Driver Drowsiness Detection Using Skin Color Algorithm and Circular Hough Transform

Abstract

The fatigue state of the driver is one of the important factors that cause traffic accidents. Vision based facial expression recognization technique is the most prospective method to detect driver fatigue. Therefore, a system that can detect oncoming driver drowsiness and issue timely warning could help in preventing many accidents and consequently save money and reduce personal suffering. By mounting a small camera inside the car the face of driver can be monitored. Firstly the face is detected by using skin color algorithm and then eyes are detected by using Circular Hough transform. This paper describes a method to track the eyes and detect whether the eyes are closed or open. If the eyes are found closed for 8 consecutive frames, the system draws the conclusion that the driver is falling asleep and issues a warning signal. In this paper non-intrusive technique is used in which no sensors are used on vehicle part as well as on body of the driver which was used in intrusive technique cause irritation in long time driving. The designed system is working properly in diverse conditions such as changes in light, shadow, and slightly dark background.