A New Edge Computing Method Based on Deep Learning in Intelligent Transportation System

Vehicle tracking and accident risks brought by transportation are becoming more and more serious. This paper takes traffic flow detection in intelligent transportation system as the research base point, combines deep learning with edge computing technology, and proposes a new method of intelligent transportation edge computing based on deep learning. Firstly, the intelligent traffic detection algorithm based on deep sorting is implemented, and the vehicle reidentification data set is used to train the deep sorting tracking model, extract deep learning vehicle features, and establish the vehicle intelligent tracking algorithm model. After testing, the model has good tracking accuracy and real-time performance. Secondly, the low delay and high reliability of edge computing technology are used to monitor the road vehicle data, analyze abnormal vehicle information and abnormal traffic flow information, establish an intelligent traffic accident risk assessment model, evaluate the road traffic risk, and then use the minimum risk Bayesian decision method to verify the judgment results. Finally, the effectiveness and superiority of the proposed method are verified by comparison experiments of various prediction models, the minimum reduction in waiting time obtained was at least 11.3%.