Personnel identification and distribution density analysis of subway station based on convolution neural network

In this paper, a method based on convolution neural network and multi-camera fusion is proposed to improve the recognition accuracy of crowd then personnel distribution subway station platform analyzed. method, tensorflow used as deep learning training framework yolov4 algorithm identify area using three videos synchronously. Through affine transformation time average statistics, passenger density each sub-area calculated in whole The results show that number people recognized by multiple cameras 58% higher than single camera. new has high rate for actual scene with large more obstacles. Finally some areas risk aggregation have been found, which should be focus safety monitoring.