Content Based Image Retrieval Using Motif Cooccurence Matrix

We present a new technique for content based image retrieval using motif cooccurence matrix(MCM). MCM is derived using a motif transformed image. The whole image is divided into 2 2 pixel grids. Each grid is replaced with the scan motif which minimize the local gradient while traversing the 2 2 grid forming a motif transformed image. MCM is then de ned as a 3 dimensional matrix whose (i,j,k) entry denotes the probability of nding a motif i at a distance k from motif j in the transformed image. Conceptually the motif cooccurence matrix is quite similar to color cooccurence matrix (CCM). MCM performs much better than CCM since it captures the third order image statistics in the local neighborhood. Experiments con rm that use of MCM considerably improves the performance of CBIR system.

[1]  Jian-Kang Wu,et al.  Fuzzy Content-based Retrieval in Image Databases , 1998, Inf. Process. Manag..

[2]  Ramin Zabih,et al.  Comparing images using color coherence vectors , 1997, MULTIMEDIA '96.

[3]  Paul A. Viola,et al.  Boosting Image Retrieval , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[4]  John Krumm,et al.  Object recognition with color cooccurrence histograms , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[5]  Jing Huang,et al.  Image indexing using color correlograms , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[6]  Bertrand Zavidovique,et al.  Content based image retrieval using optimum Peano scan , 2002, Object recognition supported by user interaction for service robots.