International

OPEN

Journal

ACCESS

Of Modern Engineering Research (IJMER)

Driver Fatigue Monitoring System Using Eye Closure

Miss.G.Victoreia1, Miss.D.Yazhini2, Miss.G.Parameswari3, Mr. E.Gurumoorthi4,
Mrs. G.Vijayabarathy5
1,2,3

MCA student, Sri Manakula Vinayagar Engineering College, Puducherry

Assistant Professor, Sri Manakula Vinayagar Engineering College, Pondicherry
5
Assistant Professor, Perunthalaivar kamarajar arts college, Puducherry

A Abstract: Now-a-days so many road accidents occur due to driver distraction while he is driving.
Those accidents are broadly depends upon wide range of driver state such as drowsy state, alcoholic
state, depressed state etc. Even driver distraction and conversation with passengers during driving
can lead in major problems. To address the problem we propose a Driver fatigue Monitoring and
warning system based on eye-tracking, which is consider as active safety system. This system is useful
and helpful for drivers to be alert while driving. Eye tracking is one of the major technologies for
future driver system since human eyes contains much information. Sleepiness reduces reaction time of
safe driving. The driver distraction is measured by the person eye closure rate for certain period while
driving. It is implemented by comparing the image extracted from video and the video that is currently
performing. The percentage of eyes is compared from both the frames, if the driver is suspected to be
sleeping then a warning alarm is given to alert the driver.
Keywords: Driver fatigue, Eye Tracking, Drowsiness detection.

I. INTRODUCTION
Now-a-days the society faces more number of accident rates. The Driver fatigue problems have
become important factors for causing road accidents. Driver fatigue is a major cause for four wheeler accidents
since the drivers are unable to take sudden decision while they are sleepy. This kind of accident can be
predicted and given solutions. The fatigue behaviors are predicted whether the drivers stay awake or not and
make a warning when they begin to fall asleep. The system can detect the changes in eyes by measuring
closed/opened eye state. This technology is used to track the eyes of the driver and alert them.
This system capture the eye sight from the face and it measures the eye rate by closed/opened state. In
addition to detect the drowsiness of the driver, if the eyelid is closed for more than 8-10 seconds then it is the
indicator of fatigue state and by that time the system alerts the driver. The system detects driver fatigue status
by measuring the proportion of eyes closed in a certain period of time and the continued closure time.
The Algorithm is used to detect the drowsiness of the driver is PERCLOS. It is abbreviated as
(PERcentage ClOSure) is a measure of driver alertness. This algorithm is helpful in measuring the person
eyelid closure. It measures the slow eyelid closures rather than blinks, and is the measurement of choice for
drowsiness detection. This Algorithm is been considered as the Standard for drowsiness detection. In this it
measures the percentage of eyelid closure over the pupil and reflects the slow eyelid closure with the movement
of eye blinking.
PERCLOS is the Real-Time measures of alertness for drowsiness detecting system. It measure by
capturing the face and focused on the eyes and detects the pupil from the eyes. Eyes has the various parameters
of measures and colors the white area will not be detected, the Retina of the eyes will detected from the whole
eyes and it has a various point and individual color to be measured from the retina.
In this system the face and the eyes are localized by comparing the image and the video input, but the
camera will be focused on the eyes. When the camera first detects the face, the image is saved to a file for the
usage of comparing. The points determine whether the person is feeling sleepy or not. Once the person closed
his eyes or moves outside of the frame, the warning system alert the driver from drowsiness or distraction or it
is helpful to other passengers to be alert and wake the driver from fast asleep.

| IJMER | ISSN: 22496645 |

www.ijmer.com

| Vol. 4 | Iss.11| Nov. 2014 | 26|

Driver Fatigue Monitoring System Using Eye Closure

II. EXISTING SYSTEM
The existing system evaluate whether changes in the eye-steering correlation that can indicate
distraction. The auto-correlation and cross-correlation of horizontal eye position and steering wheel angle
show that eye movements associated with road scanning procedure a low eye steering correlation. The eye
steering correlation will control the relationship on a straight road. The straight road led to a low correlation
between the steering movement and eye glances. In this system it is aim to detect the driver distraction based
on visual behavior or the performance of the driver so for this purpose it is used to define the relationship
between the visual behavior and the vehicle control.
This system evaluates the eye-steering correlation associated with the straight road with the
assumption that it might show a qualitatively and quantitatively different relationship compared with curvy
road and that it might be sensitive to distraction. Here in the visual behavior and vehicle control relationship
reflects a fundamental perception-control mechanism which plays a major role in driving and a strong eyesteering correlation associated with this process has been observed on curvy roads.
1. Driver Distraction Detection Algorithm
Driver Distraction Detection Algorithm is used in earlier system which is the combination of glance
behavior and vehicle control. Two types of distraction are focused:
a) Cognitive

b) Visual

Cognitive described as Mind-off-road and Visual has been described as Eye-off-road. Cognitive
tasks tend to reduce the lane position variance and visual tasks tend to increase the variance. Although both
can determine the driving performance, their effects are different.
2. Disadvantage in Cognitive distraction detection
Although it improves drive performance it degrades. Longitudinal control with hazards perception. It
is complex because the mechanism involved in cognitive distraction has not been precisely described.
There are two approach has given in distraction detection algorithm. First approach focuses
distraction detection systems on visual behavior and another approach focuses on driver behavior such as
speed, lane position and steering. Considering a combination of both glance behavior and vehicle control
might increase algorithm sensitivity to detecting different types of inattention.
This paper presents a new approach to algorithm design that focuses on the relationship between the
eye movement and steering wheel movement.
3. Eye-steering coordination
Two modeling approaches are there to describe and predict the allocation of drivers visual attention.
They are:
a) Models of visual search:
This model describes how people locate targets in the environment to complete the task.
b) Models of visual sampling:
This model describes a scanning pattern of particular areas to find relevant information
4. Problems related with existing system
For identifying the drivers fatigue level, detecting the eye state is important more than detecting the eyesteering correlation.
Correlation between the eye-steering is complex and might produce wrong results undetermined cases.
Steering movements consists of large calculations and graphs and it is difficult to merge with eye
movements in all type of roads. Lane determination is not necessary for driver distraction.

III. PROPOSED SYSTEM

In the proposed system, the driver fatigue and distraction is detected only by processing of eye region.
The main symptoms of driver fatigue and distraction appear in the drivers eyes because of sleeping while
driving. Nowadays, there are many fatigue detection methods and the best is capturing the eyes in real time
using web camera to detect the physical responses in eyes. Moreover, the processing of the eye region instead
of the processing of the face region has less computational complexity.

| IJMER | ISSN: 22496645 |

www.ijmer.com

| Vol. 4 | Iss.11| Nov. 2014 | 27|

Driver Fatigue Monitoring System Using Eye Closure

3.1. Researches and Statistics
India has the highest proportion of deaths due to road traffic accidents in South East Asia. India
accounts for as high as 6 per cent of the worlds RTAs, although it has 1 per cent of the worlds vehicles. The
RTA (Road Traffic Accident) rate of 35 per 1,000 vehicles in India is one of the highest in the world and so is
the RTA fatality rate of 25.3 per 10,000 vehicles.
Recent years have been witnessing an increasing amount of traffic on the roads leading to increased
risks for road traffic accidents to occur. Evidence from developed and especially developing countries indicates
that road traffic accidents are on the rise and are found to be fifth among important causes of deaths globally,
leading to a significant proportion of injuries, deaths and disabilities in the population.
3.2. Process of proposed system
In the existing system, it detects the drivers fatigue using eye-steering correlation monitoring. Thus it
uses the Driver distraction detection algorithm which concentrates on both visual and cognitive behaviors.
We consider that for detecting the fatigue of the driver only eye state or behavior is sufficient and provides a
easy way to use in all the vehicles.
The detection of cognitive method is tedious and it may go wrong sometimes which may provide
wrong results. The steering movement consists of large calculations and graphs and it is difficult to merge with
eye movements in all type of roads.
So in the proposed system we detect only the eye region and monitor the eyelid closure using
PRECLOS (PERcentage CLOSure). A commonly used measure of eye closure duration is PERCLOS
percentage of time a persons eyes are 80-100% closed over a period of time. When the eye is not detected the
person is considered to be distracted or in the sleepy state.
Fig 1: Overall process of Proposed approach
Face recognition

Acquire video using open

CV
Extract image from video

Save image in file

Eye detection

Right

Left

Eyes open?

Eyes detected

Eyes not detected

Warning alarm

| IJMER | ISSN: 22496645 |

www.ijmer.com

| Vol. 4 | Iss.11| Nov. 2014 | 28|

Driver Fatigue Monitoring System Using Eye Closure

3.3 PERCLOS algorithm
A real time driver attention monitoring system is proposed using PERCLOS algorithm which is said
to be the estimator of fatigue level. The main purpose of this algorithm is eye detection for a general driver
under varying illumination conditions and finding class of the detected eye.
When the driver sits in the driver seat and start the vehicle, the webcam started to detect the face
automatically refer figure 1.1. Unless the web camera is on, the vehicle cannot be started.
Fig 1.1: Face detection and extraction of eyes

When the eye region is detected from the video a photograph is taken and stored in the file to identify
the persons eye as shown in the figure 1.2. Then video is captured throughout the journey and detects the eye
movements to find the state of eyes.
Fig 1.2: Add face module

The drivers eye in the video is compared with the stored image and when the drivers eye is located
in the video as detected, the eyes is said to be in active state (refer figure 1.3).
Fig 1.3: Eyes detected

| IJMER | ISSN: 22496645 |

www.ijmer.com

| Vol. 4 | Iss.11| Nov. 2014 | 29|

Driver Fatigue Monitoring System Using Eye Closure

The drivers eye in the video is compared with the stored image and when the drivers eye is not
located in the video then he is said to be distracted and a message is given as shown in figure 1.4. When the
eyes is found to be in the closed state automatically an alarm signal is given to awaken the driver. This system
not only detects the drowsiness of the driver it also detects the driver distraction such as moving, looking
around etc. The timely warning alerts the driver as well as the passengers to be aware of the happening.
Fig 1.4: Eyes not detected

3.4 Advantages
Accidents can be avoided by alerting the drivers distraction and drowsiness using warning signals.
Comparing to the driver detection algorithm this PERCLOS system reduces the time complexity.
It is used to track the alertness of the driver throughout the journey.
The warning alarm alerts the driver as well as the passengers to be conscious about the drivers behavior.
This system typically behaves as a user-friendly application.

IV. CONCLUSION
This system will analyze the state of drivers eye state whether it is closed or opened. If it is closed
then produce the warning alarm to awaken the driver and prevent the vehicle from accident. By monitoring the
drivers eye state using the PRECLOS algorithm by web camera the system can detect symptoms of driver
fatigue early enough to avoid an accident and save people life by detecting the distraction an drowsiness. So
this project is useful and easy to detect the drivers fatigue level and give the warning output in the form of
sound to alert the driver as well as the passengers.

REFERENCES
Journal Papers:
[1]
[2]
[3]
[4]
[5]

[6]

[7]
[8]

Lora yekhshatyan, John D.Lee, Changes in correlation between eye and sterring movements indicate driver
distraction IEEE trans. Intell. Transp. Syst.,vol .14, no.1,pp. 136-145,Mar 2013.
Mohamad-Hoseyn Sigari, Mahmood Fathy, Mohsen Soryani, A Driver Face Monitoring System for Fatigue and
Distraction Detection International Journal of Vehicular Technology, Volume 2013 (2013), Article ID 263983.
Boon-Giin Lee , Wan-Young Chung, Driver Alertness Monitoring Using Fusion of Facial Features and BioSignals IEEE sensor journal., vol. 12, no. 7, pp. 2416-2422, july. 2012.
Yanchao Dong, Zhencheng Hu ,Uchimura K, Murayama. N Driver Inattention Monitoring system for Intelligent
Vehicles IEEE trans. Intell.Transp. Syst., vol .12, Issue . 2, no.1,pp. 596-614, 2011.
M.Omidyeganesh,Javadtalab.A,Shirmohammadhi.S Intelligent driver drowsiness detection through fusion of
yawning and eye closure Virtual Environments Human-Computer Interfaces and Measurement Systems
(VECIMS), 2011 IEEE International Conference on Sep-2011
I.Daza, N. Hernandez, L. Bergasa, I. Parra, J. Yebes, M. Gavilan, R. Quintero, D. Llorca, and M. Sotelo,
"Drowsiness monitoring based on driver and driving data fusion" in Intelligent Transportation Systems (ITSC),
2011 14th International IEEE Conference on, oct. 2011, pp. 1199-1204.
M.Wollmer, C.Blaschke, T.Schindl, B.Schuller, B.farber, S.Mayer, and B.Treffich. Online Driver Distraction
Detection using short-term Memory IEEE trans.intell. transp.syst., vol.12, no.2, pp.574-582, jun 2011.
Hardeep Singh, J,S Bhatia, Jasbir Kaur, Eye Tracking based driver fatigue monitoring and Warning System
Power electronics (IICPE),2010 India International Conference on Jan-2011, ISBN:978-1-4244-7883-5, INSPEC
Accession no: 11873780, publisher: IEEE.

| IJMER | ISSN: 22496645 |

www.ijmer.com

| Vol. 4 | Iss.11| Nov. 2014 | 30|

Driver Fatigue Monitoring System Using Eye Closure

[9]
[10]

[11]
[12]
[13]
[14]

[15]

Marco Javier Flores, Jose Maria Armingol, Arturo de la Escalera, Real Time Warning System for Driver
Drowsiness Detection Using Visual Information Vol. 59, Issue. 2, pp. 103-125, August 2010.
L. Yunqi, Y. Meiling, S. Xiaobing, L.Xiuxia, O. Jiangfan, Recognition Of Eye States in Real Time Video in
Proceedings of the International Conference on Computer Engineering and Technology, IEEE Computer Society,
Vol. 10, No.105, 2009, pp. 554-559.
M. Wilson, M. Chattington, and D. E. Marple-Horvat, Eye movements drive steering: Reduced eye movement
distribution impairs steering and driving performance J. Motor Behav., vol. 40, no. 3, pp. 190202, May 2008.
E. Rogado, J.L. Garca, R. Barea, L.M. Bergasa, E. Lpez, Driver Fatigue Detection System in the Proceedings
of the IEEE International Conference on Robotics and Biomimetics Bangkok, Thailand, 2009, pp.1105-1110.
M.H Sigari, Driver Hypo-Vigilance Detection based on Eyelid Behavior in Proceedings of the Seventh
International Conference on Advances in Pattern Recognition, IEEE Computer Society, 2009. pp. 426-429.
Dr. P.R Bajaj, M.S Devi, Driver Fatigue Detection Based on Eye Tracking in Proceedings of the First
International Conference on Emerging Trends in Engineering and Technology, IEEE Computer Society, 2008, pp.
649-652.
Branzan, A., Widsten, B., Wang, T., Lan, J., Mah, J. A computer vision-based system for real-time detection of
sleep onset in fatigued drivers In: IEEE Intelligent Vehicles Symposium, pp. 2530 (2008).

| IJMER | ISSN: 22496645 |

www.ijmer.com

| Vol. 4 | Iss.11| Nov. 2014 | 31|