Real Time Driver’s Drowsiness Detection System Based on Eye

This paper presents a design of a unique solution for detecting driver drowsiness state in real time, based on eye conditions. Many approaches have been used to address this issue in the past. But this paper presents a simple algorithm based solution with min imum hardware requirements. Under the controlled environment, the proposed system is successfully operated to generate results with approximately 90% accuracy. The system tends to use a web camera to capture a series of images. These captured images may further be stored as individual frames in our system. The frames so formed, are provided as input to face detection software. In terms, our required feature (eye) is extracted from the image. Individually working on each eye, the system establishes a condition and suggests a specific number of frames with the same eye condition that may be registered. The result of these images may be taken as input to obtain the level of drowsiness that a driver may encounter at any certain stage while driving a vehicle.