Camshift Combined With Kalman Filter Tracking Technology Research

Target tracking algorithm for computer vision is of great significance, it will pattern recognition, artificial intelligence, image processing, automatic control and many other fields of advanced technology, widely used in intelligent machines, medical diagnostics, intelligent surveillance, traffic controlareas. and other fields. At this stage, the accuracy and robustness of the moving target detection and tracking algorithms, has been gradually difficult to meet the increasingly high demand of a variety of complex applications, which make it become one of the research focuses at present. This paper will be used in the study of moving target detection and tracking algorithm, on the basis of it, i will make further improved and perfected, in order to achieve better tracking results.In moving target detection, this paper first introduces the basic principles of the three detection methods, aiming at the limitations of traditional methods, combines the background difference method and frame difference method, put forward a kind of improvement method. Then, in order to better perform the next target tracking, carried out a series of treatments to the detected image, including the opening operation, closing operation and connectivity analysis.In the research of tracking algorithms, this paper will focus on analysis of the advantages and disadvantages of camshift algorithm and Kalman filterof in Tracking area. Then, combine the two together, put forward an improved algorithm, and use experimental results show that, the improved algorithm not only improves the tracking accuracy, but also can make the operation time was shortened greatly, improve the tracking in real time.