Abstract—In this paper, a multi-feature fatigue decision making method based on Bayesian conditional probability formula is proposed. The captured video is detected by a face detector and then positioned to the eyes and mouth portion. The eye closure degree, the mouth opening degree, and the nodding frequency data are collected, and the driver is judged to be in a fatigue state according to the Bayesian probability formula. If fatigue is detected, an alarm is issued. According to the method, it is possible to accurately detect whether the driver is in a fatigue state, and to help the driver return to the awake state by means of the alarm method, thereby reducing the occurrence of traffic accidents.

Index Terms—Bayesian probability formula, eye chosure degree, mouth opening degree, nodding frequency

Cite: Li Zhao, Yang Liu, and Nianqiang Li, "Fatigue Driving Detection System Based on Bayes' Theorem," Journal of Image and Graphics, Vol. 7, No. 3, pp. 76-81, September 2019. doi: 10.18178/joig.7.3.76-81