Intelligent Video Surveillance System In The Target Detection And Tracking Algorithm

Video surveillance has been widely used in many fields where a real-time surveillance is needed, such as national defense, traffic control, the intelligent public security and so on. Nowadays, video surveillance system still depends on manual operation, which wastes resources and affects the efficiency, so studying the typical algorithms used in video surveillance and designing an intelligent video surveillance system is very important.In this dissertation, basing on studying algorithms of moving objects detection and moving objects tracking, a set of intelligence video surveillance has been designed and realized, which is used for detecting and tracking single moving body automaticly. On the research of moving object detection, temporal differencing algorithm and background subtraction algorithm which based on mixture Gaussian model are investigated and the detecting results are analyzed and compared. At last, the two kinds of algorithms are applied to the video surveillance for objects tracking.On the research of objects tracking, Mean Shift tracking algorithm is deeply studied. First, the dissertation discusses two kinds of improve algorithms based on Mean Shift algorithm, analyses their tracking effects. In addition, this paper suggests getting grayscales distribution by regions to add geometric information to the feature vector in order to increase robustness, and applies the improved algorithm to the video surveillance. In addition, this paper studies particle tracking algorithm, and applies it to the video surveillance, and analyses its tracking effect, and compares the tracking effects with Mean Shift algorithm.