Research On Tracking Of Moving Object In Intelligent Video Surveillance System

Since the 21st century, with the development of science and technology, the intelligent video surveillance technology is developing rapidly, and has been wildly applied in industrial production, defense construction, commercial, transportation, finance, office automation, security and other fields. The intelligent video surveillance system is a monitoring system which uses intelligent video analysis technology, it goes beyond the traditional digital monitoring system and can maximum reduce the artificial intervention, improve surveillance efficiency and reduce the people's work burden. It can achieve target detection, separation, tracking and effective identification, make intelligent judgment, and realize the intelligent monitoring in the true sense. The detection and the tracking of the moving target is the key to the intelligent video surveillance system, therefore the research of detection and tracking of the moving target has great significance.The thesis studied on the main problem of the target detection and tracking algorithm of intelligent video surveillance system. The main research word in this article as follows:On the research of target detection, the thesis focus on the three differential algorithm, give an improved three differential algorithm for the moving target detection, and use contour detection and motion template of opencv to obtain the location of moving target, makes target detection more accurate. On the research of target tracking, according to the insufficient of based on color histogram mean-shift algorithm in practical application, the thesis establish the motion model using kalman filter, forecast the location of the target in the next frame of video image, carry on target matching and searching in this area, combine the template matching algorithm which characterize by the statistical properties of image with cam-shift algorithm to achieve the target tracking.Finally, for verifying the feasibility of the improved algorithm, according to the improved algorithm of the moving target detection and tracking, using MFC development platform and opencv algorithms library to design a intelligent video surveillance program. The experiment has indicated that the program can detect and track the moving target from the video image, and achieve simple intelligent judgment.