Eye Blink Detection in Real Time Video for Driver Drowsiness Detection System

In this paper, an efficient algorithm using Haar classifiers like features for real time face detection is devised then motion analysis techniques are used to locate the user’s eye by detecting eye blinks. The eye is tracked in real time using correlation with an open eye template. If the user’s depth changes significantly or rapid head movement occurs, the system is automatically reinitialized. The principle of the proposed system is based on the real time eye blink detection for warning the driver of drowsiness or in attention to prevent traffic accidents. The facial images of driver are taken by a camera with frame rate of 30fps. An algorithm is proposed to determine the level of fatigue by measuring the eye blink duration and tracking of the eyes, and warn the driver accordingly. The system is also able to detect when the eyes cannot be found. These experiments on four drivers/subjects yielded an overall blink detection accuracy of 87.01% and overall drowsiness detection accuracy of 81.14%.