See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/314466114

Driver Drowsiness Detection System

Article  in  IOSR Journal of VLSI and Signal processing · January 2014

DOI: 10.9790/4200-04113437

CITATIONS READS

6 1,347

2 authors, including:

Smitha P S
Sree Chitra Thirunal College Of Engineering
13 PUBLICATIONS   29 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Smitha P S on 29 January 2020.

The user has requested enhancement of the downloaded file.

IOSR Journal of VLSI and Signal Processing (IOSR-JVSP)
Volume 4, Issue 1, Ver. I (Jan. 2014), PP 34-37
e-ISSN: 2319 – 4200, p-ISSN No. : 2319 – 4197
www.iosrjournals.org

Driver Drowsiness Detection System

Jayasenan J. S1, Mrs. Smitha P. S2
1Student, M.Tech Signal Processing, Department of electronics and communication engineering,Sree Chithra
Thirunal College of Engineering and Technology, Pappanamcode,Trivandrum
2Assistant Professor, Department of electronics and communication engineering,Sree Chithra Thirunal College
of Engineering and Technology, Pappanamcode,Trivandrum

Abstract: The advancement of technologies for averting drowsiness at the wheel is a key dilemma in
the field of accident p r e v e n t i o n systems. Preventing drowsiness during driving necessitates a scheme
for precisely perceiving deterioration in driver’s vigilance and a means for alerting and reviving the
driver. Drowsy Driver Detection System has been developed, using a non-intrusive machine vision based
concepts. This system offers a method for driver eye detection, which could be used for observing a
driver’s fatigue level while he/she is maneuvering a vehicle. The system uses a small monochrome
security camera that points directly towards the driver’s face and monitors the driver’s eyes in order to
detect fatigue. In such a case when fatigue is detected, a warning signal is issued to alert the driver. This
paper describes the method that has been proposed for finding the eyes, and also for determining if the
eyes are open or closed. The system deals with using information obtained for the contour of the image to find
the edges of the face, which narrows the area of where the eyes may exist. Once the face area is found, the eyes
are found by computing the contour. Once the eyes are located, converting to binary and by counting the
bright pixels in the eye area determine whether the eyes are open or closed. The system is also able to detect
when the eyes cannot be found, and works under reasonable lighting conditions.

I. Introduction
Sleep-divested driving is the process of functioning a motor vehicle while being
continuously messed up by lack of sleep. It is a key trigger of many motor vehicle catastrophes, and it can
blight the human brain a s m u c h a s alcohol c an. Drowsiness i n f l u e n c e mental alertness,
intensifying the risk of can be triggered by numerous causes such as fatigue, lack of sleep and the use of
medication. In addition, an alternative reason that can b e c o n t e m p l a t e d i s the dullness o f
driving on expressways or in jam-packed traffic. The p e r s i s t e n t construction of highways
and upgrading of vehicle operation have made it feasible for drivers to benefit from enjoyable a n d r e s t
f u l d r i v i n g . Conversely, d r i v e r s a r e more pertinent to maneuvers their vehicles under tedious driving
circumstances.
The development of a scheme for distinguishing or avoiding drowsiness at the controls is a major
contest in the domain of accident averting practices Probable procedures for perceiving drowsiness
in drivers can be generally classified into four major classes: sensing of physiological characteristics, sensing
of driver operation, sensing of vehicle response, monitoring the response of driver. Among these different
methods, the best detection precision is accomplished with methods that are supported on
physiological prodigies. Physiological measurements can be executed in two approaches: determining
variations in physiological signals, such as brain waves, heart rate etc using electrodes; and measuring
physical changes such as sagging posture, leaning of the driver‟s head and the open/closed states of the
eyes. An EEG-based system developed by [1] was able to detect fatigue with an error percentage of around
10%, which the authors allege to be very steadfast. Related EEG centered analyses were also conducted by
[2], [3] and [4]. The foremost disadvantage with these systems is that electrodes have to be appended to
driver, producing frustration to him.
The vehicle restraining schemes that might be used for getting the impression of
driving operation comprises the steering wheel, accelerator, brake pedal or transmission shift lever.
The vehicle performance perceived might be the vehicle speed, lateral acceleration, yaw rate or lateral
displacement. Since these practices permit noncontact recognition of drowsiness, they do not provide the
driver any impression of uneasiness. On the damaging p a r t , t h e y ar e f o c u s e d to s e v e r a l p r e c i n c t s
depending on the vehicle category and driving circumstances. It would also be essential to
formulate distinct detection logic for each type of vehicle. Still a different dilemma with this tactic is that
detection would not be feasible at minimal pace.
An analysis of the human eyes under a provision of d i m i n i s h e d vigilance specifies that

www.iosrjournals.org 34 | Page
 Driver Drowsiness Detection System

the eyes are constricted than in a wide-awake status and that there are occasions when the eyes actually close.

Fig 1: No of times eyes close and alertness level

Figure1 offers experimental findings demonstrating the vigilance level and the number of times
the driver's eyes closed for two or more seconds while driving on a test course. Good correlation is seen
between the two sets of data. This result indicates that a reduced level of alertness could be detected
with good accuracy by m o n i t o r i n g changes in the degree of openness of the driver's eyes.
Second section explains the proposed method. In the third section m e t h o d o l o g y is explained.
Results have been explained in fourth section and fifth section is conclusion and future scope

II. Proposed Method

In the proposed system, the optical cues of the driver, captured using a camera has been used
for building on a non-invasive monitoring scheme that will not divert the driver and simultaneously which will
assure precision in d e t e c t i n g d r o w s i n e s s ..Fig 2 s h o w s t h e b a s i c b l o c k diagram. A small CCD camera
positioned in front of the driver behind the steering in a position will be taking continues videos of the
driver's face. The camera should be positioned such that the driver‟s face takes up the greater part of the
image and also that the driver‟s face should be more or less in the centre of the image.
The v i d e o w i l l b e t r a n s f o r m e d t o f r a m e b y frame data and remitted to the frame memory
of the image processor. The f r a m e m e m o r y w i l l b e hoarding e a c h image in a appropriate pixel
format. A personal computer will be linked to the image processor for manipulating the image p r o c e s s i n g
p r o c e d u r e a n d for a s c e r t a i n i n g t h e processed results. The functions of the system can be
broadly divided into face detection and eye detection function, embracing the first half of the
processing schedule, and a drowsiness detection function, including the second half. Figure 2 shows the block
diagram of the eye detection process.

Fig 2: Block Diagram

III. Methodology
3.1. Flowchart of major functions
It has been stated that the proposed method work very well for IR images. Fig 3 shows the
flowchart of major functions.
3.2. Face Detection
At first the video of driver is captured using an IR camera which is fixed in front of the driver behind
www.iosrjournals.org 35 | Page
 Driver Drowsiness Detection System

the steering in a position so that driver‟s face can be clearly captured. Then the input video is converted to
frames and each frame is processed separately. Viola Jones method of face detection is then used. This
procedure classifies images based on the value of simple features. The simple features used are reminiscent
of Haar basis functions. More specifically, three kinds of features are used. The value of a two-rectangle feature
is the difference between the sum of the pixels within two Rectangular regions. The regions have the same size
and shape and are horizontally or vertically adjacent. A three-rectangle feature computes the sum within two
outside rectangles subtracted from the sum in a centre rectangle. Finally a four-rectangle feature computes the
difference between diagonal pairs of rectangles. A cascade of classifiers is then constructed to achieve
increased detection performance while radically reducing computation time. The overall form of the detection
process is that of a degenerate decision tree. A series of classifiers are then applied to every sub-window.
The initial classifier eliminates a large number of negative examples with very little processing. Subsequent
layers eliminate additional negatives but require additional computation. After several stages of processing the
number of sub-windows has been reduced radically. Further processing can take any form such as
additional stages of the cascade (as in detection system) or an alternative detection system. Much like a
decision tree, subsequent classifiers are trained using those examples which pass through all the previous
stages. The cascade design process is driven from a set of detection and performance goals.

3.3. Eye Extraction

The circumference of the iris is then estimated using the Hough transform for circle detection from
curves. So that iris is considered as a circular object and using the principle of Hough transform, the circular iris
is extracted and the visible circumference of iris is calculated. In Hough transform, the points are linked by
determining first if they lie on the curve of specified shape. This method detects both straight line and circle.
Circles are then drawn on the original image.

Fig 3 Flowchart of major functions

3.4. Perclos Estimation

The percentage of closure of eye is same as the number of pixels visible at the circumference of the
iris. So the instantaneous circumference value is compared with the pervious circumference value. Once the
percentage of closure is found, and if fatigue is detected a warning signal is issued to alert the driver.

3.5. Drowsiness detection

The alert to driver is issued based on the decision from face detection section and perclos estimation
section. The alert is given when the face is not detected and when perclos value of adjacent 2 frames is less
than 0.7*maximum value „c‟.
IV. Results
In this method a novel algoritm is introduced .This method is effective in normal web cam videos
in sufficient lighting conditions. Fig 4 shows the face and eye detected image and Fig 5 sho ws the
estimated circle of the detected eye. shows the final output.

www.iosrjournals.org 36 | Page
 Driver Drowsiness Detection System

Fig 4: Face detection and eye detection

Fig 5: Detected and estimated circle of eye

V. Conclusion and future scope

In this paper a novel method is introduced which is simple in terms of algorithm and gives more
accuracy compared to existing methods A non-insidious system to localize the eyes and monitor fatigue was
developed. Information about the head and eyes position is obtained through various image processing
algorithms. During the monitoring, the system is able to decide if the eyes are opened or closed. When the eyes
have been closed for too long, a warning signal is issued. Image processing achieves highly precise and
consistent detection of drowsiness and offers a non-insidious approach to detecting drowsiness without the
infuriation and hindrance. A non-insidious system developed judges the driver‟s alertness level on the basis of
continuous eye closures. There are a number of issues that remain to be addressed in the system. These include
improvement of its flexibility to changes in ambient brightness, assurance of consistency and attainment of a
more compact system design.

Reference
[1] S. G. Klauer, T. A. Dingus, V. L. Neale, , and J. D. Sudweeks, “The impact of driver inattention on near-crash/crash risk: An
analysis using the 100-car naturalistic driving study data,” National Highway Traffic Safety Administration, DC, DOT HS, vol.
810, 2006.
[2] T. Akerstedt, G. Kecklund, and L. H¨orte, “Night driving, season, and the risk of highway accidents.” Slee, vol. 24, pp. 401–
406, 2001.
[3] J. Connor, R. Norton, S. Ameratunga, E. Robinson, I. Civil, R. Dunn, J. Bailey, and R. Jackson, “Driver sleepiness and risk of
serious injury to car occupants: Population based control study.” British Medical Journal, vol. 324, pp. 1125– 1129, 2002.
[4] J. Horne and L. Reyner, “Vehicle accidents related to sleep: A review.” Occupational and Environmental Medicine, vol. 56, pp.
189–294, 1999.
[5] L. M. Bergasa, J. Nuevo, M. A´ . Sotelo, R. Barea, and M. E. L. Guille´n, “Realtime system for monitoring driver vigilance,”
IEEE Transactions on Intelligent Transportation Systems, vol. 7, no. 1, pp. 63–77, 2006.
[6] C. Papadelis, Z. Chen, C. Kourtidou-Papadeli, P. Bamidis, I. Chouvarda, E. Bekiaris, and N. Maglaveras, “Monitoring sleepiness
with on-board electrophysiological recordings for preventing sleep-deprived traffic accidents.” Clinical Neurophysiology, vol. 118,
no. 9, pp. 1906–1922, September 2007.
[7] J. Faber, “Detection of different levels of vigilance by eeg pseudo spectra.” Neural Network World, vol. 14, no. 3-4, pp. 285–
290, 2004.
[8] N. Wright, B. Stone, T. Horberry, and N. Reed, “ A review of in-vehicle sleepiness detection devices,” TRL Limited,
Published Project Report PPR157, 2007.
[9] T. Wakita, K. Ozawa, C. Miyajima, K. Igarashi, K. Itou, K. Takeda, and F. Itakura, “Driver identification using driving
behavior signals,” IEICE - Trans. Inf. Syst., vol. E89-D, no. 3, pp. 1188–1194, 2006.
[10] Y. Takei and Y. Furukawa, “Estimate of driver‟s fatigue through steering motion.” IEEE International Conference on
Systems, Man and Cybernetics., vol. 2, p. 1765–1770, 2005.

www.iosrjournals.org 37 | Page

View publication stats