The Research Of Tunnel Traffic Video Vehicle Detection And Tracking Algorithms

Video vehicle detector is playing an increasingly important role in the national economy. How to perform intelligent scene detection and to realize real-time and effective management of information are the core technologies of Video vehicle detector.Because of the complexity of tunnel surveillance, we use images from active freeway cameras to study. On the basis of analysis popular motion object detection and extraction algorithm, at the same time considering the complexity of the video monitoring scenes, this paper researches on several key technologies involved in video surveillance system, such as motion detection, object extraction, anomaly detection, etc.According to the practical requirements of video surveillance engineering, firstly, On the basis of analysis several background differencing algorithms and study on the complexity of surveillance tunnel scenes, this paper throws light on the algorithms of the median filter and three frame differencing, these algorithms is tested and analyzes the validity of these algorithm. Secondly, on the basis of analysis popular motion detection algorithm, considering practical hardness of motion detection in the tunnel surveillance scenes, several motion detection algorithms is integrated and modified, we bring forward a fast method to segment targets and get their feature at the same time, and compare it with normal algorithms and find it is more fast. Thirdly, we put emphasis on the research of moving vehicle tracking method; by building up the matrix of moving status of target we can solve the problem of occupation.In this paper, extensive experiments using images from active freeway cameras are reported. The results presented here give a valid proof of this algorithm, which don't require direct camera control or placement of a calibration object in the environment. This paper further argues that it is straightforward to extend this algorithm to other related traffic applications, such as accident detection, vehicle classification, moving time estimation.