Digital Video Sequence Of Moving Target Detection And Tracking Technology

This paper mainly discusses the fundamental theories and key technologies of Intelligent Video Surveillance. The following topics are researched in details, such as detection and extraction of moving objects with static background, detection of shadow of moving objects and tracking of moving objects and so on.Concerning the object detection with static background, current theories of object detection and extraction algorithms are summarized and classified into three classes. Temporal Differential method and Adaptive Background Subtraction method are compared with each other through experimental results. At the same time, Adaptive Background Estimation method is mainly analyzed concerning the step of updating parameters. What's more, during the implementation of Adaptive Background Subtraction method, the paper proposes an improved scheme on updating background for specific application. The experimental results show that the improved scheme can satisfy the need of system to vast extent and improve the accuracy of detected result.On the research of detection of shadow of moving objects, in the view of optical characteristic of the shadow of moving objects, the paper discusses the difference of the background region, object region and shadow region in the current frame and background frame. The feature of color space that is often made use of during shadow detection is also introduced. On the basis of summary and experiments of current theories in this field, the paper proposes the algorithm of detecting moving objects combining with two color spaces and it can suppress shadow simultaneously. Numbers of experiments show that the proposed algorithm can successfully achieve shadow detection with complicated background.About the tracking of moving objects, current algorithms concerned are summarized. Object matching and mutual occlusion in context of multi-objects tracking is discussed in detail. Not only all possible occlusion is classified into different kinds, but also an improved version by increasing parameters of Kalman filter is proposed to accomplish effective tracking.