Vehicle accidents are widespread these days. They cause loss of invaluable human lives as well as huge loss of property. An efficient accident avoidance system has been a great need since the invention of motor vehicles. We propose a vehicle accident warning system based on image processing techniques. It uses two criteria for enabling accident warning. These are driver drowsiness and front vehicle distance from our vehicle. Drowsiness, especially in long distance journeys, is a key factor in traffic accidents. We use visually observable facial behaviors as indicators of driver drowsiness. For face tracking, we use Viola-Jones face detection algorithm. The eyes region is found using a novel approach to check whether they are close or open. Other image processing techniques are used to calculate distance of front vehicles. Along with this software design, an easy to build hardware is used to complete the said system to be used in realtime. To evaluate the effectiveness of proposed system, a number of drowsy persons were tested and evaluated. Experimental results show high accuracy in each section which makes the system efficient and reliable for accident warning.
[1]
Paul A. Viola,et al.
Robust Real-Time Face Detection
,
2001,
Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.
[2]
John M Sullivan,et al.
The influence of rear turn-signal characteristics on crash risk.
,
2012,
Journal of safety research.
[3]
Martin Bates.
Interfacing PIC Microcontrollers
,
2006
.
[4]
Mahmood Fathy,et al.
A driver face monitoring system for fatigue and distraction detection
,
2013
.
[5]
Hsu-Yung Cheng,et al.
Lane Detection With Moving Vehicles in the Traffic Scenes
,
2006,
IEEE Transactions on Intelligent Transportation Systems.
[6]
U Svensson,et al.
Blink behaviour based drowsiness detection: method development and validation. Master's thesis project in applied physics and electrical engineering. Reprint from Linkoping University, Dept. Biomedical Engineering, LiU-IMT-Ex-04/369
,
2004
.
[7]
Heidi D. Howarth,et al.
An Evaluation of Emerging Driver Fatigue Detection Measures and Technologies
,
2009
.