Abstract:

A new statistical model based on wavelet domain singular value decomposition is introduced for image texture classification. Probability density function of the singular values is modeled as exponential distribution. The model parameters are used as features for the classification. Closed form expressions for the various probabilistic distances such as Bhattacharyya, Matusita, Divergence, Patrick-Fisher, Kolmogorov and Kullback Leibler are devised for exponential probability density functions. These distances are used for the classification. The performance of the method is evaluated using two texture sets. Experimental results show the superiority of the proposed approach compared with some existing methods.

Keywords: Image texture classification, Wavelet packet transform, Singular value decomposition

Biography:

	S. Ramakrishnan, is a Senior Lecturer at PSG College of Technology, Coimbatore, India. He has 6 years of teaching experience.  He is pursuing PhD degree from Anna University. He has published fifteen papers in international and national journals and conference proceedings. His areas of research include digital image processing, soft computing, and digital signal processing.
	Dr.S.Selvan, is Professor and Head, Department of Information Technology at PSG College of Technology, Coimbatore, India. He has 27 years of teaching experience.  He has published more than 60 papers in international and national journals and conference proceedings. His areas of research include   digital image processing, soft computing, digital signal processing and computer networkin

BibTex:

@ARTICLE{P1150634004,

AUTHOR = {S.Ramakrishnan and S.Selvan},

TITLE = {A New Statistical Model Based on Wavelet Domain Singular Value Decomposition for Image Texture Classification},

JOURNAL ={ICGST International Journal on Graphics, Vision and Image Processing},

}

(Full Paper571KB)