The Research On Detection And Tracking For Moving Object Based On Video Monitor

According to the demands for moving object detection in video surveillance, this paper improves and designs algorithm based on image block margin technology. The main work in this paper includes three aspects:1. After the image preprocessed, we used the time and spatial information to track moving object. First, the position of moving objects is roughly determined by comparing the image differences in blocks, and then the authenticity of the moving object can be captured by the time information in video sequence. This algorithm eliminates the interferer from other nonmoving objects, and has a good anti-jamming ability. The result shows that this algorithm has the advantage of self-adaptation and high speed, and it can be used to detect the moving objects based on video sequence.2. As for the shade detection, in the view of optical characteristic of the shadow of moving objects, the paper discusses the differences of the background region, object region and shadow region in the current frame and background frame. This paper improves the algorithm of detecting moving objects under the HSV spaces. We first get the HSV differential and differential histogram between the background and the foreground, and then define the primary value between the object and the shadow by the OTSU. We suppress shadow by adding the power under the HSV component. Experiment results show that the proposed algorithm can improve performance in shadow detection under complicated background.3. We design and realize the system software based on the algorithm. In the thesis, we give the solving of the projects such as structure, hardware and each module, and also give the realization of the surveillance system. The achievement of the thesis has been applied.