras have an image
format of 1/4 inch, and along with a compact zoom lens,
provide a small compact design resulting in high pan-tilt

speeds. Overall camera resolution is typically 480 TV lines for
color and 570 TV lines for monochrome.
Values of 1 lux sensitivity for color and 0.06 lux or less for
monochrome are typical. One system using a patented signal
level compression technique can provide images that have over
60 times the dynamic range compared to other cameras.
Cameras are also provided with automatic brightness
compensation (ABC) so that the camera can view scenes
containing both bright and dark areas. This overcomes the
problem that if a camera is located in a poorly illuminated room
and pointed at a window with a brightly illuminated scene
outside, the camera will either set its iris level to optimize the
inside or outside scene. This results in one part of display being
normal while the other part is either too light or too dark. This
also occurs in the evening when viewing oncoming traffic with
the headlights turned on. The ABC enables the camera to see
both the light and dark areas of the display with reduced flair
from the oncoming headlights.
The zoom lenses generally have a 20 to 1 optical zoom
(magnification) range that is extended by electronic digi-tal
zoom by another factor of 10 providing an overall 200 to 1
zoom range. Sensitivity of the color cameras are down to 1 lux
for color and to .05 lux for monochrome. Switchover from color
to monochrome is automatic when the light level falls below a
predetermined level. To capture image detail in both light and
dark regions, Panasonic Inc. uses the Super Dynamic SDII
technology which records the scene at two different exposures
and then electronically integrates both of them into a single
image to preserve the detail throughout the bright and dim
areas. This added to additional precise color reproduction
creates a dynamic range that is about 64 times greater than that

of conven-tional cameras.
Pan/Tilt Mechanism. The speed dome panning mecha-nism
provides 360 of continuous horizontal panning rota-tion. To
obtain the 360 rotation slip-rings are used. Some systems use a
light transmitter and receiver to transmit the signal information
rather than a metal slip ring assem-bly. The tilt mechanism
provides for at least a 90 vertical range of travel. In most cases
the camera assembly can tilt
up above the horizontal a few degrees and down −95 to provide
a tilt range beyond looking straight down to look-ing slightly
above the horizon. Precise manual panning and tilting is
achieved through a combination of a variable speed control in
the form of different speed ranges, with an automatic
adjustment of the speed range depending on the zoom position
of the lens. For wide-angle zoom-ing the speed is increased,
whereas for high magnification (telephoto zooming) the panning
and tilting speeds are decreased. Depending on the
manufacturer, the panningand tilting are done using DC servo
motors or stepper motors. To provide high torque and precise
pointing abil-ity, the DC servo design uses pulse-width
modulation and speed feedback to control the acceleration,
speed, and de-acceleration of the motors, ensuring a smooth,
precise, accurate, and fluid movement. Most manufactures
design the drive systems so that there are no belts or pulleys
insuring long-term reliable operation.
Dome Cameras
377
An example in which panning speed is important fol-lows: A

person walks past a dome pan/tilt unit 15 feet away the dome
(Figure 14-2). If the person is walking at a normal rate of about
4 feet per second and the dome is panning at a rate of 1 foot per
second (12 /sec), the monitor scene at 15 feet is moving at a
rate of 3 feet per second. The subject is quickly lost because the
pan/tilt cannot pan fast enough to follow the subject. With a
high-speed, 60 /sec panning system, a target at 15 feet from the
camera produces a picture going by at a rate of 5 feet per
second (1 foot per second faster than the target), and the subject
is not lost. In this example, the panning speed would be reduced
to 4 feet per second to keep the target in the center of the
picture.
Slip-Rings. Most standard pan/tilt platforms use a mech-anical
stop at each end of the horizontal and vertical pan-ning ranges
to prevent the wires connected to the moving
· LIMITED 355°
CONVENTIONAL PANNING

5 CONTINUOUS 360°
HIGH SPEED PANNING
360° PANNING MOTION
60°/sec MAX
15
PERSON WALKING AT
4 ft /sec
P/T PLATFORM MUST STOP
AND REVERSE DIRECTION
TO REACQUIRE MOVING TARGET

EVEN AT 24°/sec
IT REQUIRES 15 sec TO
ROTATE 360° AND
REACQUIRE TARGET
MAX PANNING SPEED 60°/sec
PRODUCES 5 ft /sec @ 15 ft
FROM THE CAMERA
15 ft
LENS
FOV
T = 4 sec
PERSON WALKING AT
4 ft/sec

FIGURE 14-2 Target speed vs. panning speed
camera/lens assembly from getting twisted (the wire ends are
terminated in the stationary wall mount). This means that the
camera cannot scan more than 355 horizontally before it must
stop and then pan in the opposite direc-tion. Even at a 24 /sec
pan speed, nearly 15 seconds is required to acquire a subject or
target that is moving past the end of the panning range. During
most of the 15 sec-onds the target is out of sight of the camera
and probably lost. The speed dome camera does not have this
limita-tion as it continues to follow the target. This is one of the
salient reasons why speed dome systems are such effective
surveillance cameras and have replaced many pan/tilt platforms.
In the panning system using slip-rings, the camera/lens
combination rotates continuously and beyond 360 with-out any
concern for twisted wires, since the electrical sig-nals and
power pass through the stationary slip-rings. No matter where
the target moves in the lens FOV, the pan-ning motion can
continue: the subject never leaves the FOV. There are no
restricting mechanical stops to limit the pan/tilt unit’s rotation.
Most dome manufacturers use gold plated metal slip rings to
transfer the video control and power signals from the camera to
the dome base and on to the communica-tion channel. Others
use optical slip rings for the video. The all-optical connection
between the moving camera and the base can provide a higher
quality image with less video noise than the metal gold
contacts. Transmission of the video signal by a light that
requires no physical con-tacts makes for a better “slip ring.”
This eliminates the possibility of image noise and enhances the
reliability of the dome unit.

14.4.2 Housing
Most indoor speed-dome housings are manufactured from ABS
or polycarbonate plastic. The lower dome bubble through which
the camera lens views is manufactured from optically clear
acrylic plastic. Most systems are pro-vided with a clear or
smoked plastic bubble. Other tints available include bronze,
chrome, and gold. The clear bubble essentially transmits all of
the light and is used for maximum light throughput. The smoked
dome loses about 30% of the light, the bronze approximately
50% of the light, and the chrome (aluminum) and gold approx-
imately 75% of the light. Only the clear and smoked versions
are generally used for outdoor applications.
If the camera/lens pointing axis is not perpendicular to the
dome surface (Figure 14-3) and looks at an oblique angle the
images may appear elongated vertically or hor-izontally. If the
dome and camera are in a fixed position with respect to one
another, the distortion is generally less noticeable than if the
lens is panning or tilting while the dome remains still. Figure
14-4 shows widely used dome housing configurations.
For outdoor applications, the domes are equipped with
thermostatically controlled heaters, blowers, and protec-tive sun
shrouds. Standard housing colors include gray, white, or black
baked on enamel. The lower domes through which the camera
views are clear or gray (smoked).
14.4.3 Hardware
There are many manufacturers producing high speed dome
camera systems. Table 14-1 shows some of the

CENTER
CAMERA
OF DOME
•
OFF CENTER
360°

0°
VERTICAL TILT
–90°
HORIZONTAL PANNING
FIGURE 14-3 Camera viewing through dome
Dome Cameras
379
FIGURE 14-4 Representative
speed-dome systems

11 480 TVL COLOR CAMERA
18× OPTICAL MAGNIFICATION MPEG/JPEG INTERNET
(IP)
10 510 TVL COLOR CAMERA
22× OPTICAL MAGNIFICATION
ANALOG OUTPUT

17 COMPACT SELF CONTAINED CCTV PLATFORM
18 INTEGRATED, ENVIRONMENTALLY CLOSED DOME
19 HIGH SPEED PANNING —200°/sec
20 HIGH POINTING ACCURACY: ±0.1°
21 360° CONTINUOUS PANNING
22 COMPACT, UNITIZED CAMERA/LENS/PAN/TILT
MODULE
23 HIGH ZOOMING (MAGNIFICATION) RANGE:
OPTICAL: 10–20×
DIGITAL: 10–20×
OVERALL: 10–200×
24 AUTO-REVERSE FOR DOWNWARD VIEWING
25 PRESETS: PAN, TILT, ZOOM, TOUR
26 MENU-DRIVEN-REMOTE SETUP
Table 14-1 Key Features of Speed-Dome Systems
features of speed-dome systems available. Section 14.7
describes many extra features not described in Table 14-1.
The two high-end designs by Pelco and Panasonic rep-resent the
most complex and full-featured systems. With technology

advancing regularly, these systems will contin-ually be updated
and supersede the capabilities of those listed in the table. Most
of these systems have many fea-tures in common but with
different specifications (see Section 14.7 for additional
features). Table 14-1 briefly outlines the key features of speed
domes. Most contain a color camera that is switched
electronically, or mechani-cally moves an optical filter in or out
of the image light path to the camera sensor. They contain a
high-speed pan/tilt servo or step motor drive system, and a
clear, smoked, or other tinted viewing material. The high-
quality dome material is of high-quality acrylic and is optically
clear with no distortion in any portion of the dome that is
viewed through by the camera/lens. These domes are available
for indoor wall mounting, ceiling mounting either recessed or as
a pendant on a building, or pole
mounted. They are available for outdoor applications for
parapet building mounting, on fixed poles in parking lots or
highways. The panning speed for most speed domes varies from
0.1 to 360 /sec continuous rotation. The verti-cal tilt ranges
from +2 above the horizon to −92 below the horizon. These
systems have manual override for speed control that ranges
from 0.1 to 80 /sec in panning to 0.1 to 40 /sec in tilting. In the
automatic preset mode, the panning speed can be up to 360 /sec
and the tilt speed up to 200 /sec. Most speed domes have
capability of pro-gramming presets including the ability to
select auto-focus modes, iris level, and light compensation.
Some systems have the ability to copy a preset command from
one cam-era to another. Programming can be via keyboard
through the dome system on-screen menu. Preset accuracy can
be as low as ±0 1 . These systems are provided with limit stops
that are programmable and used when the operator uses manual
panning. Most are provided with an opaque mechanical/optical
liner that rotates with the dome to ensure that the camera and

pan/tilt assemblies are not visible to the observer. The domes
are available with alarm inputs and outputs. Programmable
patterns can be user-defined including pan, tilt, and zoom for
the preset point-ing directions. The security manager can block
out spe-cific areas and specific viewing directions to eliminate
viewing secured areas and areas requiring privacy. The domes
are almost all available with a menu-driven setup and
operational modes. The menus can be displayed in different
languages for initial installation and operational use. Many have
an image flipping feature that inverts the dome image 180 at the
bottom of the tilt travel, so that the image is always right-side-
up when the camera view-ing angle passes through the vertical
downward rotation. Depending on the system, communication to
and con-trol from the monitoring console is performed through
multi-conductor cable, coaxial cable, UTP, fiber-optic, or
third-party control systems. Video motion detection is available
on most systems when in the preset mode of operation, with
alarm outputs activated. Most indoor and outdoor systems are
fabricated using painted aluminum construction with outdoor
systems available with stainless steel construction; either non-
pressurized or pressurized models are available depending on
the application.
Figure 14-4 illustrates some of the many standard types of
speed-dome systems available. Since most of these domes are
mounted at ceiling level, on a parapet atop a building, or on the
top of a pole, they must be designed for easy installation and
maintenance. Each has a unique quick-disconnect mechanical
install and removal inter-face for mounting the dome section to
the permanently mounted base section.
Pelco. Figure 14-4a shows a speed dome having a variable

panning speed from 360 /sec continuous down to 0 1 /sec. The
manual control range is from 0.1 to 80 /sec, and pan at 150 /sec
in what is called turbo mode. The tilt speed ranges from 0.1 to
40 /sec. When in the automatic preset mode, the panning speed
is up to 360 /sec and the tilting speed is up to 200 /sec. The
vertical unobstructed tilt is from +2 above the horizon to −92 .
Panasonic. Figure 14-4b shows a speed dome with a color CCD
camera having a 22 times zoom, auto-focus lens, and rotating
chassis in a 4.3-inch diameter housing suitable for most indoor
locations. It has an additional 10 times electronic digital zoom
for a total zoom range of 220. The color camera operates at
light levels of 1 lux and produces monochrome images at 0.06
lux. It has full 360 horizontal rotation and 90 vertical panning,
and has a speed of 300 /sec. It incorporates digital motion
detection for advanced alarm applications. The camera has 510
TV line horizontal resolution.
14.5 DOME MOUNTING HARDWARE
Many manufacturers produce attractive dome housings and
mounting configurations for indoor and outdoor fixed and
pan/tilt dome systems (Figure 14-5). For indoor applications,
the domes are securely attached to a wall or ceiling mounting
bracket. The electrical cables connected from the camera and
the pan/tilt mechanism are directed into the wall or ceiling.
14.5.1 Fixed Dome
The fixed dome module consists of the camera, lens, and

housing with dome and is installed on the surface of a wall,
ceiling, building exterior, and pole with appropriate mounting
hardware.
14.5.2 Moveable Speed Dome
The speed dome structure consists of two basic parts:
(2) the rear box which is installed or mounted on the mounting
surface (wall, ceiling, and pole) and (2) the dome with the
camera pan/tilt mechanism. Most manu-facturers use a quick,
positive, mechanical, and electrical disconnect between the rear
box and the camera/dome assembly that does not require the use
of any tools. This is particularly important in retail stores,
warehouses, parking lots, and highway applications since the
dome is usually mounted at elevations requiring ladders or other
means to reach the dome. This installation and maintenance
issue has been addressed by several companies that now pro-
duce dome systems that can be installed and maintained at
ground level (Section 14.8.2). The domes for these pole-
mounted systems are raised and lowered mechani-cally. The
dome is brought down to ground level during installation or
servicing and they are raised for operation at the elevated level
at the top of the pole. For these video domes the pole is part of
the dome system.
In harsh outdoor environments or for chemical protec-tion, type
316 stainless steel enclosures are available with a height of 11
inches including mounting and dome, and a 10 inch diameter.
These enclosures require no painting and withstand all outdoor
environmental conditions as well as having higher impact
ratings that are each important when the systems are located in
areas of vandalism or other attacks. Where required, pressurized
stainless steel pen-dants are available with an overall height of

12 inches and an 11 inches diameter. These domes have
Schrader type fill and pressure relief valves and operate at 5
lbs/square inch gage (psig) pressure typical and 7 psig pressure
relief. These systems usually incorporate internal sensors for
pressure, humidity, condensation, and temperature, and are
usually equipped with heaters or blowers where the environment
requires. These systems are equipped with internal sensors
reporting with on-screen displays of sensor indications and
sensor out-of-range reporting.
14.6 CABLING-VIDEO SIGNAL AND CONTROLS
The speed-domes communicate to the console and net-work via
built-in multi-protocol receiver/driver assem-blies for use with
matrix switching systems and other equipments. The types of
protocols supported by many manufacturers include: (1) AD
Manchester control code using a single 18 AWG shielded
twisted pair (STP) to sup-port several daisy chained domes at a
maximum of about 5000 feet, (2) 22 AWG UTP to support up to
32 daisy chained domes to a maximum of 3200 feet, (3) AD-
UTC and RG-59U video cable to control a dome to a maximum
of 1600 feet. These receiver drivers located in the dome provide
all the voltage necessary for camera controls, pan and tilt
functions, and all motorized lens functions. Most
Dome Cameras
381

(A) FLUSH CEILING MOUNT (B) PENDANT CEILING
MOUNT
(C) PENDANT WALL MOUNT (D) FLUSH
WALL/CEILING MOUNT

FIGURE 14-5 Indoor video dome mounting configurations
dome interfaces support selected third-party protocols for
integration into other systems. These can take the form of fiber-
optic communications or other types. The dome includes
standard support for UTP dome connections that allows the use
of CAT cabling for transmission of video or video up the coax
dome control signals up to 1000 feet. Communication protocols
provided by many manufactures include RS-422, RS-232, and
RS-485.
There are several techniques for the console controller to
communicate with and control the remote moveable speed-dome
camera:
· Direct Wire—video coax with multi-conductor for con-trols
· UTP—video with multiplexed controls
· Single Coaxial Cable—multiplexed video and controls on
coaxial
· Wireless—video and controls transmitted via RF or
microwave.
Direct Wire. The simplest control of the PTZ lens mech-anism
is via direct wire, using one wire for each control function and a
separate video coaxial cable. This straight-forward technique is
in widespread use for many small or short-run (under a thousand
feet) installations. This technique requires no additional driver
electronics for transmitting the control signals and no additional
receiver electronics at the camera unit. The controller consists

of switches that control all functions set manually by the oper -
ator or memorized by the system for automatic operation. Wire
size must be large enough to minimize voltage drop to the
motors and electronics.
Unshielded Twisted Pair (UTP). For longer distances or when
there are many different camera sites, a significant reduction in
the number of conductors and wire runs is accomplished by
multiplexing (time-sharing) the control signals at the control
console onto two UTP wires, sending them to the camera site,
and then de-multiplexing them or
separating them again to provide the signals necessary to drive
the PTZ unit. Since the two wires need to carry only
communications information and not current to drive the
motors, any long-distance two-wire communication system
suffices. Two popular transmission codes (protocols) are the
EIA RS-422 and RS-485. The video signal is transmitted on a
separate coaxial cable or UTP.
Single Coaxial Cable. Several companies manufacture systems
that multiplex or time-share the control signals in video signal
on the same video coaxial cable, thereby allowing video to be
transmitted from the camera to the monitor console site, and
camera control signals to be transmitted from the security
console to the camera site, all on one coaxial cable.
For direct wiring, this is an efficient solution since only a singl e
coaxial cable is required. The system requires a simple
multiplexer that combines the video and control signals at the
camera and the monitor ends. An advantage of multiplexing the
control signals onto the video signal is that additional
transmission or control signals can be added to the system
without adding new cable. These addi-tional functions can

include lens controls, alarm functions, or tamper switches.
Wireless. Control signals can be transmitted from the console to
the camera location via wireless remote con-trol
communication. The control signals are multiplexed onto a
single channel and transmitted on RF, microwave, or light-wave
(visible or infrared) communication links. In extreme security
environments (such as military or nuclear sites), wireless
transmission of video, command, and con-trol signals is used as
a backup to a hard-wired (copper or fiber-optic) system.
Fiber-optic. The fixed and speed-dome systems have
compatibility with fiber-optic transmitters used for long-
distance cabling runs. Fiber-optic transmission is an alter-native
to copper wire, and many manufacturers have equipment that
transmits the control signals, alarms, and video signal on a
single fiber-optic channel. As mentioned in Chapter 6, the fiber-
optic advantages include noise immunity, long transmission
distance, absence of ground loops, high security (difficult to
tap), and reliable opera-tion from different building sites in
harsh environments.
Third-party Communicators. The fixed and speed-dome systems
have compatibility with and capability to be con-nected into
optional boards that convert the control signals into a suitable
form for the selected third-party controllers.
Digital Network. Fixed and speed-dome camera systems are now
available that can be connected directly into ana-log or digital
networks. When the camera is connected to a LAN, WAN, or
Ethernet network, the operator can view and operate the system
and monitor the images locally or remotely using a PC.
Wiring Access Panel. The installation of the dome base is

normally accomplished prior to the purchase or instal -lation of
the dome housing itself. The dome base should have an easy
access door that allows complete access to the installation
wiring, and when closed it should provide complete separation
of this wiring from the dome drive.
14.7 SPECIAL FEATURES
Camera Sensitivity. Most dome systems have dual-mode
cameras that operate in color mode during daytime and
monochrome mode during nighttime. In addition, some cameras
have the feature to provide temporary image enhancement under
low light level conditions via manual override. This override
reduces the shutter speed from the normal 30 fps to 2 fps
resulting in a 15 times increase in camera sensitivity.
Memory. Non-volatile memory storage and location-specific
dome settings such as presets and patterns are built-in for the
camera. If a new dome drive is installed in the system, all the
settings are downloaded automatically into the new dome drive.
Motion Detection. Domes support VMD within a preset. The
motion detection trigger action includes activating a preset
command, activating a pattern, and sending a dome output to the
console.
Presets and Patterns. Most domes support camera pre-sets
programmed into the dome module so that the dome can point
(pan/tilt) to a preset direction. Models with as many as 96
presets and 60 patterns of presets are pro-grammable. Domes
are also designed to support a Home Position that automatically
returns the dome to a Preset, Pattern, or Preset Sequence after a
specified period of inactivity anywhere between 1 minute and 1
hour. Also provided is a freeze frame function that maintains a
static image on-screen during dome movement and lens adjust-

ment when presets and patterns are called. This freeze frame
function helps preserve hard drive space when a VCR or DVR is
used.
The speed-dome parameters that can be preset include:
· auto-focus mode, (2) iris level, (3) back-light compen-sation,
(4) the ability to command to copy the camera settings from one
preset to another to reduce setup time, and (5) to preset
programming the control keyboard or the dome system on-
screen menu. The preset accuracy can be as high as ±0 1 .
Proportional Pan/Tilt Speed. The system panning and tilting
speed can be increased or decreased depending on the
instantaneous zoom focal length. To optimize the viewing of the
image on the monitor for different zoom positions, when the
zoom lens is in wide-angle position the speed is increased, and
when it is in the telephoto (high magnification) position it is
decreased, and proportionally optimized in between.
Digital Flip. The speed-dome should have a provision for quick
image reversal that automatically pans the camera
· when the bottom −90 tilt limit is reached to allow for
continuous tracking of a target passing directly beneath the
dome. This is important when following a person who is passing
directly under the camera from one side to the other.
The digital flip feature allows for more convenient mon-itoring
when viewing objects that pass directly below the camera. As
the camera pans in the vertical direction to follow the object,
DSP automatically flips the image to the bottom as the object
passes beneath the camera so that the image remains right-side-
up for easier viewing. In addition, the system contains an
image-hold feature that prevents blurring when the camera

moves and does the 180 flip. It maintains the image prior to flip
after the 180 flip.
Privacy Zone-Window Blanking. Some domes support privacy
zones to prevent users from viewing sensitive or secured areas.
So as not to interfere with normal surveil-lance operations,
these on-screen shields must block out only the area that has
been defined as sensitive. The pri-vacy cell should not cause the
screen to blank out.
These privacy windows are available in: (1) four-sided user-
defined shapes, (2) opaque gray or translucent smear,
· blank all video above a user-defined tilt angle, and
· blank all video below a user-defined tilt angle.
Zoom-Distance Compensation. Whether the dome cam-era is in
the privacy zone or the lens is zooming from wide-angle to
telephoto the system should compensate for a specific focal
length in use at the time. For any specific focal length, the
zoom lens should adjust the alarm or pri-vacy zone window to
compensate for the changing FOV. This is called zoom-distance
compensation.
Monitor Display, Menu. The speed-dome systems sup-port on-
screen programming of the dome parameters including image
flip, direction indicators and azimuth, maximum zoom stop,
camera line lock or internal crys-tal synchronization, AGC,
white balance, VMD selection, alarm actions and default states,
and home position. They also display on-screen programming
of: dome names, area names, preset names, pattern names, and
alarm names. Most systems provide most of these attributes in
English, French, Italian, German, and Spanish, as well as in
other languages. The on-screen text characters are available as
user-selectable in solid or translucent white, with or with-out a
black outline.

Alarm Inputs. The dome assemblies have single or multi-ple
alarm inputs as an option and are field programmable to receive
normally open or normally close contacts. If the system is
operating on an RS-422 network, the domes are capable of
receiving the alarm and transmitting it back to the switching
system, and/or reacting to the alarm event independent of the
switching system. If a Manchester net-work is used, the dome is
capable of processing the alarm
Dome Cameras
383
internally in the dome and automatically activating a Pre-set,
Pattern, or Preset Sequence.
Twist Lock Release. Maintenance is an important factor to
consider in ceiling or pole-mounted dome camera sys-tems. To
simplify installing and servicing these domes, most systems
contain a quick disconnect or twist lock release at the base of
the dome. The standard base of the dome is hard mounted to the
wall, ceiling, or pole mount and contains a receptacle for direct
wiring to the dome assembly. All wiring is done before lifting
the camera pan/tilt assembly onto place. The base assembly
includes a tamper switch so that if the dome cover is removed,
an alarm is sounded. The quick disconnect base allows wiring to
be done once directly in place and then installing or servicing
the dome assembly without disturbing any of the wires or
connections. Normally each base includes diagnostic LEDs to
indicate power and proper commu-nications to and from the
console or matrix switcher. Some designs require a simple tool
to remove the dome assembly; however, others require no tools
and are simply installed or removed using a twist lock release.
It is impor-tant that the dome and the base are available
separately so that the installation of the base can be

accomplished by the installer prior to the purchase of the dome
hous-ing/camera assembly.
14.8 SPECIAL APPLICATIONS
The use of fixed and speed domes in elevated locations in
buildings, on exterior walls of buildings, and outdoors, in
general, has resulted in the design of many different
configurations for mounting these domes.
14.8.1 Outdoor Building Mounts
Figure 14-6 illustrates outdoor speed-domes mounted on a
building roof edge and capable of scanning 270 and
· horizontally to view parking garages and lots. With such a
large angular FOV to cover (an entire parking lot), this solution
should be used where only sporadic activity is monitored, since
panning with a standard unit from one end of a building would
not keep most of the parking lot under surveillance. Adding
additional speed domes would increase coverage.
14.8.2 Pole Mounts
Figure 14-7 shows dome camera pan/tilt assemblies mounted on
poles and pedestals to provide wide-angle video surveillance at
entry and exit roadways, parking lots, streets, etc. Mounting the
camera away from the building on a pole provides good viewing
of the entire

FIGURE 14-6 Standard outdoor speed-dome mounting
configurations
(A) STANDARD (B) BUILDING MOUNT

(A) PEDESTAL/WALL MOUNT DOME (B) CORNER WALL
BRACKET (C) POLE/WALL MOUNT
FIGURE 14-7 Outdoor dome and mounts for buildings,
roadways and parking lots
building entry area with a single camera. The presence of the
camera system serves as a deterrent to crime while it captures
the necessary visual information for possible apprehension and
prosecution. The same scan-ning limitations as described in the
previous system apply.
There is one disadvantage of the dome pole camera or any
camera mounted on a pole: the difficulty of performing
maintenance on it. Several companies have pursued designs that
permit easier maintenance. The widespread use of the speed-
dome in parking lots, on walkways, and on streets,
highways, etc. has motivated manufacturers to design inge-
nious means to raise and lower the entire dome assembly from
ground-level (Figure 14-8). The video dome in Figure14-8a, b is

raised and lowered using an electric drill.
14.9 SUMMARY
There are many varieties of camera housings and inte-grated
camera systems for video surveillance applica-tions. The
configuration that has become most popular
Dome Cameras
385

(C) HEAVY DUTY
(D) HEAVY DUTY
(A) RAISING AND LOWERING (B) NORMAL POSITION
FIGURE 14-8 Pole-mounted dome assemblies maintained from
ground-level
is the dome housing that is available in a fixed or speed-dome
configuration. These dome camera systems are suitable for
indoor and outdoor applications avail-able with monochrome
cameras, or color cameras that can automatically switch from

color for daylight use to monochrome for extended low light
level sensitivity and
produce optimum performance at most all light levels. The
speed dome provides a very powerful video surveil-lance tool
for gathering maximum visual intelligence and is in widespread
use in retail establishments, casinos, ware-houses, outdoor
parking lots, pathways, building exteriors, and streets and
highways.
Chapter 15
Integrated Cameras, Camera Housings, and Accessories
CONTENTS

15.1 Overview
15.2 Indoor Housings
15.2.1 Functional Requirements
15.2.2 Indoor Types
15.3 Outdoor Housings
15.3.2 Outdoor Design Materials
15.3.3 Outdoor Types
15.4 Integrated Cameras
15.4.1 Indoor
15.4.2 Outdoor
15.5 Specialty Housings
15.5.1 High Security
15.5.2 Elevator
15.5.3 Dust-Proof and Explosion-Proof
15.5.4 Pressurized and Air- or Water-Cooled
15.6 NEMA Environmental Ratings
15.7 Housing Accessories

15.8 Housing Guidelines
15.9 Summary
15.1 OVERVIEW
There are many varieties of video camera housings and
integrated cameras available for indoor and outdoor secu-rity
applications. Standard shapes and forms they take include: (1)
rectangular—mounted on a wall or ceiling,
· dome—mounted on a ceiling, wall, pole, and pylon,
· triangular—mounted in a corner, and (4) wedge— ceiling
mounted. The two primary functions for these environmental
housing are to protect the camera and lens from vandalism and
the environment. To meet these

requirements, indoor and outdoor housings and inte-grated
camera modules are fabricated from a variety of materials
including aluminum, painted steel, stainless steel, and molded
high-impact plastic.
There has been an increasing demand for aesthetically designed
housings and cameras to match the decor of a building interior
or exterior. While the primary function of the housing is to
protect the camera, lens, and electri-cal wiring, these aesthetic
camera housings are especially attractive and unobtrusive as
dictated by architectural con-siderations. To satisfy these
requirements, manufacturers have produced attractive designs
using injection-molded plastic and other materials and forming
techniques.
Housings are used to protect vital electronic video equipment;
consequently, the material used for their construction must be
chosen carefully. Underwriters Laboratories (UL) has developed
guidelines for minimum fire-safety requirements and suggested
tests and ratings for fireproof or fire-retardant designs. This is
especially important for non-metallic designs. The Electronic
Industries Association (EIA) has guidelines for improved
interchangeability among manufacturers’ products. The National
Electrical Manufacturers Association (NEMA) has detailed
specifications describing the requirements for indoor and
outdoor housing requirements of elec-trical equipment. These
guidelines and ratings relate to materials and finishes,
mechanical design parameters such as mounting-hole locations,
and electrical-cable entry and fittings.
This chapter describes rectangular, triangular, dome, and all the
other special indoor and outdoor housings, including
accessories such as heaters, fans, thermostats, and windshield
wipers and washers. Most housings have locks or tamperproof
hardware to prevent vandalism or theft of the camera and lens.

387
Camera Housings. The indoor round hemispherical dome-shaped
housing has become very popular because it is attractive and
has excellent functionality. The dome’s symmetrical shape and
tinted viewing “window” prevents the observer from seeing the
direction in which the cam-era is pointing. This adds a
deterrence factor to the surveil-lance function. Many security
installations require discreet video surveillance equipment that
blends in with the surrounding environment, not eye-catching or
obtrusive housings. Corner-mounted triangular and wedge-
shaped housings are also in widespread use.
Outdoor housings used on facility properties are designed to
match landscaping and grounds and/or spe-cific lighting
conditions. Outdoor environmental housings that are subject to
wind loading or ice buildup should be no larger or heavier than
required to house the camera, lens, and associated wiring and
accessories. They should be constructed to withstand the harsh
outdoor environment and added abuse from vandalism or attack.
The camera housing enclosures should have easy access into
them via a hinged or sliding interior assembly or removable
cover.
The housing, camera, and lens are often within reach of
personnel who could damage or remove the equipment. Of
particular concern are high-risk locations such as jail cells,
building exteriors, and public-access locations that require a
more rugged housing fabricated from stainless steel or high-
impact polycarbonate plastic. Figure 15-1 shows two examples
of standard indoor and outdoor cam-era housings.

Integrated Cameras. With the increased use of video
surveillance cameras, manufacturers, video integrators, and end-
users have sought to simplify the purchasing and installation of
camera systems. To that end the integrated camera has become
very popular and an efficient means to accomplish that function.
The integrated camera is a plugand play surveillance camera
including the camera, lens,and any internal wiring associated
with it, and mounted in a small housing that is ready to install
at the site in a minimum amount of time. These integrated
cameras take on shapes similar to some of the housings
described in
the previous section but are smaller and more compact. Very
popular types are domes, corner mount, wedge, with
environmentally enclosed day/night camera with integral
bracket mounting. Figure 15-2 shows examples of these
integrated cameras.
15.2 INDOOR HOUSINGS
Indoor housings must protect the camera and lens from
pollutants such as dust and other particulate matter, a cor -rosive
atmosphere, and tampering or vandalism. Indoor housings are
constructed of painted or anodized alu-minum, painted steel,
stainless steel, and several types of plastic. The material for
plastic housings must be flame-proof or flame-retardant, as
designated by local codes and UL recommendations. The
housings must have sufficient strength to protect the lens and
camera, and be sturdily mounted onto a fixed wall or ceiling

mount, or recessed in a wall or ceiling. The lens should view
through a clear window made of safety glass or plastic.
Recommended plastic window material is either high impact
acrylic or polycarbonate with a mar-resistant finish. The
electrical input/output access locations should be designed and
positioned for easy maintenance. For easy access and servicing
of internal parts, the top half of the housing should be hinged or
be able to slide open, or be remov-able. In some designs, the
entire camera/lens assembly is removable for servicing. Figure
15-3 shows the interior of a typical rectangular indoor housing.
The common rectangular housing is available in many sizes and
is the least expensive. For vandalism protection, many housings
are available with key locks or tamperproof hardware that
allows the cover to be removed only with a special tool. In very
high risk areas, welded stainless-steel housings with thick
polycarbonate windows (3/8 or 1/2 inch) and high-security locks
are used. Some housings
(A) INDOOR (B) OUTDOOR

FIGURE 15-1 Standard indoor and outdoor camera housings
389
(A) STAINLESS STEEL CORNER MOUNT (B) CEILING
MOUNT-FIXED DOME

(C) SPRINKLER CEILING MOUNT (D) HARDENED
WALL/CEILING
FIGURE 15-2 Popular integrated cameras
are designed to provide concealment and improved aes-thetics
by recessing them into the wall or ceiling. The five housing
types that account for most security installations are: (1)
rectangular, (2) dome, (3) wedge, (4) triangular, and (5) wall -
and ceiling-recessed and surface-mounted.
15.2.2 Indoor Types
Rectangular. The most popular type of housing is the standard
rectangular design since it can be fabricated at low cost, is
sturdy, and is available from many manufactur-ers in many sizes
and attractive styles.

Under normal circumstances, indoor housings do not require
any special corrosion-resistant finishes. The housings are made
from painted or anodized alu-minum, painted steel, or high-
impact plastic, such as polyvinyl chloride (PVC), acritile
buterated styrene (ABS), or polycarbonate (General Electric
Lexan, etc.). In high crime areas and jails, stainless steel
housings are used.
Accessibility to the camera/lens assembly for installation and
servicing is important. Video surveillance cameras are always
mounted near or at ceiling height, on a pedestal, or at some
elevated location requiring service personnel to be on ladders or
power lifts. The housing design must permit

FIGURE 15-3 Indoor housing showing interior
easy access and serviceability under these conditions. Man-
ufacturers provide one of several means to gain access to the
housing: (1) removable top cover, (2) hinged top cover, (3) top
cover or camera/lens on slide, (4) remov-able front and/or rear
cover, (5) hinged bottom cover (dome), or (6) top cover on slide
(Figure 15-4).
Dome. A second category of indoor housing is of a round or
hemispherical, clear or tinted dome in which a cam-era, lens,
and an optional pan/tilt mechanism are housed. Chapter 14
described dome cameras in detail. The ceiling-mounted
hemispherical dome and the below-the-ceiling and wall-
mounted domes on brackets look totally differ-ent from the
rectangular housing, and often blend in better with architectural
decor. Since they look like a lighting fixture, they are less
obtrusive than rectangular housings. Since the hemispherical
dome is circularly sym-metrical, it can be in a fixed position
and the CCTV cam-era pointed in any direction to view the
scene. A pan/tilt unit used in a dome can rotate and tilt the
camera and lens while still remaining inside the confines of the
dome. This is in contrast to cameras inside rectangular and
other housings: if the camera moves, the entire housing assem-
bly has to move as a unit.
If the dome is tinted so that the person down at floor level

viewing the dome cannot see the camera and lens, it is possible
to point the camera in any direction without the observer seeing
it move. This capability can act as an additional security
deterrent because the observer does not know when he or she is
under surveillance.
There are three different types of plastic dome materials
through which the lens views the scene: (1) clear, (2) semi-
transparent aluminum- or chrome-coated, and (3) tinted or
smoked plastic. When the dome housing is used for
REMOVABLE
TOP COVER
MAIN
MAIN
HOUSING
HOUSING
SLIDE

(1) REMOVABLE TOP COVER
(2) HINGED TOP COVER
(3) CAMERA/LENS SLIDE
REAR
COVER
MAIN
CEILIING LEVEL
SLIDE
HOUSING
HINGED

FRONT DOME
COVER
(4) REMOVABLE FRONT/REAR COVER (5) HINGED
BOTTOM COVER (6) TOP COVER ON SLIDE
FIGURE 15-4 Camera housing access methods
protection only and its pointing direction need not be con-
cealed, the clear plastic dome is the best choice, since it
produces only a small 10 or 15% light loss. If the camera’s
pointing direction is to be concealed for additional secu-rity a
coated or tinted dome is required. The aluminized dome is the
earliest version of the coated dome and atten-uates the light
passing through it by approximately two f-stops (equivalent to
approximately a 75% light reduction or loss). While this type of
dome is still in use, the pre-ferred dome material is a smoked
plastic or tinted plastic that attenuates the light approximately
one f-stop, or 50%.
In contrast to rectangular housings using flat plas-tic or glass
windows with excellent optical quality and transmission, some
dome systems add slight optical dis-tortion to the video picture.
In high-quality domes the image distortion is almost negligible,
but in some systems the distortion or loss in resolution is
noticeable. In any dome-housing application the camera/lens
should view through the surface of the dome perpendicularly as
shown in Figure 15-5a.
Under this condition, there is at least symmetry of dis-tortion,
that is, the primary effect is that of a weak lens producing a
small change in the focal length of the total

391
lensing system and is usually not noticeable. If the cam-era/lens
pointing axis is not perpendicular to the dome surface (Figure
15-5b) and looks at an oblique angle through the dome housing
material, noticeable distortion will occur; for example, images
may appear elongated ver-tically or horizontally. If the dome
and camera are in a fixed position with respect to one another,
the distortion is generally less noticeable than if the lens is
scanning or tilting while the dome remains still. Figure 15-6
shows four widely used dome housing configurations.
Wedge Housing. One version of the wedge housing is designed
to replace an existing standard 2 feet × 2 feet drop ceiling tile
(Figure 15-7a) and another version (Figure 15-7b) is designed
for surface mounting. The wedge housing in Figure 15-7a is a
manually rotatable 16-inch high impact white plastic center
section with a wedge-shaped camera protruding about 5 inches
below the ceiling line. There are no additional accessories
required. The design allows for manual pan adjustments of 360
and minor tilt adjustments. After final pointing the center
camera/lens section is restricted from rotating by tight-ening
thumbscrews. The camera’s wedge shape aims the camera about
15 down from the horizontal. The front of
(A ) LENS AXIS PERPENDICULAR TO DOME SURFACE:
EXCELLENT IMAGE

DOME
SURFACE
· LENS AXIS NOT PERPENDICULAR TO DOME SURFACE:
POOR IMAGE
DOME
SURFACE
LENS AXIS
OBLIQUE
ANGLE
LENS AXIS

FIGURE 15-5 Indoor ceiling mounted dome camera with lens
axis perpendicular to dome surface
(A) INDOOR-N CEILING MOUNT (B) OUTDOOR-
BUILDING/POLE MOUNT

(C) INDOOR-SURFACE MOUNT (D) OUTDOOR-SURFACE
MOUNT
FIGURE 15-6 Dome housing configurations
the protrusion has a viewing window of clear acrylic with no
distortion and virtually no light transmission loss.
Another version is a small surface-mounted wedge-shaped
housing that can be attached to any ceiling. These are available
in either a surface- or recessed-mounting configuration.
Corner Mount. Figure 15-8 illustrates examples of aes-thetic
and hardened camera/lens housings designed specifically for
corner mounting in rooms, elevators, stair-wells, jail cells, etc.
Figure 15-8a shows a high-security housing of welded stainless
steel with a polycarbonate win-dow. The tamperproof corner
mount camera housing has a camera bracket assembly
permitting the camera to be tilted vertically ±10 for minor
adjustments of the vertical pointing angle. The lens viewing
window permits viewing a 95 horizontal FOV and 75 vertical
FOV. The optimum pointing direction for the lens and camera is
45 with respect to both adjacent walls and 45 down from the
ceiling horizontal plane. For an elevator cab application this
housing with a wide-angle, 95 horizontal FOV can view entire
elevator cab with no hidden areas and provide 100% video
coverage of the cab area. The high-security housing has a
hinged, lockable cover for easy, controlled access to all internal

parts, and a tough mar-resistant poly-
carbonate (Lexan) window. All mounting, video, and elec-trical
power access holes are located on the rear and top surfaces and
inaccessible to the public. The installation meets codes that
require unbroken firewalls. Three differ-ent housing sizes of
this design accommodate most CCD solid-state cameras and
wide-angle manual- or automatic-iris lenses or variable focus
(vari-focal) lenses. Since the housing is exposed to the public, it
is securely locked and manufactured using tamperproof
materials, such as steel or stainless steel, and a polycarbonate
(Lexan) window.
Figure 15-8b shows a housing fabricated from high impact
plastic and is a configuration suitable for applica-tion requiring
moderate security. The plastic housing has a lockable front
cover and all mounting and electrical access holes are out of
sight, and not accessible to the public. The housing has an
adjustable bracket for tilting the camera vertically. There are
many manufacturers supplying these types of corner mount
housings in materials ranging from stainless steel, steel, and
plastic. Finishes include brushed stainless steel and painted
aluminum, steel, and plastic.
Figure 15-8c shows a mirror-view corner mount hous-ing that
has a tinted or aluminized one-way window. It is
· small 7 inch × 7 inch × 7 inch unobtrusive housing that
renders the camera and lens covert.
393

(A) ROTATABLE: 2' × 2' PANEL (B) FIXED HOUSING
(C) TYPICAL ABOVE CEILING HOUSING (D) COMPACT:

7" LONG
FIGURE 15-7 Wedge camera housings
Ceiling- or Wall-Recessed or Surface Mount. Recessed or
partially concealed housings are often mounted in ceil-ings and
walls. Figure 15-9 shows examples of these hous-ings, including
the wedge and dome-shaped types. The round, semicircular, and
tapered housings shown offer design flexibility since the camera
and lens can be pointed in any horizontal direction while the
square or rectan-gular ceiling tile remains in place. These
housings are used where a low-profile (but not covert) type of
surveil-lance camera is required. These cameras are well suited
for looking down hallways, at cash registers, etc. In ceiling
installations, most of the housing, camera, and lens are mounted
above the ceiling level. The only portion below ceiling level is
a small part of the housing and the window through which the
camera lens views. The cameras and lenses are accessible from
below ceiling level by unlocking a cover that swings down, or
by gaining access from the rear of the housing above the ceiling
from an adjacent ceiling tile. It is important that all ceiling tile
mount hous-ings be securely attached to a structural member of
the
building above the ceiling with a chain or cable so that if the
hanging ceiling support fails, the housing and con-tents do not
fall to the floor or possibly injure personnel below.
With the increased use of video surveillance in pub-lic
locations, be they government, industrial, or private, more
attention is being given to the decorative and aes-thetic features

of the housing. These housings often have finishes of brass,
gold, or chrome, with satin or polished finishes. They are also
available with custom paint colors and textures, and custom-
colored plastics. Several manu-facturers offer special shapes
and custom configurations for matching specific architectural
designs.
15.3 OUTDOOR HOUSINGS
Like the indoor housing, the outdoor housing protects the
camera and lens from vandalism and adverse out-door
environments. Most outdoor housings are provided

(A) DISCRETE TRIANGULAR ONE-WAY MIRROR (B)
DISCRETE CONVEX TINTED MIRROR
(C) STAINLESS STEEL (D) HIGH IMPACT PLASTIC
FIGURE 15-8 Corner mount housings
with key locks to prevent unauthorized opening of the housing.
Outdoor housings must protect the camera from vandal-ism as
well as adverse environmental conditions. The van-dalism
encountered can range from rocks or sticks thrown at the

housing to bullets and other explosives. These secu-rity
housings are prime targets since they are mounted on ceilings,
walls, building exteriors, and poles and pedestals.
In outdoor installations the camera is mounted in a protective
enclosure to protect it against environmental factors such as
precipitation: rain, hail, snow, sleet, ice, and condensing
humidity. The outdoor housing must also protect against many
types of particulate matter including
dirt and dust, sand, fly ash, soot, and any other material local to
a particular site. Outdoor locations with a cor-rosive atmosphere
can cause rapid deterioration, failure, and premature
replacement of the camera and lens if not properly protected.
These substances include industrial chemicals, acids, and salt
spray. Outdoor housings should have external finishes that
withstand the atmosphere in which they are to operate. In hot
climates, a sun shield or shroud and a bright aluminum or white
finish is desir-able to reflect sunlight and eliminate heat buildup
in the housing.
Outdoor housings share many of the same require-ments as
indoor housings. Accessibility to the camera and lens during
installation and maintenance are more impor-tant in outdoor
applications since video equipment is often mounted high above
the ground and serviced under adverse conditions.
395

(A) CONCEALED CEILING (B) WEDGE
(C) DOME
FIGURE 15-9 Recessed and concealed ceiling, wedge, and
dome housings

15.3.2 Outdoor Design Materials
Outdoor housings are manufactured from aluminum, painted
steel, stainless steel, and outdoor-rated plas-tic, including
polycarbonate, ABS with a UV protective layer. It is important
that plastic outdoor housings be fabricated from UV-inhibiting
materials, to prevent the housing from deteriorating due to
sunlight. Plastics not treated will crack, and colors will fade.
High-quality baked-enamel, painted-steel, and stainless-steel
housings will
15.3.3 Outdoor Types
The outdoor camera housings are similar to the indoor except
that they must be furnished with an exterior finish that can
resist and withstand the outdoor environment. They should be
fitted with a thermostatically controlled heater and fan so that
when the temperature extends beyond the range of the camera
and lens specifications they can either be heated or cooled.
last many years. Where long-lasting, high-security, vandal-
Rectangular. For outdoor applications the rectangular
proof housings are required, stainless steel is the choice
plastic, painted aluminum, or stainless steel housings
since it does not rust or corrode and is extremely tough.
are the most popular choices. These housings are eas-
Aluminum is a good choice for an outdoor hous-
ily mounted from a bracket on a building, wall, or pole,
ing when anodized and finished in baked polyurethane
or hung from a building overhang to provide a solid
enamel paint and anodized. Anodized and painted alu-
mounting.

minum is the most durable finish. Aluminum and steel
housings should not be used when a salt or other cor-
Dome. Dome housings can be mounted on an individual
rosive atmosphere is expected. Stainless steel and special
pole or pylon, under the eaves of a building, or on a
plastics are the best choice for a salt-spray environment.
bracket mounted off the wall of a building. These housings
Consult the housing or materials manufacturer for the
must also use outdoor materials that will withstand the
proper choice.
environment.
15.4 INTEGRATED CAMERAS
The integration of the video camera, lens, housing, and mount
into one unit has been a natural evolution in the security
industry. This evolution has occurred as a result of the
availability of small CCD and CMOS cameras and asso-ciated
small lenses. It has made technologic and economic sense for
manufacturers to integrate these components into a single
finished product available to the video sys-tems integrator or
end-user as a plug and play video surveil-lance module ready
for mounting on a wall, a ceiling, outside a building, etc. These
integrated cameras have taken the form of domes (see Chapter
14), triangular-corner, wedge, and covert. There are many
manufacturers producing hundreds of models for indoor and
outdoor applications. They are available in monochrome and
color for daylight and nighttime use.
15.4.1 Indoor

Indoor integrated cameras have housings that take the form of
those described in Section 15.2. The housing types
include the dome, triangular-corner, wedge, and semi-covert
models (Figure 15-10).
Dome. The integrated dome camera uses a dome hous-ing with a
camera and lens installed. Most dome applica-tions now use the
integrated dome camera instead of the component form because
of the ease of installing a com-plete plug and play module and
the concomitant lower overall cost. These modules are available
for monochrome and color use as well as total darkness using
infrared LED illumination. Figure 15-11 illustrates an
integrated dome camera and its interior assembly in an
electrical duplex outlet box showing the manually adjustable
and tilt bracket for the camera lens assembly.
Triangular-Corner. A triangular-shaped integrated cam-era
housing using a one-way mirror installed in the corner of a
room at the ceiling level provides an excellent semi-covert
surveillance camera. Typical locations are in a small room or
lobby, an elevator or a stairwell. Figure 15-12 shows this design
using a wide-angle (90 FOV) lens that can view the entire area
of a small room or other space.

(A) DAY/NIGHT RUGGEDIZED DOME (B) SPRINKLER
HEAD
(C) CORNER MOUNT MIRROR (D) RUGGEDIZED WALL
MOUNT
WITH LED IR ILLUMINATION
FIGURE 15-10 Indoor integrated cameras

397
FIGURE 15-11 Integrated dome camera assembly FIGURE 15-
13 Wedge integrated camera
15.4.2 Outdoor
FIGURE 15-12 Discrete triangular corner mount mirror
integrated camera
The camera installed in the triangular housing is at a 45 angle

pointing down from the ceiling to view the entire area. The
triangular housing can be mounted in protected outdoor
locations at entrances or exits to buildings, etc. where two walls
meet, resulting in a very unobtrusive instal-lation. When
mounted in hot or cold environments, the housings must be
provided with a thermostatically con-trolled heater or fan.
Wedge. The wedge-integrated camera is available as a small,
unobtrusive assembly suitable for mounting directly to a hard
ceiling or on a ceiling tile. These cameras are lightweight and
generally require no additional support structure—they can be
mounted directly onto the ceiling tile (Figure 15-13).
Covert. There are many variations of integrated covert-type
video surveillance cameras used to augment overt cameras.
These can take the form of a sprinkler head, smoke detector,
passive infrared detector, temperature thermostat, etc. (see
Chapter 18 for many versions of covert integrated cameras).
Most integrated camera units for outdoor applications take the
form of a dome camera assembly in a plug and play form for
maximum ease of installation and servicing. Some other forms
used include ruggedized camera hous-ings with the camera,
lens, heater, and fan, all enclosed in the housing, ready for
mounting on an exterior bracket pole or pedestal. Dome
assemblies such as those shown in Chapter 14 for outdoor
applications are representative of these types.
15.5 SPECIALTY HOUSINGS
There are security applications in which cameras must be
located in very hostile environments. To protect the camera and
lens from damage and downtime, manufac-turers offer housings

that can withstand high mechanical impact from hand-thrown or
fired projectiles, extreme high temperature, dust, sand, liquid,
corrosive chemicals, and explosive gas. The following housings
have unique characteristics for solving these extreme security or
special environmental applications.
15.5.1 High Security
There are numerous armored camera/lens enclosures for
installation in correctional institutions. Figure 15-14 illustrates
several high-security housings designed specif-ically for
mounting in jails and detention and holding cells, to provide
maximum protection from vandalism. These integrated cameras
have no exposed hardware or cabling and all use heavy-duty
high security locks with tamper switches. The housings are
fabricated from 10-gauge (0.134-inch thick) or heavier welded
steel. The win-dow material is 3/8–1/2-inch polycarbonate or
cast acrylic plastic having an abrasion-resistant finish. These
housings withstand blows and impacts from hammers. Rocks
and
FIGURE 15-14 High security integrated cameras

(A) CEILING MOUNT (B) IN-WALL
(C) CORNER MOUNT (D) WALL MOUNT
some firearm projectiles cannot penetrate or destroy the
integrity of the housing.
15.5.2 Elevator

Figure 15-15 illustrates an example of a hardened cam-era/lens
housing designed specifically for elevator appli-cations. The
photograph of the elevator interior illustrates that the full
interior of an elevator can be monitored from one wide-angle
camera/lens system.
The elevator housing style is available in three sizes: 6, 8, and
12 inches high. These high-security housings are fabri-cated
from welded stainless steel with a 1/4-inch thick poly-carbonate
window. The tamperproof integrated camera assembly is
complete with a monochrome or color CCD camera and a wide-
angle, 90 FOV lens in the stainless-steel housing. In this
configuration, the camera can be tilted
±10 for minor adjustments of the vertical pointing direc-tion.
The high-security housing has a hinged, lockable cover for easy,
controlled access to all internal parts, and
a tough mar-resistant polycarbonate (Lexan) viewing win-dow.
All mounting and camera power and video electrical cable
access holes are located on the rear and top surfaces, and are
inaccessible to the public. The installation meets codes that
require unbroken firewalls. The three housing sizes
accommodate most CCD solid-state cameras using wide-angle
manual- or automatic-iris or vari-focal lenses. These integrated
cameras can also accommodate cameras with infrared LED
lighting to obtain excellent viewing under completely dark,
unlighted conditions. The small 6-inch high unit accommodates
and protects all small 1/4-and 1/3-inch format cameras and
associated wide-angle lenses. Figure 15-16 illustrates the
camera viewing and pointing parameters for elevator-cab
surveillance.

The optimum pointing direction for the lens and cam-era is 45
with respect to both adjacent walls and 45 down from the
ceiling horizontal plane. With a wide-angle, 90 horizontal FOV
the entire elevator cab is viewed with no hidden areas,
providing 100% video coverage of the cab area. Since the
housing is exposed to the public, it is securely locked and is
manufactured using tamperproof steel and stainless steel, and a
polycarbonate window.
399
(A) STAINLESS STEEL HOUSING (B) CAMERA VIEW
FIGURE 15-15 High security corner mount elevator integrated
camera

CEILING
WALL WALL
VERTICAL
FOV = 75°
HORIZONTAL
FOV = 95°
LENS VERTICAL POINTING
DIRECTION: 45°
LENS HORIZONTAL POINTING FLOOR
DIRECTION: 45°
FIGURE 15-16 Elevator cab viewing parameters
15.5.3 Dust-Proof and Explosion-Proof
The dust-proof housing is similar to many other camera
housings except that it is totally sealed from the out-side
atmosphere and therefore can be used in sandy and dusty
environments (Figure 15-17). When fabricated from stainless
steel, these housings can withstand the effects of corrosive

environments. The window material is tem-pered glass to
provide safety and maximum resistance to abrasion and
corrosion. To provide some cooling of the camera and lens, a
fan is used to circulate the air inside the housing, and an
optional sun shield above the camera housing protects it from
direct solar heating. The housing is provided with air fittings so
that an external, filtered,
compressed-air supply can be used to maintain moderate
operating temperatures. These housings are not consid-ered
explosion-proof.
Explosion-proof housings are designed to meet the rig-orous
safety requirements of explosion-proof and dust-ignition-proof
electrical equipment, for installation and use in hazardous
locations (Figure 15-18). These security housings and cameras
meet the requirements of the National Electric Code Class 1,
Division 1, and Class 2, Division 1, and are certified as per the
require-ments of UL 1203 specifications and procedures. These
housings are generally of heavy-wall, all-aluminum con-
struction and are available in 6, 8, and 10 inch diam-eters to
accommodate most camera/lens combinations.

FIGURE 15-17 Dust-proof integrated housing and camera
assembly
FIGURE 15-18 Explosion-proof integrated housing and camera
assembly
They are fitted with explosion-proof, sealable fittings for
electrical power/control input and video signal output. Optional

sun shrouds are available for operation in hot environments.
15.5.4 Pressurized and Air- or Water-Cooled
Pressurized housings are used in hazardous atmospheres. They
meet these requirements by purging (filling) them with an inert
gas at a pressure in accordance with National Fire Protection
Association specification Number 946 (Figure 15-19).
The housings are fabricated from thick-walled alu-minum with
corrosion-resistant finishes. The window is 1/2-inch-thick
tempered and polished plate glass. These housings can be back-
filled (purged) with low-pressure nitrogen gas to a pressure of
15 pounds per square inch gage (psig). Nitrogen is completely
inert and prevents an explosion from occurring if there is any
spark or electrical malfunction in the housing. The housings
have hermeti-cally sealed O-ring seals located between the
access cover and the housing. All electrical terminations are
made and brought out through hermetic seals. To purge the
housing, the access cover is mounted and secured, and the
housing is filled with dry nitrogen to a pressure of 15 psig by
means of a filling valve and pressure-relief valve. The purge is
then closed and the nitrogen filling tube removed. These
housings are significantly more expensive than standard
housings, since they must be designed to be

(A) PRESSURIZED OUTDOOR DOME (B) PRESSURIZED
AND NITROGEN PURGED (C) WATER COOLED
FIGURE 15-19 Pressurized and water cooled environmental
housings
401
hermetically sealed to provide a positive pressure of 15 psig
differential pressure, and to withstand an explosion.
Water-cooled housings are designed for use in extremely hot
indoor or outdoor locations. They require a constant supply of
cooling water for proper operation. A 1-inch-thick water jacket
built into the housing effectively shields the camera/lens from
the outside environment. Depending on the application, the
housings are made
on these ratings is included on the manufacturer’s liter-ature,
and detail information can be obtained from the NEMA
organization. Table 15-1 summarizes several NEMA housing
ratings for indoor and outdoor designs.

15.7 HOUSING ACCESSORIES
from aluminum or stainless steel. An internal fan pro-
There are numerous accessories available for indoor and
outdoor housings. Some of the more common types
vides constant air circulation within the housing, aids in
include thermostatically controlled heaters and fans, win-
efficient heat transfer to the water jacket, and prevents
dow wipers and washers, sun shields and shrouds, and
heat buildup. The housing is supplied with a 1/4-inch-
many types of mounts and brackets.
thick Pyrex heat-resistant window for operating at temper-

atures up to 550 F (288 C). Consult the manufacturer to obtain
recommendations for the specific operating environment.
15.6 NEMA ENVIRONMENTAL RATINGS
The NEMA has developed a comprehensive set of specifi -
cations and ratings for indoor and outdoor electrical hous-ings.
Many of the manufacturers of video security housings and
integrated camera modules have designed their prod-ucts to
meet some of these housing ratings. Information
Heater and Fan. In warmer climates where the tem-perature does
not drop below freezing, only a fan and thermostat are required
to maintain suitable operating temperatures for the camera and
lens. The thermostat is designed to automatically turn on the fan
when the tem-perature in the interior of the housing rises above
some value, usually between 90 and 100 F (32–38 C), and turn
it off when it falls a few degrees below the set tempera-ture. In
cold climates, a heater and thermostat are used to keep the lens
and camera above about 45–55 F (7–13 C). The heater prevents
condensation on the window and lens and keeps the automatic-
iris mechanism and cam-era operative. In freezing weather, it
prevents moisture
PROVIDES PROTECTION AGAINST THE
FOLLOWING ENVIRONMENTAL CONDITIONS

APPROXIMATE IP EQUIVALENT **
INCIDENTAL CONTACT WITH ENCLOSED EQUIPMENT
INDOOR
OUTDOOR
FALLING DIRT
DRIPPING AND LIGHT SPLASHING LIQUIDS
RAIN, SLEET AND SNOW
CIRCULATING DUST, LINT, FIBERS, DEBRIS
SETTLING DUST, LINT, FIBERS, DEBRIS
EXTERNAL ICE
HOSEDOWN AND SPLASHING WATER OIL AND COOLANT
SEEPAGE
OIL AND COOLANT SPRAYING AND SPLASHING
CORROSIVE AGENTS
OCCASIONAL TEMPORARY SUBMERSION
OCCASIONAL PROLONGED SUBMERSION
· NEMA—NATIONAL ELECTRICAL MANUFACTURERS
ASSOCIATION
· IP—INGRESS PROTECTION CLASSIFICATION
4 AND 4X ARE THE MOST COMMONLY USED OUTDOOR

TYPES 12 AND 13 ARE THE MOST COMMONLY USED
INDOOR TYPES
NEMA ENCLOSURE TYPE *
1
3
4
4X
6
6P
12
13
IP30
IP64
IP66
IP66
IP65
IP65
Table 15-1 NEMA Housing Ratings for Non-Hazardous
Locations
from freezing on the window and within the environmen-tally

enclosed housing. The thermostat applies power to the heater
when the temperature goes down below the dew point. When the
camera/lens housing is located in an interior close-to-the-ceiling
environment or in an out-door warm environment, a
thermostatically controlled fan is used to cool the camera/lens
combination. The fan should contain a removable filter that can
be cleaned or replaced periodically.
Heaters require considerable electrical power for their
operation. Most heater assemblies supplied by the man-ufacturer
require 24 VAC for their operation. If primary power is
supplied from a 117 VAC source then a step-down transformer
with a 24 VAC output is required. If 117 VAC power is not
locally available at the camera-housing site, the wire supplying
the power must be sized correctly. Table 15-2 lists appropriate
wire sizes vs. distance between the 117 and 24 VAC sources and
the camera.
Window Washer and Wiper. Another accessory is the win-dow
washer and wiper. If the housing is rectangular and pointing
down at 15 or 20 or more, it is generally unnec-essary to
provide the housing with a window wiper and washer, as rain
will run off the window, along with dirt, and allow proper
viewing. If, however, the housing is located in a dusty
environment or is in a more horizontal direction, it is advisable
to include a window washer/wiper assem-bly. This assembly is
mounted below and in front of the
window and operates like an automobile washer/wiper sys-tem.
The wiper motor and liquid washing pump can be energized
automatically and periodically or remotely from the control
console.

Most environmental housings, indoor or outdoor, are supplied
with plastic or safety (tempered) glass windows for the lens to
view through. These windows may be acrylic, polycarbonate, or
glass, depending on the design. The choice of acrylic vs.
polycarbonate depends on whether the application is to be
maximally tamperproof or only mod-erately so, and whether the
housing is used indoors or out-doors. Acrylic is optically clear
and will transmit over 95% of the light. Polycarbonate transmits
less—approximately 85%—but has a higher impact resistance
than acrylic. Both types are available in a mar-resistant type
which is highly recommended, and will remain optically clear
under nor-mal cleaning action and withstand outdoor
weathering. For maximum resistance to scratching and abrasion,
safety glass is used. Window thicknesses range from 1/8 inch
for light duty to 1/4 inch for normal service and from 3/8-to
1/2-inch for maximum security housings. For dome systems the
portion of the dome that is used for viewing has its surface
pointing downward and tends to be self-cleaning; however, they
must be cleaned periodically and water droplets on the surface
will reduce visibility.
Tamper Switch. In most security applications, it is impor-tant
that when the camera housing is being opened by authorized or
unauthorized personnel, the system or
POWER
CONDUCTOR
RESISTANCE
POWER TO HEATER AND CAMERA OVER TWO
CONDUCTOR CABLE

MAXIMUM CABLE LENGTH (ft)
SOURCE
SIZE
ohms/1000 ft †
VOLTAGE
AWG *

25 WATT LOAD (0.21 AMP)
22
33.0
1656
828
414
20
20.8
2628
1314
657
117 VAC
18
13.02
4198
2099
1050

16
8.18
6683
3341
1671
14
5.16
10594
5297
2649
12
3.24
16872
8398
4199

22
33.0
69
34
17.3
20
20.8
110
55
27.5
18
13.02
176
88
44
24 VAC
16
8.18
281
140.5
70
14
5.16
445

222.5
111
12
3.24
709
354.5
177
10
2.04
1127
563.5
284
POWER

SOURCE
CAMERA
HEATER, ETC.
117 VAC, 24 VAC, 12 VDC ††

· AMERICA WIRE GUAGE
· RESISTANCE REPRESENTS FULL WIRE LENGTH, I.E. 2x
CABLE LENGTH
· IF 12 VDC POWER IS USED, USE THE 24 VAC TABLE
ABOVE AND DOUBLE THE WIRE LENGTH Note: BASED
ON MAXIMUM VOLTAGE DROP OF 10%
Table 15-2 Wire Size vs. Distance for Housing Heater,
Camera, and Other Electronics
403

(A) WALL (B) CEILING (C) OUTDOOR DOME MOUNT
FIGURE 15-20 Camera housing brackets and mounts
guard be alerted. An electrical switch in the camera hous-ing is
used to activate an electrical alarm that can be sent back to the
monitoring location when the housing has been opened.
Locks, Security Screws. There are various levels of secu-rity
key locks available for indoor and outdoor housings. Most
camera housings are supplied with standard locks, but these can
be upgraded to high security locks when the application
demands it. In place of key locks various types of security
screw hardware is available. The manufacturer should be
consulted on the different types of key locks and security
screws that can be supplied.
Brackets and Mounts. A large variety of brackets and mounts
are available to mount cameras, housings, and pan/tilt platforms
safely to walls, ceilings, poles, pedestals, and other structures.
Since most mounts are not compati-ble from manufacturer to
manufacturer, the housing and bracket should be purchased

from the same manufacturer to avoid extra costs for reworking
parts that do not inter-face properly. Figure 15-20 shows some
common camera housing brackets and mounts available.
15.8 HOUSING GUIDELINES
The EIA has written a guideline of recommended design
parameters for housing manufacturers for hole configu-rations
on mounting brackets and housing mountings. At present, not
all manufacturers use the same mounting-hole configuration.
The EIA has recommended guidelines
for the electrical input/output wiring and connector con-
figurations so that there is interchangeability between
manufacturers and so that safe procedures are followed by
manufacturers and installers. Local building codes and UL
codes specify the minimum requirement for electrical enclosure
materials. They should be consulted to be sure materials are
suitable. The purchaser must be aware of the requirements for
each application and look carefully at the manufacturer’s
specifications to determine the most suitable housing. The
NEMA housing recommendations should be consulted to help
determine the specific rating for indoor or outdoor housings.
15.9 SUMMARY
The security camera housing plays an important role in
protecting the camera and lens from the environment and
vandalism, and insuring that they will be in a safe and
controlled environment to maximize life and picture quality.
Many camera housing designs are available for indoor and
outdoor applications.

In an effort to reduce the complexity of choosing a com-patible
lens, camera, and other accessories at the camera site, the
integrated camera design has evolved. This inte-grated design is
lower in cost and requires less installation time resulting in an
additional cost savings.
There are many specialty housings to protect the camera and
lens in harsh environments and from extreme vandal-ism. With
the large number of housing manufacturers to choose from,
there is a housing configuration for almost any application.
Chapter 18
Covert Video Surveillance

CONTENTS
18.1 Overview
18.2 Covert Techniques—Background
18.3 Covert Lens/Camera Types
18.3.1 Pinhole Lenses
18.3.2 Convertible Pinhole Lens Kit
18.3.3 Mini-Lenses
18.3.3.1 Off-Axis Optics
18.3.3.2 Optical Attenuation Techniques
18.3.3.3 Mini-Camera/Mini-Lens
Combination
18.3.4 Comparison of Pinhole Lens and
Mini-Lens
18.3.5 Sprinkler-Head Pinhole Lenses
18.3.6 Mirror-Pinhole Lens
18.3.7 Fiber-Optic Lenses

18.3.7.1 Configuration
18.3.7.2 Rigid Fiber Pinhole Lens
18.3.7.3 Flexible Fiber
18.3.7.4 Image Quality
18.3.8 Bore-Scope Lenses
18.4 Special Covert Cameras
18.4.1 PC-Board Cameras
18.4.2 Remote-Head Cameras
18.5 Infrared Covert Lighting
18.5.1 Concealment Means
18.5.2 IR Sources
18.6 Low-light-level Cameras
18.7 Imbeded Covert Camera Configurations
18.8 Wireless Transmission
18.9 Covert Checklist
18.10 Summary

18.1 OVERVIEW
Overt video surveillance equipment is installed in full viewof
the public and is used to observe personnel and activity
and letting people know that they are under surveillance. Overt
video has had the effect of deterring crime of all types. Covert
video ideally operates so that the offender is not aware of the
surveillance. It can be recorded to produce a permanent video
recording for later use in con-fronting, dismissing, or
prosecuting the offender. Overt video security installations are
very useful in apprehend-ing offenders; however, in special
situations, investigators, police officials, government agencies,
retail operations, and security personnel require covert or
hidden cameras.
Covert and overt video are often used together to foil
professional criminals. The criminal, seeing the overt sys-tem,
defeats or disables the overt cameras, but the covert cameras
can still record the activity. An unrelated reason for using
covert video is to avoid changing the architec-tural aesthetics of
a building or surrounding area.

Covert video cameras and lenses have become com-monplace,
and although these hidden cameras use small optics, they can
produce high-quality video images. Covert video cameras are
concealed in common objects or located behind a small hole in
an opaque barrier (such as a wall or ceiling). Cameras are
camouflaged in common objects such as lamps and lamp
fixtures, table and wall clocks, radios, or books. A very
effective covert system uses a camera and lens camouflaged in a
ceiling-mounted sprin-kler head.
This chapter will analyze covert video principles, tech-niques,
and unique pinhole lenses and cameras. Lenses are analyzed that
have a small front lens diameter thereby permitting the lens and
camera to view the scene through a 1/16-inch-diameter hole.
Most of these lenses have a medium-to-wide FOV, from 12 to
78 , to cover a large scene area, but still permit identification of
persons and the monitoring of activities and actions. Special
pin-hole lens variations including right-angle, automatic-iris,
sprinkler-head, and fiber optic are described, as well as small
pinhole cameras combining a mini-lens and sensor
445
into a small camera head and other complete minia-ture
cameras.
In low-light-level (LLL) applications, a CCD camera with a
very sensitive sensor and IR light source or an image intensifier
is used. Since many covert installations are tem-porary, wireless
transmission systems are used to send the camera signal to the
monitor, recorder, or video printer.

18.2 COVERT TECHNIQUES—BACKGROUND
The lens and camera concealment is accomplished by having the
lens view through a small hole, a series of small holes, or from
behind a semitransparent window. Figure 18-1 shows a typical
room in which covert video surveillance is installed.
A number of suitable covert camera locations include the
ceiling, a wall, a lamp fixture, a clock, or other articles
normally found in the room. Video cameras are installed in one
or more locations in the room depending on the activity
expected. Covert video systems using small lenses pose unique
optical problems compared with overt systems that use standard
lenses. Since the diameter of the front lens that views the scene
must, by necessity, be small in order to be hidden, the lens is
designed to be
optically fast, collecting and transmitting as much light as
possible from the reflected scene to the camera sen-sor. As a
consequence, small-diameter lenses called pin-hole lenses are
used. (The term pinhole is a misnomer, as these lenses have a
front diameter anywhere from 1/16 to 3/8 inch.)
There are several misconceptions regarding the factors
determining a good pinhole camera or lens system for covert
applications. Figure 18-2 shows the covert security problem.
The lens/camera must receive reflected light from an
illuminated scene. The lens must collect and transmit the light
to the camera sensor and the camera must transmit the video
signal to a remote video moni-tor and/or recorder and video
printer. Most covert pin-hole lenses are designed for 1/4- and
1/3-inch camera sensor formats. For indoor applications the
light sources are typically fluorescent, metal-arc, mercury, or
tungsten types. Outdoor light sources include sunlight in the

day-time, and mercury, metal-arc, tungsten, sodium, or xenon
lighting at night. Figure 18-3 shows two basic configura-tions
for pinhole lenses and cameras located behind a barrier.
The hole in the barrier is usually chosen to be the same
diameter (d) or smaller than the pinhole lens front lens element.
When space permits the straight-type

FIGURE 18-1 Covert CCTV lens/camera environment
447
ILLUMINATION
SOURCE
SCENE
SMALL HOLE
LENS FOV
IN WALL

COVERT
PINHOLE LENS
ROOM AND CAMERA
BARRIER
INTEGRAL SMALL CAMERA
AND PINHOLE LENS
FIGURE 18-2 Covert CCTV surveillance
installation is used. In confined or restricted locations with
limited depth behind the barrier, the right-angle pinhole
lens/camera is used. In both cases, to obtain the full lens FOV it
is imperative that the pinhole lens front lens ele-ment be located
as close to the front of the barrier as possible to avoid
“tunneling” (vignetting). When the pin-hole lens front lens
element is set back from the barrier surface, the lens is, in
effect, viewing through a tunnel, and the image has a narrower
FOV than the lens is capable of producing. This appears on the
monitor as a porthole-like (vignetted) picture.
An important installation problem often initially over-looked is
the lens pointing angle required to see the desired FOV (Figure
18-4). Many applications require that the lens/camera point
down at a shallow depression angle (30 ) from the ceiling
(Figure 18-4a). This is accomplished by using the small-barrel,
slow-taper lens. This feature allows pointing the small-barrel

lens over a larger part of a room than the wide-barrel lens. Not
all lenses can be mounted at a small angle to the ceiling because
of the lens barrel shape (Figure 18-4b). Lenses having a large
barrel diameter and fast taper at the front cannot be mounted at
the shallow angles required. The small-barrel, slow-taper design
permits easier installation than the fast-taper since less material
must be removed from the barrier, and the
lens has a faster optical speed, since the front lens element is
larger and collects more light. Figure 18-4 illustrates this
installation problem. It shows a small hole on the scene side of
the barrier and some material cut out of the barrier behind it to
permit the front lens element to be located close to the front of
the barrier surface. A pinhole lens having a small front diameter
is simple to install. The smaller tapered barrel can be mounted
at a smaller angle to the barrier than the wide-barrel lens. This
feature allows pointing the small-barrel lens over a larger part
of a room than the wide-barrel lens.
18.3 COVERT LENS/CAMERA TYPES
Pinhole lenses and cameras used for covert security appli -
cations include: standard pinhole, compact pinhole lens kit, and
mini-lens. There are many single board covert camera designs
available using a small lens mounted to a single printed circuit
(PC) board housed in a plastic or metal housing (Section
18.4.1). Special covert lens and camera designs include: fiber
optic, sprinkler-head, and covert camera/lens combinations
uniquely configured in special housings.

LIGHT SOURCE:
SUN
DETAIL OF
LAMPS: •
FLOURESCENT
PINHOLE LENS
•
TUNGSTEN (HALOGEN)
INSTALLED IN
•
SODIUM
ROOM BARRIER
•

MERCURY
•
INFRARED
REFLECTED LIGHT
FROM SCENE
SMALL
HOLE IN
WALL
SCENE

STRAIGHT
PINHOLE LENS
AND CAMERA
ROOM
BARRIER
RIGHT ANGLE
PINHOLE LENS
AND CAMERA
MONITOR

FIGURE 18-3 Straight and right angle pinhole installation
18.3.1 Pinhole Lenses
Figure 18-3 shows how pinhole lenses and cameras are mounted
behind a wall, with the lens viewing through a small hole in the
wall. Most are designed for 1/4 -, and 1/3-inch format cameras
and have a manual- or automatic-iris control to adjust the light
level reaching the camera. Figure 18-5 shows several samples of
the generic pinhole lens types available.
The right-angle version permits locating the camera and lens
inside a narrow wall or above a ceiling. The optical speed or f-
number (f/#) of the pinhole lens is important for the successful
implementation of a covert camera sys-tem. The lower the f-
number, the more light reaching the camera and the better the
video picture. The best theoret-ical f-number is equal to the lens
focal length (FL) divided by its entrance lens diameter (d):
f /# = FL/d
(18-1)
This theoretical f-number cannot be obtained in practice
because of various losses caused by imperfect lens trans-
mission that is caused by reflection, absorption, and other lens-
imaging properties. The light getting through the
pinhole lens to the camera sensor is limited primarily by the
diameter of the front lens or the mechanical open-ing through
which it views. The larger the lens entrance diameter, the more
light getting through to the camera sensor, resulting in better
picture quality, all other condi-tions remaining the same. The

light collected and trans-mitted through a lens system varies
inversely as the square of the lens f-number. If the lens
diameter is increased (or decreased) a small amount, the light
passing through the lens increased (or decreases) by a large
amount: if the lens diameter is doubled, the light throughput
quadru-ples. An f/2.0 lens transmits four times as much light as
an f/4.0 lens. The f-number relationship is analogous to water
flowing through a pipe: if the pipe diameter is dou-bled four
times as much water flows through it. Likewise if the f-number
is halved, four times as much light will be transmitted through
the lens.
Many types of covert lenses are commercially available for
video surveillance applications. Table 18-1 summarizes the
characteristics of most manual- and automatic-iris pin-hole
lenses.
Most of these lenses are designed for 1/4 - and 1/3-inch sensor
formats since covert cameras are small. In spite of their small
size they have resolutions of 380–420 TV
449
(A) SLOW-TAPER BARREL
(B) FAST-TAPER BARREL
SMALL
LARGE
DIAMETER

DIAMETER
30°
55°
FIGURE 18-4 Pinhole lens pointing angle
SLOW TAPER FAST TAPER
STRAIGHT
STRAIGHT
MANUAL IRIS

MANUAL IRIS
RIGHT ANGLE
MANUAL IRIS
STRAIGHT
RIGHT ANGLE AUTO IRIS
AUTO IRIS
FIGURE 18-5 Standard straight and right-angle Pinhole lenses
ANGULAR FIELD OF VIEW (FOV) IN DEGREES
FOCAL
CAMERA FORMAT (inch)
TYPE

HORIZ
VERT
HORIZ
VERT HORIZ
VERT
2.6
2.5
62.4
46.8
83.2
62.4
STRAIGHT

4.0
2.0
51.2
38.4
68.5
43.1
STRAIGHT

5.5
3.0
30.2
23.6
38.7
31.0
60.4
47.2
STRAIGHT
6.2
2.0
32,4
24.3
42.8
30.1
56.1
42.1
STRAIGHT
8.0
2.0
21.8
16.7
29.4
22.0
43.6
33.4
STRAIGHT
8.0
2.2
21.8
16.7
29.4
22.0

43.6
33.4
RIGHT-ANGLE
9.0
3.4
22.3
16.8
29.5
22.1
39.1
29.3
STRAIGHT
11.0
2.3
16.2
12.3
21.6
16.1
32.4
24.6
STRAIGHT
11.0
2.5
16.2
12.3
21.6
16.1
32.4
24.6
RIGHT-ANGLE
Table 18-1 Covert Pinhole Lens Parameters

MOUNT
CS
CS
C/CS
CS
C/CS
C/CS
C/CS
C/CS
C/CS
lines for a 1/4 - or 1/3-inch color camera and 450–570 TV lines
for monochrome cameras. Many pinhole lenses have very small
entrance apertures: 0.10 inch (2.5 mm) and are therefore
optically slow (f/3.5–f/4.0) by design. From Equation 18-1 a
lens with a FL of 9 mm and a 2.5 mm aperture (d) has at best a
theoretical f-number of:

f/# = 9 mm/2.5 mm = 3.6
(18-2)
Other lens losses within this type of lens give an overall optical
speed of approximately f/4.0.
A covert lens with an 11 mm FL and a 6 mm aperture has a
theoretical f-number of:
f/# = 11 mm/6 mm = 1.83
(18-3)
Other lens losses result in an overall optical speed of
approximately f/2.0. This means that the 11 mm lens col-lects
four times as much light as the 9 mm lens.
The 9 mm lens with the smaller aperture works well if there is
sufficient light. An advantage of the 6 mm-aperture
(approximately 0.25 inch) lens is that it can be used in
applications where a larger hole, that is, 6 mm diameter
adequately conceals the lens and there is insufficient light
available for the 9 mm FL lens with the 2.5 mm hole. The most
important characteristics of a pinhole lens are: (1) how fast is
the lens optical speed—that is, how low is the lens f-number
(the lower the better) and (2) ease of installation and use. When
covert operation is required in locations having widely varying
light-level conditions or in a low light level application, a high
sensitivity CCD solid-state or other intensified LLL camera
used with a pinhole lens with an automatic iris controlling the
light reaching the camera sensor is necessary. Shuttered CCD
cameras may tolerate the use of manual-iris lenses. Check with
the
manufacturer for the light range over which the camera will
operate. Figure 18-5 shows straight and right-angle pinhole

lenses with manual and automatic irises capable of controlling
the light level reaching the camera sensor over a 35,000-to-1
light-level range.
A generic characteristic of almost all pinhole-type lenses is that
they invert the video picture and therefore the cam-era must be
inverted to get a normal right-side-up picture. Some right-angle
pinhole lenses reverse the image right to left and therefore
require an electronic scan-reversal unit (Section 16.4) to regain
the correct left-to-right ori-entation. Some pinhole lenses have a
focusing ring or the front element of the lens can be adjusted to
focus a sharp image on the camera sensor.
18.3.2 Convertible Pinhole Lens Kit
Pinhole lenses have been manufactured for many years in a
variety of focal lengths (3.8, 4, 5.5, 6, 8, 9, 11 mm), in straight,
right-angle, and manual- and automatic-iris con-figurations. The
FL of most of these lenses can be dou-bled to obtain one-half
the FOV by using a 2X extender. Pinhole lenses with 16 mm
and 22 mm FLs are achieved by locating a 2X magnifier in
between the 8 and 11 mm lenses and the camera. This
automatically doubles the f-number of each lens (only one-
fourth of the light transmitted). In many applications, the
required FLs and configuration are not known in advance, and
the user (or dealer) must have a large assortment of pinhole
lenses, or take the risk that he will not have the right lens to do
the job. This dilemma was solved with the pinhole lens kit
(Figure 18-6).
Eight different FL lenses can be assembled in either a straight
or right-angle configuration within minutes with

· RIGHT-ANGLE SPRINKLER LENS ASSEMBLED FROM
KIT
(B) LENS KIT IN CASE

FIGURE 18-6 Pinhole lens kit
this kit of pinhole lens parts. An additional four combina-tions
can be assembled in the form of a disguised sprinkler-head
covert application (Section 18.3.5). All lenses have a manual
iris with automatic iris optional). Table 18-2 lists all the lens
combinations for this versatile pinhole lens kit.
451
Tables 18-3 and 18-4 tabulate the scene areas (width and
height) as viewed with the popular pinhole lenses on 1/4 - and
1/3-inch sensor format cameras.
Several points should be considered when using stan-dard, fully
assembled pinhole lenses or pinhole lenses made from the
pinhole lens kit:
· Straight pinhole lenses invert the picture; therefore, the
camera should be mounted in an inverted orientation.
· Some right-angle pinhole lenses will show a right-to-left
picture orientation instead of left-to-right, as with normal
lenses. A camera SRU will correct the problem. Check with the
manufacturer.
· The straight pinhole lens with the sprinkler-mirror attachment
displays a right-to-left picture. Use an elec-tronic SRU to
correct the problem. The right-angle sprinkler-mirror version
displays a correct left-to-right picture.
As an example: choose a pinhole lens and camera to view a

scene 6 feet high by 8 feet wide at a distance of 15 feet using a
1/4 -inch format camera. Use Table 18-3 and choose an 11 mm
FL lens. As another example, the scene area displayed on the
monitor with an 8 mm lens on a 1/3-inch format camera in a
ceiling at a distance of 20 feet is an area 22 feet wide by 16.4
feet high (Table 18-4).
Note that the FOV when using any of the medium- to long-FL
lenses is independent of the hole size through which the lens
views, providing the hole produces no tun-neling. Viewing
through a wall with a wide-angle 4 –8 mm FL pinhole lens may
require a cone-shaped hole or an array of small holes to prevent
tunneling (vignetting) of the scene image.
18.3.3 Mini-Lenses
Mini-lenses and a mini-lens camera kit consisting of five
interchangeable mini-lenses and a very small CCD camera are
described in this section. Mini-lenses are small FFL objective
lenses used for covert surveillance when space is at a premium
(Figure 18.7).
FOCAL LENGTH (mm)
f/#
CONFIGURATION
IMAGE ORIENTATION
COMMENTS
11

2.3
STRAIGHT
NORMAL
PINHOLE LENS
8
2.0
STRAIGHT
NORMAL
PINHOLE LENS
11
2.5
RIGHT ANGLE
REVERSED
PINHOLE LENS
8
2.2
RIGHT ANGLE
REVERSED
PINHOLE LENS
22
4.6
STRAIGHT
NORMAL
PINHOLE LENS
16
4.0
STRAIGHT
NORMAL
PINHOLE LENS
22
5.0
RIGHT ANGLE
REVERSED
PINHOLE LENS
16
4.4

RIGHT ANGLE
REVERSED
PINHOLE LENS
11
2.3
STRAIGHT
NORMAL
SPRINKLER HEAD
22
4.6
STRAIGHT
NORMAL
SPRINKLER HEAD
11
2.5
RIGHT ANGLE
REVERSED
SPRINKLER HEAD
22
5.0
RIGHT ANGLE
REVERSED
SPRINKLER HEAD
Table 18-2 Pinhole Lens Kit Combinations and Parameters

1/4 inch SENSOR FORMAT LENS GUIDE
PINHOLE
CAMERA TO SCENE DISTANCE (D) IN FEET
LENS
WIDTH AND HEIGHT OF AREA (W ×H ) IN FEET
FOCAL
5
10
15
20
25
30
LENGTH

(mm)
W × H
W × H
W × H
W × H
W × H
W × H
2.6
12.3
× 9.2
24.6 × 18.5
36.9 × 27.7
49.2
× 36.9
61.5 × 46.1
74.0 × 55.5
3.7
8.5
× 6.5
17.3 × 13.0
30.0 × 19.5
34.6
× 26.0
43.3 × 32.5
60.0 × 39.0
4.0

8.0
× 6.0
16.0 × 12.0
24.0 × 18.0
32.0
× 24.0
40.0 × 30.0
48.0 × 36.0
6.2
5.2
× 3.9
10.4 × 7.8
15.6 × 11.7
20.8
× 15.6
26.0 × 19.5
31.2 × 23.4
8.0
4.0
× 3.0
8.0
× 6.0
12.0 × 9.0
16.0
× 12.0
20.0 × 15.0
24.0 × 18.0
9.0
3.6
× 2.7
7.2
× 5.4
10.8 × 8.1

14.4
× 10.8
18.0 × 13.5
21.6 × 16.2
11.0
2.9
× 2.2
5.8
× 4.4
8.7 × 6.6
11.6 × 8.8
14.5 × 11.0
17.4 × 13.2
16.0
2.0
× 1.5
4.0
× 3.0
6.0 × 4.5
8.0
× 6.0
10.0 × 7.5
12.0 × 9.0
22.0
1.5
× 0.8
2.9
× 2.2
4.4 × 3.3
5.8 × 4.4
7.3 × 5.5
8.7 × 6.6

Table 18-3 Pinhole Lens Guide for 1/4-Inch Format Camera
1/3-inch SENSOR FORMAT LENS GUIDE
PINHOLE
CAMERA TO SCENE DISTANCE (D) IN FEET
LENS
WIDTH AND HEIGHT OF AREA (W × H) IN FEET
FOCAL
5
10
15
20
25
30

LENGTH
(mm)
W × H
W × H
W × H
W × H
W × H
W × H
2.6
8.6 × 6.5
16.9 × 12.6
25.8 × 19.4
33.8 × 25.2
43.1 × 32.3
50.8 × 37.8
3.7

6.1 × 4.5
11.9
× 8.7
18.2 × 13.6
23.8 × 17.7
30.3 × 22.7
35.7 × 26.6
4.0
5.6 × 4.2
11.2
× 8.2
16.8 × 12.6
22.0 × 16.4
28.0 × 21.0
33.0 × 24.6
6.0
3.7 × 2.8
7.3
× 5.5
11.2 × 8.4
14.7 × 10.9
18.7 × 14.0
22.0 × 16.4
8.0
2.8 × 2.1
5.5
× 4.1
8.4 × 6.3
11.0 × 8.2
14.0 × 10.5
16.5 × 12.3
9.0

2.5 × 1.9
4.9
× 3.7
7.5 × 5.7
9.8 × 7.4
12.5 × 9.5
14.7 × 11.1
11.0
2.0 × 1.5
4.0
× 3.0
6.0 × 4.5
8.0 × 6.0
10.0 × 7.5
12.0 × 9.0
16.0
1.4 × 1.1
2.8
× 2.1
4.2 × 3.3
5.6 × 4.2
7.0 × 5.5
8.4 × 6.3
22.0
1.0 × .8
2.0
× 1.5
3.0 × 2.4
4.0 × 3.0
5.0 × 4.0
6.0 × 4.5

Table 18-4 Pinhole Lens Guide for 1/3-Inch Format Camera
Mini-lenses shown have focal lengths of 3.8, 8, 11 mm, etc.
They have front-barrel diameters between 3/8 and 1/2 inch,
making them easy to mount behind a barrier or in close
quarters. Because these small lenses have no iris, they should be
used in applications where the scene light level does not vary
widely, or with electronically shuttered cam-eras. Mini-lenses,
like other FFL lenses and unlike pinhole lenses, do not invert
the image on the camera. Since the small and short (less than
5/8 inch long) mini-lenses have only three to six optical lens
elements, fast optical speeds of f/1.4 to f/1.8 are realized.
Pinhole lenses, on the other hand, are 3–5 inches long, and have
as many as 10–20 optical elements and optical speeds of f/2.0 to
f/4.0. This makes the mini-lens approximately five times faster
(able to collect five times more light) than the pinhole lens.
18.3.3.1 Off-Axis Optics
A useful variation of the mini-lens is one that is mounted with
its optical axis laterally offset from the camera-sensor axis
(Figure 18-8). This offset configuration allows the camera to
view a scene at an angle away from the camera-pointing axis.
The physical amount the optics must be moved to produce a
large offset angle is only a few millime-ters, which is easily
accomplished with this special mini-lens and its modified
mount. The offset angle is chosen so that, with the camera
parallel to a mounting surface, the entire lens FOV views the
scene of interest without view-ing the mounting surface. This
angle is 22 for the 8 mm lens and 15 for the 11-mm when using
a 1/4 -inch format camera. It is 18 and 13 , respectively for the
same lenses when using a 1/3-inch camera. This technique has a
direct

453
CAMERA
1/3" FORMAT CAMERA
SCENE
MINI-LENS
8 mm
lens on camera
11 mm FL
3.8 mm FL
FIGURE 18-7 Mini-lens and optical diagram
ON AXIS

CAMERA
MINI-LENS
LENS OFFSET
FOV
ON-AXIS
LENS
OFF-AXIS
MINI-LENS
CCD SENSOR
FOV

OFF-AXIS
LENS
CEILING MOUNTED CAMERA
CEILING
SIDE VIEW
WALL MOUNTED CAMERA
8 mm OFF-AXIS
WALL
MINI-LENS

TOP VIEW
FIGURE 18-8 Off-axis optics configuration
benefit when a camera/lens is mounted flat against a wall or a
ceiling or other mounting surface (Figure 18-8).
18.3.3.2 Optical Attenuation Techniques
Since mini-lenses do not have an iris, they should be used when
the lighting conditions are fairly constant and do not exceed the
dynamic range of the camera. If the scene is very brightly
illuminated with an intense artificial light or the sun, several
techniques can be used to attenuate the light to the lens/camera
(Figure 18-9).
The first technique is to mount the mini-lens behind a light-
attenuating filter (Figure 18-9a). This may take the form of a
gray neutral-density filter, a partially alu-minized film, or a
tinted/smoked glass or plastic material. Neutral-density filters
are available from photographic sup-ply stores. This technique
uniformly attenuates the light across the full aperture of the

lens. A second technique shown in Figure 18-9b through 18-9e
is to mount the mini-lens behind a small hole, a pattern of small
holes, a slit, or other hole(s). This is accomplished by either
mount-ing a small cap with the hole(s) (Figure 18-9b) onto the
lens, or mounting the lens behind a hole(s) in the barrier (Figure
18-9c –18-9e). The light level reaching the camera sensor can
be set initially by locating the lens behind a hole smaller than
the mini-lens diameter. This technique
attenuates the light reaching the lens but does not do it
uniformly. For medium-FL lenses (11 mm and above), almost
any shape hole results in a satisfactory image on the sensor.
When the 11 or 22 mm mini-lens or pinhole lens is mounted
behind a viewing barrier, a central hole as small as 1/16th of an
inch is suitable for producing a full image of the scene,
providing sufficient light avail-able for the camera. When short
focal length (2.2, 3, 8 mm, etc.) mini-lens or pinhole lens views
through a small hole, an undesirable porthole effect occurs,
which is eliminated by having the lens view through a central
hole and a series of concentric holes located around the central
hole. The hole pattern must extend to the outer limits of the lens
so that the full FOV of the lens is maintained. These concentric
holes enable the lens to have peripheral vision or wide-angle
viewing, and they eliminate vignetting. Figure 18-9(b, d) shows
two examples of this extended hole pattern. Either technique
can provide attenuations required for sunlit or brightly
illuminated scenes.
18.3.3.3 Mini-Camera/Mini-Lens Combination
A high-sensitivity pinhole camera results when a very fast mini -

lens—f/1.4 to f/2.0—is coupled directly with the camera sensor.
Figure 18-10 illustrates a mini-lens camera kit with three
standard on-axis mini-lenses having focal
(A) UNIFORM LIGHT ATTENUATION
ACROSS LENS APERTURE: (C)
• NEUTRAL DENSITY FILTER
• SMOKED OR TINTED GLASS
OR PLASTIC
(D)
(B–E) DISCRETE APERTURE ATTENUATOR
· SINGLE HOLE
· MULTIPLE HOLES
• SLIT(S) (E)
(B)

FIGURE 18-9 Lens optical attenuation techniques
6 MINI-CAMERA MINI-LENS 3.8 mm FL ASSEMBLED
FROM KIT

12 MINI-CAMERA MINI-LENS KIT
FIGURE 18-10 Mini-camera/mini-lens combination and kit
lengths of 3.8, 8 and 11 mm and two off-axis mounts for the 8
and 11 mm FL lenses, and a very small, sensitive, high-
resolution color CCD camera. The complete camera is only 125
× 125 × 100 inches long. The 11-mm FL lens extends 0.3 inch
in front of the camera. The camera oper-ates directly from 12
volts DC, requires only 1.5 watts of power, and produces a
standard composite video output.
The small lens size and direct coupling to the camera sensor do
not leave room for a manual or automatic iris. The camera has
excellent electronic light-level compensa-tion, but optimum
performance is achieved if the lighting is fairly constant. Under

bright light conditions an atten-uation technique shown in
Figure 18-9 is used.
455
18.3.4 Comparison of Pinhole Lens and Mini-Lens
To compare different pinhole and mini-lenses with respect to
their ability to transmit light to the camera sensor, a light power
factor (LPF) is defined, with a slow pinhole lens (f/4.0) as a
base reference. Table 18-5 summarizes the optical speed (f-
number) and LPF for standard pinhole and mini-lenses.
The f-number is usually critical in nighttime applications with
low light levels and where auxiliary lighting cannot be added.
Table 18-5 illustrates the significantly higher amount of light
passing through the mini-lenses as com-pared with the pinhole
lenses. A camera/lens using an f/1.8 mini-lens transmits almost
five times as much light to the camera sensor as an f/4 pinhole
lens. The f/1.4 mini-lens transmits more than eight times as
much light as the f/4 pinhole lens.
18.3.5 Sprinkler-Head Pinhole Lenses
A very effective covert system uses a camera and lens
camouflaged in a ceiling-mounted sprinkler head. Of the large
variety of covert lenses available for the security video industry
(pinhole, mini, fiber-optic), this unique, extremely useful
product hides the pinhole lens in a ceiling sprinkler fixture,
making it very difficult for an observer standing at floor level
to detect or identify the lens and camera. Figure 18-11a shows

the sprinkler pin-hole lens attached to a standard camera
mounted on a ceiling.
The covert surveillance sprinkler installed in the ceil -ing in no
way affects the operation of the active fire-suppression
sprinkler system; however, it should not be installed in
locations that have no sprinkler system, so as not to give a false
impression to fire and safety personnel that there is a sprinkler
system installed.
The only part of the lens system visible from below is the
standard sprinkler head and the small (3/8×5/8-inch) mirror
assembly. In operation, light from the scene reflect-ing off the
small mirror is directed by the mirror to the front of the pinhole
lens. The 11 or 22 mm pinhole lens transmits and focuses the
scene onto the camera sensor. In the straight version the image
is reversed. In surveil-lance applications this is often only an
annoyance and not really a problem. However, if it needs to be
corrected an electronic SRU will correct this condition. The
right-angle version (Figure 18-11b) corrects this condition and
pro-duces a normal left-to-right image scan. The small mirror
can be adjusted in elevation to point at different scene heights.
To point in a particular azimuth direction, the entire camera-
sprinkler lens assembly is rotated with the mirror pointing in
the direction of the target of interest. When installed, most of
the pinhole lens and the entire
456
CCTV Surveillance

POWER
COMMENTS
LENGTH
f/#
LENS TYPE
CONFIGURATION
1/4-inch FORMAT
1/3-inch FORMAT
FACTOR
(mm)

2.6
2.0
MINI
STRAIGHT
4.0
62.4
46.8
83.2
62.4
ULTRA WIDE-ANGLE
3.8
1.4
MINI
STRAIGHT
8.16
39.1
31.8
52.1
42.4
ULTRA WIDE-ANGLE

8.0
1.6
MINI
STRAIGHT
6.25
21.8
16.7
29.1
22.3
LONG TAPER
11.0
1.8
MINI
STRAIGHT
4.94
16.2
12.3
21.6
16.4
LONG TAPER
25.0
4.0
MINI
STRAIGHT
1.0
7.0
5.3

9.3
7.1
ULTRA WIDE-ANGLE
3.8
2.0
PINHOLE
STRAIGHT
4.0
39.1
31.8
52.1
42.4
ULTRA WIDE-ANGLE
3.8
2.2
PINHOLE
RIGHT-ANGLE
3.31
39.1
31.8
52.1
42.4
ULTRA WIDE-ANGLE
5.5
3.0
PINHOLE
STRAIGHT
1.78

32.3
25.6
43.1
34.1
WIDE-ANGLE
6.2
2.0
PINHOLE
STRAIGHT
4.00
28.0
21.5
37.3
28.7
WIDE-ANGLE
8.0
2.0
PINHOLE
STRAIGHT
4.00
21.8
17.7
29.1
23.6
SHORT, WIDE-ANGLE
8.0
2.2

PINHOLE
RIGHT-ANGLE
3.31
21.8
17.7
29.1
23.6
LONG TAPER
9.0
3.5
PINHOLE
STRAIGHT
1.31
19.4
15.1
25.9
20.1
LONG TAPER
11.0
2.3
PINHOLE
STRAIGHT
3.02
16.2
12.3
21.6
16.4
SHORT TAPER

11.0
2.5
PINHOLE
RIGHT-ANGLE
2.56
16.2
12.3
21.6
16.4
SHORT TAPER
16.0
4.0
PINHOLE
STRAIGHT
1.00
11.0
8.3
14.7
11.1
NARROW-ANGLE

11 INCREASE IN LIGHT LEVEL REACHING SENSOR
BASED ON USING VALUE OF 1.00 FOR AN f/4 PINHOLE
LENS
Table 18-5 Pinhole Lens and Mini-Lens Light Transmission
Comparison
camera is concealed above the ceiling, with only a modi -fied
sprinkler head, a small mirror, and small lens in view. For many
applications this stationary pinhole lens pointing in one specific
direction is adequate. To look in different directions the camera,
sprinkler head, and moving mir-ror assembly are made to pan
(scan) via a motor drive. A motor drive sprinkler scanning
system can provide remote panning capability. A scanning
version of the sprinkler concept has a remote-control 360
continuous panning capability (Figure 18-12).
18.3.6 Mirror-Pinhole Lens
Large plastic domes are often used to conceal a PTZ video
surveillance system from the observer (Chapter 14). The
purpose for concealing the camera and lens in the dome is so
that the observer cannot see the direction in which the camera
lens is pointing or whether there is actually a surveillance
camera. Using this subterfuge, one camera system can scan and
view a large area without the observer knowing at any instant
whether he is under observation. Most domes are from 5 to 10

inches in diameter and drop below the ceiling by 5–8 inches.
The requirement that the lens view through the dome results in a
typical light loss of 50%. A more covert camera/lens assembly
takes the form of a camera, pinhole lens, and small mirror.
If the right angle lens of the sprinkler-head assembly shown in
Figure 18-11 and 18-12 is removed, all that pro-trudes below
the ceiling is a small mirror approximately
3/8 × 5/8 inches. This technique results in a very low pro-file
that is difficult for an observer to detect at ground level. The
pinhole/mirror system provides an alternative to some dome
applications. The system can be fixed or have a 360 panning
range.
Two advantages of the moving mirror system over the dome are:
(1) no large protruding dome suspended below the ceiling and
(2) easy installation. Installation is easy since only a small hole
about 3/4 inch in diameter is required to insert the lens and
mirror through the ceil-ing. The small mirror scanning system
has limitations: (1) it cannot view the scene directly below its
location and
27 there is no zooming. The dome system has two advan-tages
over the scanning mirror: (1) the dome serves as a deterrent
since the observer sees the dome and believes a camera is active
in it but does not know at any instant where the camera is
looking, and (2) the added capability of full-range zoom optics.
18.3.7 Fiber-Optic Lenses
When the barrier between the scene side and the camera/ lens
side is a few inches as in Figure 18-3, a pinhole or mini-lens
and camera can be mounted directly behind the bar-rier. For

difficult covert video surveillance applications in which small
cameras and mini pinhole lenses will not work, coherent fiber-
optic-bundle lenses may be the solution.
Fiber optics are used when it is necessary to view a scene on the
other side of a thick barrier or inside a confined area. The fiber -
optic bundle lens and camera are installed
457
(A) STRAIGHT

(B) RIGHT-ANGLE
FIGURE 18-11 Sprinkler-head pinhole lenses
behind the barrier and the objective lens on the scene side. The
lens viewing the scene can be a few inches or a few feet away
from the camera. There are three opti-cal techniques to transfer
the image, in effect “lengthen” the camera’s objective lens: (1)
a rigid coherent fiber-optic conduit, (2) a borescope lens, and
(3) a flexible fiber-optic bundle. These special lenses can
extend the objective lens several inches to several feet in front
of the camera sensor. The rigid fiber conduit uses a fused array
of fibers and cannot be bent. The flexible fiber lens has hair -
like fibers loosely contained in a protective sheath and can be
flexed and bent easily. These fiber-optic lenses should not be
confused with the single or multiple strands of fiber commonly
used to transmit the time-modulated video signal a long distance
from a camera to a remote monitoring site (Chapter 6). Coherent
fiber-optic lenses typically have 200,000–300,000 individual
fibers forming an image-transferring array. Rigid fiber-optic
lenses are 1/4 –1/2 inch in diameter and from 6 to 12 inches
long. Flexible fiber-optic lenses are from 1/8 to 1¼ inch in
diam-eter and up to several feet long. These fiber-optic lenses
are available with manual or automatic iris for 1/6-, 1/4-, 1/3-,
1/2-, and 2/3-inch video formats.
By combining lenses with coherent fiber-optic bun-dles, a long,
small-diameter optical lens is produced that requires a small
hole for insertion into the barrier. A small aperture hole is
drilled completely through at the barrier

surface and connected to the camera on the protected side
(Figure 18-13).
This lens/camera system has provided the solution for many
banking ATM and correctional-facility security prob-lems. A
minor disadvantage of all fiber-optic systems is that the picture
obtained is not as “clean” as that obtained with an “all -lens”
pinhole lens. These imperfections occur because several
hundred thousand individual hair-like fibers make up the fiber-
optic bundle some of which are not perfectly transmitting. For
most surveillance applica-tions the imperfections do not result
in any significant loss of intelligence in the picture. Figure 18-
14 shows complete rigid and flexible fiber-optic lenses.
18.3.7.1 Configuration
A fiber-optic lens consists of three parts: (1) an objec-tive lens
that focuses the scene onto the front end of the fiber-optic
bundle, (2) a rigid conduit or flexible fiber coherent optic
bundle that transfers the image a substan-tial distance (several
inches to several feet), and (3) a relay lens at the output end of
the fiber bundle that re-images the output image and focuses
onto the camera sensor (Figure 18-15).
The objective lens can be like any of the FFL, zoom, pinhole,
manual-, or automatic-iris lens. The objective lens

PANNING MECHANISM
ABOVE CEILING
CAMERA
RIGHT
ANGLE
CAMERA
PINHOLE
PINHOLE
LENS
LENS

ADJUSTABLE
MIRROR
270°
360°
PANNING
CONTINUOUS
PANNING
FIGURE 18-12 Panning sprinkler-head pinhole lens system

SCENE
CAMERA
3/16" TO 1/2"
DIAMETER HOLE
RIGID FIBER OPTIC
MANUAL
IRIS
(6–12 INCHES LONG)
THICK WALL BARRIER
(6–12 INCHES)
FIGURE 18-13 Fiber-optic pinhole lens installation in thick
wall
459

OBJECTIVE LENS: 8 mm OR 11 mm FL
FIBER TYPE: RIGID CONDUIT
FIBER LENGTH: 6 inches
RELAY LENS: M = 1:1
IRIS: MANUAL
MOUNT: C OR CS
(A) RIGID CONDUIT LENS
OBJECTIVE LENS: ANY C OR CS MOUNT
FIBER TYPE: FLEXIBLE BUNDLE
FIBER LENGTH: 39 inches
RELAY LENS: M = 1:1
IRIS: MANUAL
MOUNT: C OR CS

(B) FLEXIBLE BUNDLE LENS
FIGURE 18-14 Rigid and flexible fiber-optic lenses
must produce an image large enough to fill the full aper-ture
(cross-sectional area) of the fiber-optic bundle. The coherent
fiber-optic bundle consists of several hundred thousand closely
packed glass fibers to coherently transfer an image from one
end of the fiber to the other, several inches to several feet
(Figure 18-16).
Fiber 1 transmits point 1 of the image from the objec-tive lens
down the fiber to a corresponding point 1 on the exit end of the
fiber bundle. Likewise, all of the remain-ing points of the
entrance image are transferred in an exact one-to-one
correspondence to the exit end of the fiber bundle, thereby
producing a coherent image. Coher-ent means that each point in
the image on the front end of the fiber bundle corresponds to a
specific point at the rear end of the fiber bundle.
18.3.7.2 Rigid Fiber Pinhole Lens
The rigid fiber-optic bundle has individual fibers that are fused
together to form a rigid glass rod or conduit and is
usually protected from the environment and mechanical damage

by a rigid metal tube (Figure 18-14). The fiber-optic bundle is
approximately 0.4 inch in diameter for a 2/3-inch format sensor,
0.3 inch for a 1/2 inch, 0.2 inch for a 1/3 inch, and 0.15 inch for
a 1/4 inch. For the 2/3 inch format, the outside diameter is
about 0.5 inch. It should be noted that the image exiting the
fiber-optic lens is inverted with respect to the image produced
by a standard objective lens. This inversion is corrected by
inverting the camera. The fiber-optic lens speed is between f/4
and f/8 depending on the fiber length—slow in comparison with
the standard, all-lens type pinhole lens.
18.3.7.3 Flexible Fiber
When the most flexibility between the front objective lens and
the camera is required, an alternative to the remote-head CCD
camera is a coherent flexible fiber-optic bundle (Figure 18-14).
The front of the flexible fiber-optic bun-dle has a C mount and
accepts any pinhole, C, or CS
RELAY LENS
OBJECTIVE LENS
CAMERA
SCENE
COHERENT
SENSOR

FIBER OPTIC BUNDLE
6 –12 inches LONG
IMAGE
ON
SENSOR
FIBER
SCENE
OPTIC
IMAGE
OUTPUT
SCENE
FIGURE 18-15 Fiber-optic lens configuration
FLEXIBLE FIBER
BUNDLE ENDS

EPOXIED
RIGID
CONDUIT FIBER #1
12 MICRON FIBERS
LOOSELY HELD IN
PROTECTIVE SHEATH
FIBER #1
FIBER #1
FUSED
RIGID
GLASS
FIBERS
FIBER
ENDS
EPOXIED
FIBER #1
FIGURE 18-16 Fiber bundle construction
mount lens. The rear lens terminates in a male C mount, suitable
for any C or CS mount camera. One advantage the fiber-optic
lens has over a remote head camera is that there is no electrical
connection from the front objective lens to the camera sensor,

which may be important in some applications, for example
environmental protection (from adverse weather, corrosive
environment, or mechan-ical abuse). It can be twisted through
360 with no image degradation. It, too, has spots like the rigid
fiber-optic. The flexible fiber-optic lens has a 180 “twist” built
into it and therefore does not invert the picture. The flexible
fiber-optic bundle individual fibers are fused together only at
the ends, but are free to move in the length between the ends.
18.3.7.4 Image Quality
As shown in Figure 18-16, the fiber-optic bundle is assem-bled
from several hundred thousand individual glass fiber-optic
strands. Although high technology and careful assembly
techniques are used throughout the fiber bundle manufacturing
process to achieve maximum uniform opti-cal transmission,
there are small variations in transmission from one fiber to
another and some broken fibers. The result is that in almost all
fiber-optic systems, the picture obtained is not as “clean” as
that obtained with an “all-lens” pinhole lens. There are some
cosmetic imperfections that look like dust spots (actually non-
or partially transmitting fibers), as well as a geometric pattern
caused by pack-ing the fibers during manufacture. These
imperfections occur because there are several hundred thousand
indi-vidual hair-like fibers comprising the fiber-optic bundle,
and some of them are not transmitting perfectly. For many
461
applications these imperfections do not result in any loss of
picture intelligence, making the lens system adequate for
identification of people, actions, and other informa-tion. Some

fiber-optic lenses have a resolution of 450–500 TV lines,
similar to a high-quality 1/4-, 1/3-, and 1/2-inch camera system.
Figure 18-17 shows two examples of images produced from a
rigid and flexible fiber-optic lens.
The photographs were taken directly from a 9-inch monochrome
monitor using a CCD solid-state camera with resolution of 570
horizontal TV lines. Figure 18-17a shows the typical resolution
and image quality obtainable from a 1-meter, flexible fiber-
optic lens: approximately 450 TV lines horizontal and 350
vertical. The spots are caused by partially transmitting or non-
transmitting fibers. Figure 18-17b shows the same image
obtained with an 8-inch rigid fiber-optic lens. The vignetting at
the corners of the image was caused by the relay lens, not the
fiber bundle. Note the spots and honeycomb pattern in the rigid
fiber-optic monitor picture. The honeycomb is caused by the
fiber-stacking procedure and consequent heat fusing of the rigid
bundle.
18.3.8 Bore-Scope Lenses
The bore-scope lens viewing system is a long thin tube housing
with multiple relay lenses used to view inside objects (such as
safes) or through barriers. Bore-scope sizes range from 12 to 30
inches long, and from 1/8 to 3/8 inch in diameter (Figure 18-
18).
Special mini-bore-scopes are available with 1–2 mm outside
diameters, 2–6 inches long. Bore-scopes are constructed from
stainless-steel tubing and contain an

(3) FIBER: FLEXIBLE: 39 inches LONG
OBJECTIVE LENS: 25 mm FL, F/1.4
RELAY LENS: M = 1:1
OVERALL F/#: 4.0

· FIBER: RIGID: 6 inches LONG
OBJECTIVE LENS: 8 mm FL, F/1.6
RELAY LENS: M = 1:1
OVERALL F/#: 6.0
FIGURE 18-17 Resolution and image quality from fiber-optic
lenses
FIGURE 18-18 Boroscope lens viewing system

CAMERA
CAMERA
RELAY
OPTICS
TOTAL
PROBE TUBE
SCAN
20°
DIAMETER: 9/16"
106°
SCAN MIRROR

WORKING LENGTH: 18"–50"
“all-lens” optical system. The long lengths and all-lens design
mandate that such lenses have very high f-numbers: they are
optically slow. Typical designs have an f-number between f/15
and f/40. By comparison, an f/5 lens trans-mits 16 times more
light than an f/20 lens. The bore-scope must be used with high
levels of lighting or an LLL camera (Chapter 19).
18.4 SPECIAL COVERT CAMERAS
18.4.1 PC-Board Cameras
The miniaturization of 1/6 -, 1/4 -, 1/3 -, and 1/2 -inch CCD and
complimentary metal oxide semiconductor (CMOS) sensors and
camera electronics has generated a new family of small single
and dual printed-circuit (PC) board surveillance cameras. Three
PC-board and housed flat cameras are shown in Figure 18-19.
Figure 18-19a shows a color camera with a CS mount and
automatic-iris option. Figure 18-19b shows a 1/3-inch format
PC-board CCD camera with an 8 mm FL mini-lens and six IR
LEDs for night-time illumination. Other inter-changeable

lenses—3.8, 5.5, and 11 mm FL—are avail-able. Figure 18-19c
shows a compact flat camera sealed in a metal/epoxy case with
pin terminals at the rear. The 1/3-inch format camera has 380-
TV-line resolution and 0.2-fc sensitivity. All cameras are
powered by 12 volts DC.
18.4.2 Remote-Head Cameras
The small size of mini-lenses and CCD and CMOS camera
sensors permits the construction of extremely small covert
lens-sensor heads by remoting the lens and sensor from the
camera electronics via a small electrical cable. The cable link
between the camera head and the camera electronics can vary
from a few inches to 100 feet. Figure 18-20a shows a
monochrome 1/3-inch format CCD remote-head cam-era with an
11-mm FL, f/1.8 lens, and an 18-inch cable connecting the
sensor-lens with the camera electronics.
The camera has a resolution of 450 TV lines and a light
sensitivity of 0.1 fc. Figure 18-20b shows a small color CCD
remote-head camera with a 7.5 mm FL, f/1.6 lens on a 1/2-inch
format sensor. The lens-sensor head is 0.69 inch in diameter ×
225inches long and weighs only 0.64 ounce. The camera has a
resolution of 460 TV lines and a sensitivity of 1.0 fc.
18.5 INFRARED COVERT LIGHTING
Video surveillance augmented with invisible IR covert lighting
can significantly increase the usefulness of covert installations.

Since the covert camera is intended to be hid-den from its
target, if the covert video system can operate in near or total
darkness the person under surveillance will not be aware that he
is under observation. By augmenting the camera system with an
IR light, invisible to the human eye but not to the camera, the
resulting video image can be as good as that obtained under
normal visible daylight conditions. CCD, CMOS, and other LLL
cameras are sensi-tive to this IR radiation and can “see” with
this IR lighting. The amount of IR radiation the camera
responds to and the resulting quality of the picture depends on
the type of IR lamp or LED used, its power level and beam
angle (Chapter 3), and the sensitivity of the camera to the IR
radiation. This last factor depends on whether an IR cut
463

(A) 1/3" COLOR CMOS (B) 1/4" COLOR CCD
(C) HIGH RESOLUTION 1/2" COLOR CCD
FIGURE 18-19 Flat printed circuit PC-board cameras
filter is in place in the camera and on the CCD sensitivity to the
IR energy.
18.5.1 Concealment Means
Light sources that emit both visible and IR light (tungsten,
tungsten-halogen, xenon lamps, and others) can be opti-cally
filtered so that only the IR radiation leaves the source and
irradiates the scene. High-efficiency, low-power LED
semiconductors produce sufficient IR energy to illuminate an
area suitable for covert operation while being invisible to the
eye. Figure 18-22 illustrates the principle and several

techniques of producing IR illumination.
The thermal lamp or LED source emits IR radiation that reflects
off the scene and off objects in it. The lens and camera collect
the reflected IR energy to produce a video image signal. The
IR-emitting source is often con-cealed by installing it behind an
opaque (tinted) plas-tic or one-way (partially aluminized)
window. Another technique is to use a spectral beam-splitting
window that
transmits the invisible IR radiation and blocks the visible
radiation. Another technique is to conceal the IR-emitting
source just as the pinhole lens is concealed, by locating the
source at the focal plane of a pinhole lens and directing the
energy at the same target the pinhole lens is viewing. Usually
the beam from the pinhole lens IR source is made slightly larger
than the FOV of the pinhole lens–camera combination.
Alignment is necessary between the camera and IR source since
the IR beam must illuminate the same scene the pinhole lens is
looking at. When the application is to perform covert
surveillance at short distances and in small rooms (10–15 feet),
a wide-area IR illuminator is used since the alignment is not
critical.
18.5.2 IR Sources
There are numerous commercially available thermal lamp and
LED IR sources for covert surveillance applica-tions. They vary
from short-range, low-power, wide-angle beams to long-range,
high-power, narrow-angle beam

(A) ULTRA SMALL 1/4" DIA.
(B) HIGH RESOLUTION
FIGURE 18-20 Remote head cameras
types. Figure 18-23 illustrates two IR LED and thermal IR

source illuminators.
A single IR LED emits enough IR energy to produce a useful
picture at ranges up to a few feet with a CCD camera. By
stacking many (10 to several 100) LEDs in an array, higher IR
power is directed toward the scene, and a larger area at
distances up to 50–100 feet may be viewed (Figure 18-23a).
Filtered thermal lamp IR sources with power levels up to
several hundred watts can illumi-nate large areas at distances up
to several hundred feet (Figure 18-23b). These are usually used
in outdoor appli-cations where longer ranges are required and
personnel cannot come into close proximity to them. Since the
radi-ation source is not visible to the human eye personnel
should not come in close proximity to them.
18.6 LOW-LIGHT-LEVEL CAMERAS
The camera parameter most critical to the successful view -ing
of a scene under low light level (LLL) conditions with a covert
system is the camera sensor sensitivity. Most monochrome CCD
cameras have sensitivities of approxi-mately 0.2–1 fc (0.1 lux),
which does not result in satisfac-tory CCTV picture quality
under dawn, dusk, nighttime, or poorly lighted indoor
conditions. A few special CCD cameras produce sensitivity of
0.003 fc (0.0003 lux) which
substantially increases its usefulness at low light levels. It also
boasts a resolution of 570 TV lines.
When CCD camera sensitivity is not sufficient and addi-tional
lighting cannot be added, a LLL camera such as an intensified

CCD (ICCD) or intensified SIT (ISIT) must be used (Chapter
19). These light-intensified cameras oper-ate at significantly
lower light levels than the solid-state cameras. The newer ICCD
camera has a sensitivity match-ing that of the prior generation
SIT camera. All this increased sensitivity comes at a cost. Any
intensified cam-era is expensive and should be considered only
for critical security applications.
18.7 IMBEDED COVERT CAMERA CONFIGURATIONS
Video cameras and lenses are concealed in many different
objects and locations including overhead track lighting fixtures,
emergency lighting fixtures, exit signs, tabletop radios, table
lamps, wall or desk clocks, shoulder bags, and attaché cases
(Figure 18-21).
Figure 18-21a shows a popular emergency light that was
modified to house a camera and mini-lens system with the
camera viewing from behind the front bezel. The emer-gency
lighting fixture operates normally, can be tested for operation
periodically, and its operation is in no way affected by the
installation of the camera. The housing has an angled extension
that points the housing downward by about 15 so that the lens
points downward and optimally views the area. Alternatively an
off-axis mini-lens could be used instead of the on-axis mini-lens
to make the cam-era look downward. The lens views through the
smoked (tinted) plastic front window and cannot be seen by an
observer even at close range.
The exit light fixture is another convenient housing for
camouflaging a covert camera system (Figure 18-21b). A wide-
angle mini-lens on a small PC-board camera is all that is
required for this covert camera installation.

A wall-mounted clock is an ideal location for camouflag-ing a
covert camera/lens combination (Figure 18-21c). The lens views
out through one of the black numerals. In this case, the flat
camera (approximately 7/8 inch deep) and mini-lens are
mounted directly behind the numeral 11 on the clock. The
camera uses offset optics (Figure 18-8) so that the camera views
downward at approximately a 15 angle even though the clock
and camera are mounted vertically on the wall.
Figure 18-23d shows a no smoking sign into which a camera
and lens have been installed. The camera views through an
imperceptible hole in the sign. Figure 18-24 shows a ceiling-
mounted sprinkler-head camera. An option to any of these
covert cameras is a wireless RF or microwave transmitter. These
covert camera systems can also be designed using a digital IP
wireless camera and viewed using an Internet browser. The
items into which
465

(A) EMERGENCY LIGHT (B) CLOCK
CAMERA VIEWS THROUGH CAMERA VIEWS THROUGH
BLACK PLASTIC HOLE AT NUMERAL "11"
(C) EXIT SIGN (D) NO SMOKING SIGN
CAMERA VIEWS THROUGH B&W CAMERA VIEW
THROUGH
HOLE IN EITHER ARROW BLACK OPAQUE PLASTIC
FIGURE 18-21 Covert cameras installed in office building
fixtures
covert cameras can be installed are limited only by the
imagination of the user.
18.8 WIRELESS TRANSMISSION

The video signal from the covert camera is sent to the monitor,
VCR, DVR, or over the Internet via RG59/U 75-ohm coaxial
cable, UTP, LAN, WAN, or wireless LAN (WiFi). If a dedicated
telephone-grade line (two-wire) is available, the UTP using a
special line driver and receiver pair provide good transmission
of a real-time video signal over several thousand feet of
continuous telephone wire (Chapter 6). For digital video
transmission CAT-5e cable is used.
Covert video applications often require that the cam-era/lens
system be installed and removed quickly, or that it remain
installed on location for only short periods of time. This may
mean that a wired transmission link (such as coaxial cable or
fiber -optic) cannot be installed
and a wireless transmission link from camera to moni-tor or
recorder is required. This takes the form of a low power radio
frequency (RF) or microwave video transmit-ter mounted near
the video camera. A description of these transmitters is given in
Chapter 6, but those specifically applicable to covert
applications are summarized here. The RF transmitters are less
than 100 milliwatts output and transmit the video images over
ranges from 100 to 2000 feet. In the United States, the FCC
restricts the use of the higher-power transmitters to federal or
government agencies and allows only low-power units for
commercial or industrial use.
Figure 18-25a shows a low-power RF, 100-mw transmit-ter and
receiver operating at 920 MHz that can transmit an excellent
monochrome or color video picture over a distance of a few
hundred feet.
Figure 18-25b shows a 2.4 GHz microwave transmitter that
transmits excellent monochrome and color images over

distances up to a few hundred feet indoors and 2000 feet
outdoors. Using a directional (Yaggi) receiver antenna can
increase the range further. While RF and
(A) LAMP WITH FILTER
TUNGSTEN LAMP
TUNGSTEN HALOGEN LAMP
SPOT OR FLOOD
IR TRANSMITTING
LAMP (PAR)
FILTER

METAL HOUSING
WITH COOLING
FINS (HEAT SINK)
AND CONVECTION
SWIVEL
MOUNT

(B) LED ARRAY
(C) CCTV LENS WITH IR FILTER
LED
LENS
IR
FILTER
IR
FFL OR

SOURCE
ZOOM
CCTV
LENS
IR
PINHOLE
IR
SOURCE
LENS
FILTER

FIGURE 18-22 IR illumination technique
(A) IR LED ARRAY (B) IR THERMAL
FIGURE 18-23 IR source illuminators: IR LED array, IR
thermal lamp

FIGURE 18-24 Sprinkler-head covert camera
microwave transmitters can be used indoors, recognize that
these frequencies cannot pass through metal objects and
therefore the systems should be tested on site, through a steel
building or near other metallic or reinforced con-crete
structures before an installation is made. While the transmitter
may have suitable range under outdoor, unobstructed
conditions, when used indoors or between two points with
obstructions, the only way to determine the useful range of the
link is to put the system into operation. The deleterious effects
most readily observed are: (1) reduction in range, (2) ghost
images (multiple images produced by reflections of the signal
from metallic objects), and (3) unsynchronized pictures (picture
breaks up). Repositioning the transmitter or receiver equipment
often substantially improves or eliminates such problems. Most
microwave systems have a more directional transmit-ting
pattern than RF transmitters. This means the antenna directs the
energy toward the receiver, and therefore align-ment between
transmitter and receiver is more critical. Most microwave
installations are line of sight but the microwave energy can be

reflected off objects in the path between the transmitter and the
receiver to direct the energy to the receiver, at a sacrifice in
range. The higher frequency of operation and directionality
make microwave installation and alignment more critical than
the RF trans-mitters (Chapter 6).
Commercial microwave transmission systems operate in the 2.4
and 5.8 GHz frequency range and do not require FCC licensing
and approval. Other frequencies can only be used by
government agencies and some commercial customers if they
apply to the FCC for a license. One condition in obtaining
approval is to have a frequency search performed to ensure that
the system causes no interference to existing equipment in the
area.
Another line-of-sight system requiring no FCC approval is a
wireless gallium arsenide (GaAs) IR optical trans-mission
system. This light-wave system requires no cable connection
between the transmitter and the receiver and achieves ranges of
hundreds to several thousands
467

(A) RF TRANSMITTER
(B) MICROWAVE
FIGURE 18-25 RF and microwave transmitters for covert video

transmission
of feet (Chapter 6). Its major limitation is the severe reduction
in range under fog or heavy precipitation conditions.
18.9 COVERT CHECKLIST
· Optical speed or f-number is probably the most impor-tant
reason for choosing one pinhole lens over another. The lower
the f-number the better. An f/2 lens trans-mits four times more
light than an f/4. This can mean the difference between using a
standard CCD or CMOS camera and using a LLL ICCD.
· Most pinhole lenses have a FL between 3.8 mm and 22 mm
and are designed for 1/4- and 1/3-inch format cameras. Tables
18-1, 18-3, 18-4, and 18-5 show the FOVs obtained with these
lenses. For example, using these tables or the Lens Finder Kit
(Chapter 4), the FOV seen
with the 11 mm lens on a 1/3-inch camera format at a distance
of 15 feet is an area 6 feet wide by 4.5 feet high displayed on
the monitor. Note that the FOV is independent of the hole size
through which the lens views, providing a hole produces no
tunneling. When viewing through a wall with a wide-angle
pinhole lens or mini-lens (3.8, 5.5, or 8 mm), the lens may
require a cone-shaped hole or an array of small holes to prevent
tunneling (vignetting) of the scene image.
· A short FL lens (3.8 mm) has a wide FOV and low mag-
nification. A long FL lens (25 mm) has a narrow FOV and has

high magnification.
· Medium FL lenses produce FOVs wide enough to see much of
the action and still have enough resolution to identify the
persons or actions in the scene. A short FL lens sees a wide
FOV and objects are not well resolved. Long FL lenses see a
narrow FOV with objects well resolved (clear).
· Under most conditions, the small-barrel, slow-taper pin-hole
lens is easier to install and is the preferred type over the wide-
barrel, fast-taper shape. The user must weigh the pros and cons
of both types.
· The use of a straight or right-angle pinhole lens depends on
the space available behind the barrier for mounting the lens and
camera, and on the pointing direction of the lens.
· The fastest pinhole video system is a mini-lens coupled to the
camera. This is the best choice where the lowest cost and
highest light efficiency are desired.
· A manual-iris lens is sufficient in applications where there are
no large variations in light level, or where the light level can be
controlled. Depending on the camera used, where there is more
than a 50:1 change in light level, an automatic-iris pinhole lens
or an electronically shuttered camera is needed.
· Most applications are solved using an “all-lens” system. In
special cases where a thick barrier exists between a surface and
the camera location, a rigid coherent fiber-optic bundle lens or
bore-scope is used. If sufficient light
is available, an “all-lens” bore-scope type should be used to
obtain the cleanest picture. Another alternative is a remote-head

camera.
· AC power is preferred for permanent covert camera
installations. Either 117 VAC to 12 VDC or 24 VAC wall-
mounted converters are used. Using 12 VDC or 24 VAC is
preferred over 117 VAC since it eliminates any fire or shock
hazard and can be installed by security per-sonnel without
outside help. Since most small cameras operate from 12 VDC, a
117 VAC to 12 VDC converter is most popular. For temporary
installations, 12 VDC battery operation is used, with
rechargeable or non-rechargeable batteries, depending on the
application (Chapter 23).
18.10 SUMMARY
Pinhole lenses are used for surveillance problems that cannot be
solved adequately using standard FFL or zoom lenses. The fast
f-numbers of some of these pinhole lenses make it possible to
provide covert surveillance under nor-mal or dimly lighted
conditions. The small size of the front lens and barrel permit
them to be covertly installed for surveillance applications.
A large variety of mini-lenses and pinhole lenses are available
for use in covert security applications. These lenses have FL
ranges from 3.8 to 22 mm covering FOVs from 12 to 95 .
Variations, including manual- and automatic-iris, standard
pinhole, mini- and off-axis-mini, provide the user with a large
selection.
Equipment is available to provide covert surveillance under
lighted or unlighted conditions. Through the use of IR
illumination, scenes can be viewed in total darkness. Compact
lenses, small and low-power cameras, wireless RF, microwave,
and IR transmission systems make the covert system portable.

The availability of digital IP cameras has now made remote
covert video surveillance a reality. The images from these
cameras can be viewed using an Internet browser from any
Internet access location by anyone having the camera IP
address.

Chapter 11Video SwitchersCONTENT

More Related Content

Similar to Chapter 11Video SwitchersCONTENT (20)

More from EstelaJeffery653 (20)

Recently uploaded (20)

Chapter 11Video SwitchersCONTENT