The Research Of An Automatic Target Tracking System

As one of the hotspots in computer vision area, Moving Object Tracking Technology has been widely used in traffic control, social security and military investigation. The traditional tracking method of moving target require to delineate the boundary of moving target artificially. Through extracting some features from the training of the target area, the moving target can be followed. This paper proposed an automatic tracking system, tracking moving target based on a bit of priori information. The system get target scence information combined with the result of moving target detection. Around the automatic target tracking system, the main research work of this paper are as follows:(1)Study on the existing moving object detection methods, in view of the high computation of mixed Gauss background model, use the model cleaning mechanism to reduce the number of Gauss distribution mechanism, use easier addition and subtraction instead of complex operation, greatly reduced the operation of the algorithm. Experiment shows that the improved mixed Gauss background model has better simulation velocity under the premise of ensuring the accuracy of detection.(2) Study on the existing moving target tracking method, in view of the high computation of TLD tracking method and need to calibration target beforehand, we proposed a improvement method. We adopt frame to frame target analysis method,simplify the TLD tracking algorithm, applied to the area to be inspected, so we can get the location information of the target.(3)Perfected the theoretical basis of the related algorithms, described the basic idea and implementation process of the improved algorithm in detail. The feasibility of the improved method is verified through experiment comparision. At last, we evaluated the performance of the method.