Content Based Video Retrieval Using Information Theory

During the last decades a large set of video archivesis created and rapidly multimedia growth creates new challengein the image processing world. A reliable system is needed toautomate the process of this large amount of data. Video analysesare done in two different levels, low level and high level. Thereare many problems in video content analysis and in this work weanalyzed content based video analysis. Our proposed method isbased on Information theory. These systems consist of three mainparts which includes: Shot Boundary Detection, Hierarchicalvideo summarization, retrieve and index target video. Systemperformance is evaluated on TRECVID2006 Database, resultsshown the usefulness of the proposed method.

Keywords:

Information Theory , Video Retrieval , video indexing