Research On Moving Object Detection And Tracking In Sequence Image

Based on studying and organizing the relative information, this paper includes the following several topics, detecting the moving objects under astatic background, tracking the moving objects under an arbitrary, and so on.In terms of object detection, some common approaches in the field of object detecting are introduced briefly, including frame subtraction, background subtraction and optical flow. All those algorithms are analyzed and their advantages and disadvantages are concluded. Then the paper introduced the background modeling method in background subtraction including statistical average, in value method, W 4method, GMM and so on. And the related knowledge on morphologic operation and connectivity analysis is discussed.In this paper, a mixed-moving target detection algorithm based on frame subtraction and background subtraction was obtained. Firstly use frame subtraction method to detect a rough goal, and then on this basis that the image is divided into several regions, find out the region that moving target existence. These regions use the background subtraction method for the objectives of the movement is more accurate detection. The algorithm will be a combination of two methods advantages, that is integrated into the background subtraction of accurate segmentation of objects, but also integrated into the frame subtraction method strong adaptability. Experiments show that the algorithm is fast and accurate, with broad applicability. This paper also introduced the formation elements of the shadow and the different characteristics between the shadow and moving targets with different characteristics. Then obtained the shadow detection algorithm based on HSV color space of and experimental results.About object tracking, some common approaches in the field of object tracking are introduced briefly, including frame subtraction, background subtraction and optical flow in object detection and model-based tracking, region-based tracking, feature-based tracking, deformable-template-based tracking in object tracking. This paper pays great attention to Mean Shift algorithm. The algorithm has many advantages, for example: very good real-time, robust for occlusion and target distortion in recent year. Some improved algorithms are analyzed including fast moving objects and object occlusion. Finally, through experimental results analyzed the advantages and disadvantages of Mean Shift Algorithm.