Research On Video Moving Target Segmenting And Tracking Technology

With continous development of multimedia video, video Target segmenting and tracking technology have become one of core topic in computer vision. This paper researches mainly on technology about multi-moving Target segmentation and tracking, in moving Target segmentation area, The common methods of Moving Target segmentation are studied and summarized and the virtue and defect about common methods are compared. A method of Target segmentation based on GMM background modeling is proposed in this paper. The integrated feature of pixel neighbor, brightness and chromaticity which can solve the shadow problem and false Segmentation is also discussed. This method improves the performance of segmentation.In moving Target tracking area, the linear movement model of the Target and how to use the kalman filter to predict and estimate the state of tracking moving Target are studied. We extract the key features of tracking moving Target to finish tacking in series frames. After extracting key features, The similarity metric by Mahalanobis distance are got. at last, the matching relationships are determined. The obstruction problem in the process of tracking, we solve it by the establishment of tracking Target templates to save the information of Shielding Target.At last, the processing system of video moving Target based on method of GMM background modeling is implemented, the integrated feature of pixel neighbor, brightness and chromaticity and applying the kalman filter to tracking Target is achieved too.The experiment proves that the proposed algorithm can better solve the multi- Target segmentation and tracking in the video.