Research On Moving Object Detecting And Tracking In Video Surveillance System

As the rapid development of networks and multimedia technology, Image information with its intuitive and vivid forms is very popular. Compared to static images, moving target video sequences provide more information. Intelligent video surveillance system can automatically analyze the sequence of images by computer vision and the methods of video analysis. The system can detect, locate, recognize and track objects in a moving environment in real-time. Furthermore, it can also analyze and judge the movement of objects.Currently, Motion object detection and tracking technology has a very common application in computer vision, intelligent video surveillance, traffic management and automatic monitoring, detection and tracking of the human body. In this thesis, the motion object detection and motion object tracking technology under the fixed camera gathering video image sequence are studied. The content is composed of motion object detection and motion object tracking.In the motion object detection aspect, the thesis studies analysis and compares the typical motion object detection algorithms, and has pointed out the respective advantages and disadvantages. By combining the frame difference algorithm and the background subtraction algorithm, the thesis uses the background subtraction algorithm to detect the moving objects based on the background build with the frame difference algorithm, which can achieve the detection of moving targets rapidly.In the motion object tracking aspect, the thesis focuses on MeanShift tracking algorithm. And then make the improvements on the shortcomings of the MeanShift tracking algorithm. By the result of motion object detection, the automatic tracking of the target is realized. The results of experiments show that the algorithm has a better accuracy and real-time performance.