Research On Video Tracking Technology In Intelligent Transportation

With the constant improvement of the national economy and the continuous development of urbanization, urban traffic congestion and traffic accidents are also increasing. Intelligent transportation system applies digital image processing, pattern recognition, computer vision, and information technology in traffic management system, which can effectively help intelligentize the traffic management, keep the roads clear, reduce the traffic accidents. It can also help reduce the national investment of manpower and material resources in the road traffic, and maintain traffic order better. Intelligent transportation system becomes the key areas of the world transportation research and development of hot spots.In recent years, the widespread use of computer vision and traffic webcams has provided convenient conditions for the intelligent traffic video tracking. In order to realize the video tracking technology of intelligent transportation, the moving objects such as men and vehicles should be detected and tracked firstly. Foreground detection and tracking of moving targets in the figure has practical significance and great value. This paper studies the present stage of target detection and target tracking technology, and selects the inter frame difference method and background subtraction division, which is based on the Mean Shift(mean shift tracking algorithm for target tracking algorithm) as a target detection method.This paper studies all kinds of intelligent video surveillance technology of intelligent transportation, adopts the video server hardware and software OpenCV, which is an open source image processing library, and applies intelligent traffic video tracking technology based on target detection and target tracking. The front-end hardware collects and stores videos through the hd camera and video server. The back-end software writes programs based on OpenCV to analyse and deal with the real-time behavior surveillance video. Through using the method of computer vision, video analysis and digital image processing technology, instead of the human brain to analyse real-time video sequence and video file data, the traffic targets such as vehicles can be detected and tracked. Meanwhile the object movement monitoring, the stranded monitoring, move objects monitoring, sudden objects monitoring, starting monitoring, moving path monitoring, directional movement monitoring, wandering speed monitoring, and traffic monitoring can be completed. Thus the intelligent traffic video tracking can be achieved...