CONTOUR BASED FOREGROUND MOVING OBJECT EXTRACTION IN VIDEO SEGMENTATION

Abstract

Video segmentation plays an important role in the MPEG-4 standard for multimedia applications. In this paper the video segmentation is done by detecting the edge of the objects of the each frames and keep tracking it. Then morphology motion filter is used to find the edges of the foreground objects. Next, a filling technique is used to connect edge information to generate final object. Though there is no need for the moving camera situation in the surveillance system, it can deal with both still and moving camera situations. The object based or shape based retrieval is better than the other video on MPEG-4 standard. The main objective of selecting the edge based video segmentation is, the object can be easily stored, indexed and retrieved in the MPEG-4 standard for multimedia applications. The edge detection process serves to simplify the analysis of images by drastically reducing the amount of data to be processed, while at the same time preserving useful structural information about object boundaries.

Keywords: Video segmentation, MPEG-4,  Edge detection