Research on A Driver Fatigue State Detection System

In view of the distraction and fatigue driving of the driver, the detection research on the fatigue state and the sitting state based on machine vision is carried out. When there is an abnormal situation, the detection system will issue an alarm signal in time. Through comprehensive face detection, human eye detection and mouth area detection, the characteristics of human eye state and mouth area under fatigue driving state are discussed, and image processing technology is used to extract driver facial image information, and then the extracted eye area or The mouth area is processed to complete the detection and analysis of the driver's fatigue state. The sitting posture monitoring is performed by discriminating the driver's sitting posture. A closed-eye discrimination based on human eye detection, a yawn discrimination based on the fusion edge statistics of the mouth active region, and a monitoring experiment based on sitting tilt were performed. The experimental results show that this paper can solve the problem that the single indicator can judge the fatigue state of the driver and the details of the human eye information are not full. The experimental results are better and the accuracy of the fatigue driving state detection is improved.