Intelligent Non-motor Vehicle Behavior Monitoring System Based on Convolutional Neural Network

This paper aims to use convolutional neural network (CNN) to develop an intelligent system, mainly to detect the helmet wearing and overload behavior when driving non-motor vehicles on the road. The intelligent non-motor vehicle behavior management system has important market and application value for the urban traffic system, which can play a positive role in promoting the driver's behavior and road safety management, and can be used to assist the supervision of non-motor vehicle drivers to comply with the road traffic safety law. This system can be combined with the street monitoring equipment to achieve 24 hours a day, the whole process without dead Angle detection, capture the real-time picture of the electric vehicle drivers, timely monitoring and identifying the drivers violating the road traffic safety law, and upload the corresponding violations.