Research on Abnormal Driver Behaviors Monitoring based on Bone Key Points

With the development of social economy, the number of road vehicles is increasing, and road traffic safety has become one of the major factors threatening people's life and property safety. Abnormal driving behavior of drivers represented by mobile phone operation is one of the important causes of road traffic safety accidents. Real-time monitoring of drivers' mobile phone use and implementing appropriate measures when abnormal behavior occurs are of great significance to protect the life and property safety of drivers and passengers. Due to the differences in the driver's own appearance, the various ways of operating the mobile phone, and the chaotic background in the car, the routine abnormal behavior monitoring is prone to wrong detection and missing detection. Based on the existing human pose estimation model, this paper proposes to classify the driver's mobile phone use by using the method of calculating the Angle between the arm and the bone key points: Did not use the mobile phone, answer the phone, edit the information, to prepare for the subsequent use of the model classification. Based on this model, it can be extended to monitor other abnormal behaviors of drivers. It has practical application value to promote civilized driving and improve vehicle safety.