Study of the Eye-tracking Methods Based on Video

Driver fatigue is a chief cause of the traffic accidents. The key technologies for detecting driver fatigue are the real-time and effectively detecting and tracking of driver's eyes. This paper studies the eye-tracking methods by the images of the driver's face based on video cameras. Firstly, a Haar cascade classifier for the face is designed on the arithmetic of Viola-Jones and AdaBoost. Then the eye-tracking is realized by two-step location methods. The experimental results show that the methods discussed in this paper are accurate and robust.