Driver fatigue detection based on fuzzy fusion

Abstract

A method of driver fatigue detection is proposed in this paper. Kalman filter is used to track the eyes. The curvature of up-eyelid and the distance of eyelid are fused to recognize the states of eye. This method can overcome the disadvantage of using single feature which is half-baked, not exact and uncertain. The test result shows that it can get high correct rate and precision rate.