EYE TRACKING BASED DRIVER DROWSINESS MONITORING AND WARNING SYSTEM

International Journal of Technical Research and Applications e-ISSN: 2320-8163,
www.ijtra.com Volume 3, Issue 3 (May-June 2015), PP. 190-194
190 | P a g e
EYE TRACKING BASED DRIVER DROWSINESS
MONITORING AND WARNING SYSTEM
Mitharwal Surendra Singh L., Ajgar Bhavana G., Shinde Pooja S., Maske Ashish M.
Department of Electronics and Telecommunication
Institute of Knowledge College of Engineering
Pune, India.
surendra.mitharwal1@gmail.com, bhavana.ajgar@yahoo.com,
pooja.shinde1993.ps@gmail.com, ashishmaske@rediffmail.com
Abstract- This project represents a way of developing an
interface to detect driver drowsiness based on continuously
monitoring eyes and DIP algorithms. Micro sleeps that are short
period of sleeps lasting 2 to 3 seconds are good indicator of
fatigue state. Thus by continuously monitoring the eyes of the
driver by using camera one can detect the sleepy state of driver
and timely warning is issued.
Aim of the project is to develop the hardware which is very
advanced product related to driver safety on the roads using
controller and image processing. This product detects driver
drowsiness and gives warning in form of alarm and as well as
decreases the speed of vehicle.Along with the drowsiness
detection process there is continuous monitoring of the distance
done by the Ultrasonic sensor. The ultrasonic sensor detects the
obstacle and accordingly warns the driver as well as decreases
speed of vehicle.
Keywords—drowsiness, image processing, ultrasonic sensor,
detection, camera, speed.
I. INTRODUCTION
A. Concept of Driver Drowsiness Detection System
The main idea behind this project is to develop a non-
intrusive system which can detect drowsiness of the driver and
issue a timely warning. An accident involving driver
drowsiness has a high fatality rate because the perception,
recognition, and vehicle control abilities reduces sharply while
falling asleep. Driver drowsiness detection technologies can
reduce the risk of a catastrophic accident by warning the
driver of his/her drowsiness. The development of technologies
for preventing drowsiness at the wheel is a major challenge in
the field of accident avoidance systems. Preventing
drowsiness during driving requires a method for accurately
detecting a decline in driver alertness and a method for
alerting and refreshing the driver. Micro sleeps that are short
period of sleeps lasting 2 to 3 seconds are good indicator of
fatigue state. Thus by constantlyobserving the eyes of the
driver one can detect the sleepy state of driver early enough to
avoid accident.
The current project is a prototype of the model that can be
proposed or built to detect the drowsiness of the driver and
save the life of the driver. The components used in the project
are microcontroller Atmega AT89S52 which plays the vital
role in the functioning of the project, the other components are
L293D the motor driver IC used to drive the DC motor, DC
motor used as a vehicle, MAX 232 IC is used to convert the
voltage levels to TTL and vice versa. The camera is being
utilized to capture the images of the driver’s eye to detect the
state of drowsiness. For powering up the whole system we are
using transformer based regulated supply.
B. Motivation
Drowsiness is one of the main issues in road accidents.
The fatality rate due drowsiness is higher. An accident
involving driver drowsiness has a high fatality rate because the
observation, acknowledgement and vehicle control abilities
reduce sharply while falling asleep. The growing number of
accident fatalities in world in recent years has become a
problem of serious concern for the society, so accidents must
be prevented before they happen and this thing lies with the
driver. Accidents usually lead both economic as well as social
loss to the society. If accidents are prevented we can save
many lives and along with that the environment is also
preserved. The earlier systems used were costly and heavy.
Preventing accidents caused by drowsiness requires a
technique for detecting sleepiness in a driver and a technique
for arousing the driver from that sleepy condition. The project
describes a system that uses an image processing technique to
recognize the open or closed state of the driver's eyes as a way
of detecting drowsiness at the wheel. The electrodes that were
attached to the body of the driver were both annoying and
tiring for the driver. The driver was also unable to concentrate
on driving due to this. Along with that electrodes needed time
to time replacement. The other systems that were implemented
to detect the drowsiness of the driver used GSR, ECG, EMG.
These also needed the attachment of the sensors to the body of
the driver. Thus detecting drowsiness and alerting the driver
helps in preventing accidents and lives can be saved.
C. Objective
We are searching for a system which will automatically
detect drowsiness based on driver’s performance. Current
systems used to detect drowsiness are slow and consume more
time to give the output and warn the driver accordingly. The
systems utilized are also very costly and are implemented only
in high class or very expensive vehicles. Due to which the
normal vehicles lack such systems and even the safety is low.
This project aims at building a safety system for vehicles
which can be implemented and build at low cost. This will in
turn provide both security as well as protection to the vehicle
as well as the driver driving the vehicle. If such systems come
into existence lots of lives can be saved.
II. CURRENT SYSTEMS USED IN DROWSINESS
DETECTION
A. Vision- Based Visual Cues Extraction
Fatigue monitoring starts with extracting visual
parameters typically characterizing a person's level of
vigilance. This is accomplished via a vision system in
computer. In current section, we discuss the computer vision

191 | P a g e
system we developed to achieve this goal. Figure below
provides an overview of our visual cues extraction system for
driver fatigue monitoring. The system consists of two
cameras: one wide angle camera focusing on the face and
another narrow angle camera focusing on the eyes. The wide
angles camera monitors head movement and facial expression
while the narrow angle camera monitors eyelid and gaze
movements.
Fig. Vision based visual cues extraction
B. Controller Based Drowsiness Detection
The proposed integrated system architecture is depicted in
below figure. As seen the driver monitoring system outputs
are used as an input for the controller and the control
commands are augmented with driver’s commands for the
vehicle control in adverse conditions. A diagnosis system
constantly decides based on the risk level given by the driver
monitoring system, the vehicle and the controller conditions.
Using these three information channels, the diagnosis system
can activate or deactivate the controllers according to the
particular situation. In the following sub-sections, monitoring
and controller systems are detailed. The diagnosis system
structure requires several controllers and scenarios to be
considered in a more extensive way. Therefore, in this case
study, the augmentation of the controller and driver for
controlling the vehicle and the role of the monitoring system
are the focus. The system is a specific solution for accident
avoidance in the case of drowsiness/sleepiness with the assist
of an adaptive robust lateral controller with speed regulation
as an auxiliary system.
Fig. Controller based drowsiness detection system
C. Detecting The Physiological Response Of Driver
In this method the driver drowsiness is monitored by
planting various sensors on the driver’s body. The sensors
used are EKG (Electrocardiogram), GSR (Galvanic Skin
Response), and EMG (Electromyogram). The outputs received
from these sensors are used in deciding the alertness of the
driver. All these sensors are to be continuously attached to the
body of driver. The main drawback of this system is the aging
of the sensor response.
Fig. Sensors measuring the physical response
III. EYE TRACKING BASED DRIVER DROWSINESS
MONITORING AND WARNING SYSTEM
Fig. Block diagram of overall system
The above figure shows the block diagram of the overall
project. The images taken from the camera are provided to the
PC unit for processing. The result from the PC unit is
displayed on the display screen. The ultrasonic sensor keeps
monitoring the distance and along if obstacle detected it alerts
the driver. Thus this is a safety system which helps in
preventing accident.
IV. FUNCTIONALITY OF COMPONENTS
A. Operations
All the system and hardware components are initialized.
We take the video input via camera and the GUI input data is
given to the system. Eyes of the driver are continuously
scanned. If found drowsy the alarm system is activated and the
speed of the vehicle is gradually reduced. The driver manually
stops the alarm and the same process is carried further. Along
with the drowsiness detection process there is continuous

192 | P a g e
monitoring of the distance done by the Ultrasonic sensor. The
ultrasonic sensor detects the obstacle and accordingly warns
the driver.
B. Software Input
Fig. GUI used to take the input via camera
Using MATLAB graphical user interface is created.
Graphical user interface enables us to create an interface
between the hardware and software. It consists of various push
buttons like open eye template, close eye template, initialize
camera. It also consist of the edit button to edit the number of
frames. With the help of handle we can move from one object
to another within that GUI.
Fig. Open eye template input
Continuously eyes are scanned by the camera and video
preview is taken after the camera is initialized. After clicking
the pushbutton of open eye template camera takes the preview
and image is cropped later to fit into the axes of open eye
template.
Fig. Close eye template
Continuously eyes are scanned by the camera and video
preview is taken after the camera is initialized. After clicking
the pushbutton of close eye template camera takes the preview
and image is cropped later to fit into the axes of close eye
template.
C. Software Output
Fig. Driver is found to be drowsy
Continuous scanning of eyes is done to check whether
the driver is drowsy or not. And this is decided by the open
eye and close template. For a given number of iterations if
consequently three frames of close eye template are found by
the system, the driver is said to be drowsy. Command window
in MATLAB displays the number of iterations been carried
out and display ‘Drowsy’ if found drowsy.
D. Hardware Input
Fig. Hardware input
When the power supply is provided to the hardware
the LCD, DC Motor and buzzer are powered on.

193 | P a g e
E. Hardware Outputs
Fig. No obstacle detected
If no obstacle is detected by the ultrasonic sensor
LCD displays the message “No Obstacle”. Hence alarm
system is also not activated and DC motor continues to rotate
with the same speed, i.e. its speed does not change.
Fig. Obstacle detected
If obstacle is detected by the ultrasonic sensor, LCD
displays the message “Obstacle detect” as the distance at
which the obstacle is detected is below the threshold value
given in the program and also displays the distance of the
obstacle detected. Along with the LCD display, driver is
alerted in the form of alarm and the speed of DC motor is
reduced gradually.
Fig. Obstacle detected at greater distance than the desired
distance
As the distance at which the obstacle is detected is above the
threshold value, which is given in program, LCD does not
display the message “Obstacle detected” and hence alarm
system is not activated. Speed of DC motor is not changed it
continues to rotate at the same speed. But at what distance the
obstacle is detected by the ultrasonic sensor is displayed on
the LCD display.
F. Results
The results given by the system are more accurate and the
output is given within 6to7 seconds and the accuracy is greater
as compared to other processes.
G. Comparison Between Different Techniques
As mentioned above we have seen that many of the
methods are being implemented to detect the drowsiness and
the level of fatigue in the driver. The technique bears results
which are very much effective. In the detecting the
physiological response of driver method wherein we use ECG,
EKG to detect the drowsiness of the driver in this scenario we
come to know that various instruments are mounted in the
vehicle as well as on the drivers body. This method tires the
driver to a greater extend because the driver has to be seated in
a constant position for the duration he drives the vehicle. In
case if any accident is caused it therefore becomes very risky
for the driver. In such cases the driver may even lose his life.
In the vision based system the cameras are mounted on the
dashboard which continuously keeps on scanning the face and
produce the results as programmed. The system consists of
two cameras: one wide angle camera focusing on the face and
another narrow angle camera focusing on the eyes. The wide
angles camera monitors head movement and facial expression
while the narrow angle camera monitors eyelid and gaze
movements. This method also produces very positive result.
Thus the response produced by all the systems is not accurate
therefore it is very much necessary to bring into picture such a
system which will help in detecting drowsiness at faster rate
and produce efficient results. This will in return help in saving
the driver’s life.The development of technologies for

194 | P a g e
preventing drowsiness at the wheel is a major challenge in the
field of accident avoiding systems. Drowsiness prevention
during driving requires a method for accurately detecting a
decline in driver’s alertness and a method for alerting and
informing the driver. As compared to all the above methods
the outputs given by the “Eye Tracking Based Driver
Drowsiness Monitoring and Warning System” yields better
results and time taken is also very less. The efficiency and
accuracy of the model is greater as it takes very less time to
give the output. The output is produced within few couple of
seconds.
V. ADVANTAGES AND APPLICATIONS
A. Advantages
 Component establishes interface with other drivers
very easily.
 Life of the driver can be saved by alerting him using
the alarm system.
 Speed of the vehicle can be controlled by controlling
the fuel supply of the car.
 Ultrasonic sensors are utilized for monitoring the
distance and alerting the driver accordingly.
 Traffic management can be maintained by reducing
accidents.
 Practically applicable.
B. Applications
 Can be used in vehicles to detect drowsiness.
The drowsiness detection system can be used to
detect the drowsy state of the driver. If found drowsy
the alarm system gets activated and the driver is
alerted. In the same way the ultrasonic sensor keeps
on monitoring the distance and if the distance
between the vehicle and obstacle detected is less the
warning is given to the driver.
 It can be implemented in factories to keep a check on
the machine operator.
In factories such systems can be used to keep a check
on the machine operator.
This system can also be used for the safety of the
machine operator.
 Can be used in railway engines for the safety of the
driver.
The railway drivers mostly have to travel long
distance. This system can be used to alert the driver
of the drowsy state. And the ultrasonic sensor warns
them about the obstacles on the track.
 Also it can be used for military purpose.
VI. CONCLUSION
 A non-intrusive method of drowsiness detection is
possible.
 Same method may be applied to detection of fatigue
or other related driver performance.
 By monitoring the eyes using camera and using this
new algorithm we can detect symptoms of driver
fatigue early enough to avoid an accident.
 So this project will be helpful in detecting driver
fatigue in advance and will gave a warning output in
form of sound.
 Also an ultrasonic sensor which continuously
monitors the distance helps in avoiding accidents.
VII. ACKNOWLEDGEMENT
It is our pleasure to get this opportunity to thank our
beloved and respected guide, Prof. A. M. MASKE who
imparted valuable basic knowledge of Electronics. We
sincerely thank him for consistent guidance, inspiration
and sympathetic attitude throughout the paper work. We
express our sincere thanks to all our staff and colleagues
who have helped us directly or indirectly in completing
this project work. Given an opportunity we also like to
thank our parents for wishes and moral support during the
project work and all concerned for helping and
encouraging us. We are grateful for the many useful
comments and suggestions provided by everyone, which
have resulted significant improvements in paper work.
REFERENCES
[1] “Development of drowsiness detection system” Vehicle
Research Laboratory, Nissan Center NISSAN MOTOR
Co., Ltd. Hiroshi Ueno Masayuki Kaneda Masataka
Tsukino
[2] “Smartcard: Detecting Driver Stress “Massachusetts
Institute of Technology Media Laboratory Jennifer Healey
and Rosalind Picard
[3] “Real Time and Non-intrusive Driver Fatigue Monitoring
“Department of Electrical, Computer, and Systems
Engineering2004 IEEE Intelligent Transportation System
Conference Washington, D.C., USA, October
36.2W4TuC4.4 Zhiwei Zhu* Qiang Ji**
[4] “Face and eye tracking algorithm based on digital image
processing “Department of Electrical Engineering,
Universidad de Chile AV. Tupper 2007. Claudio A. Perez,
Alvaro Palma, Carlos A. Holzmann and Christian Pera
[5] “EEG-Based Drowsiness Estimation for Safety Driving
Using Independent Component Analysis”2726 IEEE
transactions on circuits and systems –I: Regular papers,
VOL. 52, NO. 12, December2005. Sheng- Fu Liang, Wen-
Hung Chao, Yu-Jie Chen, and Tzyy-Ping Jung
[6] “Active Accident Avoidance Case Study: Integrating
Drowsiness Monitoring System with Lateral Control and
Speed Regulation in Passenger Vehicles” Proceedings of
the 2008 IEEE International Conference on Vehicular
Electronics and SafetyColumbus, OH, USA. September 22-
24, 2008. Pınar Boyra, John H. L. Hansen
[7] “Communication, Control and Automation Vision-based
Vehicle Detection in the Nighttime” Department of
Electrical Engineering National Chin-Yi University of
Technology Taichung, Taiwan 2010 International
Symposium on Computer. Ying-Che Kuo Hsuan-Wen
Chen

EYE TRACKING BASED DRIVER DROWSINESS MONITORING AND WARNING SYSTEM

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to EYE TRACKING BASED DRIVER DROWSINESS MONITORING AND WARNING SYSTEM (20)

More from International Journal of Technical Research & Application (20)

Recently uploaded (20)

EYE TRACKING BASED DRIVER DROWSINESS MONITORING AND WARNING SYSTEM