A Deep Learning Approach to Driver Distraction Detection of Using Mobile Phone

Abstract

Using mobile phone while driving is a big threat to traffic safety. In the rail transit, in order to prevent the driver from being distracted by the mobile phone, the real-time monitoring of drivers' behavior through video analysis technology is especially important. At present, the driver's cell phone usage detection methods are prone to object occlusion, image rotation, illumination change and are difficult to extract deep features of the image, etc., which results in a low accuracy. Therefore, this paper proposes a driver's cell phone usage detection algorithm based on deep learning. The proposed algorithm comprises 2 steps, Firstly, face detection and face tracking using PCN (Progressive Calibration Networks), determining the calling detection area. Secondly, the convolution neural network-based driver's cell phone detection method is used to detect the cell phone in the candidate area. Our experiments show that the accuracy of the proposed algorithm reaches 96.56%, the false positive rate reaches 1.52%, and the processing speed reaches 25 frames per second. It can effectively detect the driver's behavior of using cell phone.