The Study of Driver Distraction Characteristic Detection Technology

Driver´s distraction in driving is one of the major causes of the traffic accidents. The abnormal behavior of the driver´s head movement and the facial expressions were studied in detail in order to get the characteristics of the inattention status. With real-time monitoring on the driver´s attention characteristics: the position and movement status information of eyes and mouth, the detection mechanisms is established to detect the scattered mental of drivers state and estimate the driver´s distracting degree. Based on the detection results of eyes and mouth region, BP neural network is used to evaluate the various model of the driver´s inattention. The adoption of the D-S evidence theory is effective for the decision-level fusion of the multi-information of the driver´s distraction status. The experimental results demonstrate that the reliability and accuracy of the scattered mental of drivers state detection is highly improved by using the BP neural network and D-S rule multi-information fusion.