Research on Real-time Traffic Flow Collection System Based on YOLO Model
For a modern city,reasonable traffic planning is the key to efficient operation of a city.As the key information of traffic planning,urban vehicle flow information originally needs manual identification,acquisition and verification of extraction methods,with the vigorous development of computer vision technology,will eventually withdraw from the stage of history.In order to improve the accuracy and timeliness of urban vehicle flow information,a real-time vehicle flow acquisition system based on YOLO model is designed by using the existing computer technology.Based on the YOLO visual detection model,the system uses DeepSORT algorithm to track and identify the detected target vehicles,judge the running status of vehicles,realize the traffic flow statistics of the current road section,visually display the recorded traffic flow information and output data.The system can effectively replace the traditional labor-consuming rigid work,and realize automatic data collection and rapid monitoring of road traffic conditions.The system is simple and interactive,and provides accurate and real-time information data for urban traffic management and traffic planning.
国家科技期刊平台
NSTL
万方数据
