无重叠视域多摄像机运动目标匹配分析-analysis of multi-camera moving target matching in non-overlapping field of view.docxVIP

PAGE PAGE IV The Moving Object Re-identification for Multiple Non-overlapping Cameras Abstract The Moving Object Re-identification for Multiple Non-overlapping Cameras Abstract In recent years, with the rapid development of intelligent video surveillance system, as well as the urgent demand of the security situation, people’s requirements for smart intelligent video surveillance system are increasing more and more, and it is much accounted of day by day that multi-cameras network because of its expansive surveillance areas. Nevertheless, on the one hand, due to the consideration of capital and resources, it is unlikely to cover all surveillance area with a great amount of monitors. Therefore, camera tracking with non-overlapping views has become a main part in the study of wide-area video surveillance. On the other hand, the precondition in the camera tracking is object matching which is the significant research project in the object detection and matching for multiple non-overlapping surveillance cameras. This thesis contrapose the existing problem in the moving object detection of disturbed irregular background, object tracking with a single static camera and object re-identification for multiple non-overlapping surveillance cameras. With the in-depth study, the following results can be achieved: To the question the high rate of false drop of moving target detection with Gaussian mixture model in the irregular disturbed background, combined with the motion estimation method usually used in video coding, we proposed an effective method of real-time moving object detection. Gaussian mixture model can effectively cope with regular disturbance in the scene, and has an advantage of foreground area extraction, at the same time, irregular disturbing region in the background will also be detected as moving foreground region. Therefore, with the detection results of Gaussian mixture model, we further utilize the regional motion estimation method to differentiate irregul