
	Information Technology Journal
	Year: 2010 \| Volume: 9 \| Issue: 5 \| Page No.: 1003-1012 DOI: 10.3923/itj.2010.1003.1012

Video Macrosegmentation Using Automatic Analysis of Similarity Matrices

S. Haidar, B. Chebaro and B. Haidar

Abstract: In the present study, we propose an automatic method for segmenting video material using the comparison matrix. Without a priori information, neither any training phase, the similarity matrices are filtered then regional minimums and maximums are extracted over the line and column projections. Pseudo-homogeneous areas are identified between each pair of regional minimums. The pseudo-homogeneous areas in the resulting matrix are thus the macro-segment video records in question. We experiment our method over a database of short video commercials to illustrate the potential of our approach. An overall evaluation is finally established and given.

PDF Fulltext XML References Citation Report Citation

RELATED ARTICLES:

•	Full 3D Compact Histogram Segmentation of Color Images

•	An Efficient Frequent Temporal Pattern Mining Algorithm

•	Implementation of Matrix Folding in Multimedia Transmission

How to cite this article:

S. Haidar, B. Chebaro and B. Haidar, 2010. Video Macrosegmentation Using Automatic Analysis of Similarity Matrices. Information Technology Journal, 9: 1003-1012.

DOI: 10.3923/itj.2010.1003.1012

URL: https://scialert.net/abstract/?doi=itj.2010.1003.1012

COMMENT ON THIS PAPER

Navigation


		Online First
		Current Issue
		Previous Issues
		Editorial Board
		Submit a Manuscript
		Guide to Authors
		Subscribe to E-alerts

Indexed In


		ASCI-Database
		Asian Digital Library
		Google Scholar

Science Alert