IoT Based Smart Surveillance and Automation

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3253
IoT Based Smart Surveillance and Automation
Milind R. Rampure1, Kevin P. Rathod2, Kashyap M. Pandya3, Pradnya V.Kamble4
123Students, Department of Electronics and Telecommunication Engineering, K. J. Somaiya Institute of
Engineering and Information Technology, Mumbai, Maharashtra, India .
4Assistant Professor, Department of Electronics and Telecommunication Engineering, K. J. Somaiya Institute of
Engineering and Information Technology, Mumbai, Maharashtra, India.
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - In this paper, we introduce a compact and
complete solution to this problem, all packed in single
Raspberry Pi by implementing an effective object
detection model using Machine Learning and
integrating it with advanced Image Enhancement
technology, opencv to detect any human activity even
under adverse conditions, it will start recording only if
a person or a movement is identified and stop recording
once there is no movement. This technique will improve
the system's efficiency in terms of reducing the storage
needed for saving recordings, and reducing the
processing and searching time in the recordings. Body
temperature of the person is recorded, In case of
increased body temperature readings it will trigger an
alert system to notify the owner.
Key Words: Machine learning, Raspberry pi, OpenCV,
IoT
1. INTRODUCTION
Video Surveillance systems have been an extremely
essential part of security systems installed in all types of
offices, schools, hotels, businesses and even people’s
homes.
A real-time surveillance system is designed and realized
by using Raspberry Pi. The cheap price, low power
consumption, and the availability of open-source software
are major merits of using Raspberry Pi in developing
applications. Thus, it has been widely implemented in the
development of IoT technology, as its reliability and
available information on the Internet.
Recent cameras that record when a movement is detected
save storage rather than recording all the time. These new
features of recording methods in the recent cameras
indeed have affected its prices. Even though recording
videos depending on detecting a movement has enhanced
the efficiency of the surveillance system, it still consumes
large storage. Thus, the need for a camera that records
only humans has emerged. We will interface an IR
temperature sensor and send the email alerts with the
image of the person if the temperature of any particular
person exceeds the set value.
There is a requirement for a savvy and astute framework
that is ought to identify any vindictive action and quickly
report to essential individuals to carry out responsive
actions.
Since the corona outbreak, scanning body temperature has
become the “new normal” to tackle the current Covid19
pandemic. There are many temperature guns available in
the market, but none of them gives any alert or email
notification to higher authorities to take action when the
temperature exceeds a particular limit. A traditional
security camera system provides excellent support to the
security of any home or business but can entirely be
expensive.
2. LITERATURE SURVEY
Initially, many Security Systems such as RFID, OTP-based
systems, and biometrics were used to prevent
unauthorized access. But sooner, people realized merely
preventing unauthorized access was not enough and there
was a need to come up with systems that would aid in
catching the intruder, while also unfolding their motives.
Similar PIR-based systems triggered camera and sensor
networks on motion detection, then uploaded captures to
the cloud server and sent alerts using GSM technology.
All of the above solutions were heavily hardware
dependent which is subjected to high-risk failure. Also,
these solutions usually require a PC along with Wi-Fi DSP
Kit which should be continuously ON for 24 hours, this
leads to higher power consumption and also involves
additional charges for every alert being sent.
3. Our Proposed Method
The smart system is based on Raspberry Pi 3B adopts the
Internet of Things architecture and combines several
sensors and devices for room security to set up a safer
environment. In this proposed system, we introduce a
compact and complete solution to this problem, all packed
in a single Raspberry Pi by implementing an effective
object detection model using Machine Learning and
integrating it with advanced Image Enhancement
technology and OpenCV libraries to detect only human
activity even under adverse conditions. MLX90614 IR

temperature is connected to raspberry pi, which will
measure the body temperature. This will trigger the alert
system and send SMS/Mail. Also, the entire data is being
collected by this system.
We have leveraged the power of IoT and built a
Temperature Monitoring device with Email alerts using
Raspberry Pi, MLX90614, and PiCamera.
An innovative surveillance system is powered by the
Raspberry Pi 3B that provides results with minimal
latencies.
There are 6 steps involved in the working of our model:
1: Body Temperature is Recorded
2: Input Camera Feed
3: Background Subtraction
4: Detection
5: Start Video Recording and Upload to drive
6: Alert using SMTP protocol
Figure 1- Proposed Model
Figure 2- Circuit Diagram
3.1. Contactless Body Temperature Monitoring.
Since the corona outbreak, scanning body temperature has
become the “new normal” to tackle the current Covid19
pandemic. There are many temperature guns available in
the market, but none of them gives any alert or email
notification to higher authorities to take action when the
temperature exceeds a particular limit.
Figure-3: Body temperature measurement
3.2 BACKGROUND SUBTRACTION
Background subtraction (BS) is a common and widely
used technique for generating a foreground mask (namely,
a binary image containing the pixels belonging to moving
objects in the scene) by using static cameras.
The background of our video stream is largely static and
unchanging over consecutive frames of a video. Therefore,

if we can model the background, we monitor it for
substantial changes. If there is a substantial change, we
can detect it. This change normally corresponds to motion
on our video. The input video is taken from a USB
WebCam connected to Raspberry Pi, the very first frame is
considered as background frame which is updated after
particular intervals is compared with the current frame
and if detection is above the threshold value the system
starts video recording and stops when there is no
detection involved.
Figure -4: Background subtraction motion detection
3.3 SMTP & GOOGLE DRIVE API
SMTP is a Simple Mail Transfer Protocol which is an
application layer protocol responsible for email service
using TCP/IP. It provides the ability to send and receive e-
mail messages and is managed by the Internet Engineering
Task Force(IETF).
Google Drive API allows you to create apps that leverage
Google Drive cloud storage. You can develop applications
that integrate with Google Drive, and create robust
functionality in your application using Google Drive API.
4 RESULTS AND DISCUSSIONS
The system uses a three-tier architecture, where the first
tier is the algorithm running to detect motion that keeps
constantly running in a loop triggering certain functions
when deemed necessary.
The second tier is the database and cloud storage which
stores important data in cloud or SD card.
It also includes the e-mail servers that aid in notifying the
user in real-time.
The third and final tier is the part of the system that
experiences user interaction. An IR temperature sensor
sends email alerts with the image of the person if the
temperature of any particular person exceeds the set
value.
Figure 5- Body temperature Email alert
Figure 6 - Video uploaded to drive
3. CONCLUSIONS AND FUTURE SCOPE
The implementation of security cameras typically requires
running cables, as referenced in a past post, and includes
extensive labor and initial installation typically brings with
it a heftier price tag.Our proposed system provides a cost-
effective solution to the problem with a total system cost
of Rs 5600 all packed in a single Raspberry Pi by
implementing an effective object detection model to detect
any human activity even under adverse conditions which
only records when any human is detected, the body
temperature of the person is recorded while entry and exit

from premises, triggering an alert system when increased
body temperature is detected. This technique will improve
the system's efficiency in terms of reducing the storage
needed for saving recordings, and reducing the processing
and searching time in the recordings.
Automatic door lock when any theft is detected in
surveillance mode and when the body temperature of the
user entering the room is high. For further improving this
system we can develop a mesh network with thread
protocol based on IPv6 using 4K Ultra HD cameras to
enable face recognized door open system using IoT
components and also use the other benefits of the face
recognition system. Real-time video stream available
through a monitor.
4. ACKNOWLEDGEMENT
The authors gratefully acknowledge the teachers, lab
incharge and friends for the support and guidance. The
work would have not come to the present shape without
the supervision and help of many people. All the eminent
authors are in debt to Prof. Pradnya Kamble for the
guidance and without her persistent help, the goal of this
project would not have been realized.
5. REFERENCES
[1] “IoT enabled Video Surveillance System using
Raspberry Pi ”, Rahul Muppanagouda Patil, Ram Srinivas ,
Rohith Y, N R Vinay and Pratiba D
[2] “Implementation of Human Detection on Raspberry Pi
for Smart Surveillance” Jahangir Abbas Mohammed,
Member, IEEE, Agniswar Paul, Member, IEEE, Ajay Kumar,
Member, IEEE, Jaideep Cherukuri Member, IEEE
[3] “Real-Time Smart Home Surveillance System of Based
on Raspberry Pi” Yi-Chen Lee, Ching-Min Lee*
[4] “On using AI-based human identification in improving
surveillance system efficiency” Sanyukta Santosh
Pawaskar, Ashwini Mandar Chavan

IoT Based Smart Surveillance and Automation

More Related Content

Similar to IoT Based Smart Surveillance and Automation (20)

More from IRJET Journal (20)

Recently uploaded (20)

IoT Based Smart Surveillance and Automation