Video Objects Extraction Based On DFD Between The Frames and Threshold Segmentation

摘要

In order to eliminate the effect caused by noise,complex motion,and uncovered background in the process of change detection segmenting video objects,a new method for video objects extraction based on DFD between the frames and thresho|d segmentation is proposed In this method,filtering and obtained two consecutive difference between the frames,and then amended the difference images by "assimilation filled" to get the difference template,and use the templates buffer to maintain the integrity,of iteration template. The algorithm does not depend on a fixed background,and can eliminate the uncovered background in the difference images,we can get precise shape of the moving targets by use this method,and this algorithm is simple and fast.