A hybrid IMM/SVM approach for wavelet-domain probabilistic model based texture classification

Ling Chen, Hong Man

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Fisher kernel method was recently proposed to incorporate probabilistic (generative) models and discriminative methods for pattern recognition (PR). This method use parameter derivatives of log-likelihood calculated from probabilistic model(s), "Fisher scores", to generate statistical feature vectors. It is followed by discriminative classifiers such as "support vector machine" (SVM) for classification. In this work we study the potential of Fisher kernel method on texture classification. A hybrid system of "independent mixture model" (IMM) and SVM is introduced to extract and classify statistical texture features in wavelet-domain. Compared to existing methods that apply Bayesian classification based on wavelet domain "energy signatures" (ES) and stand along IMM, the new hybrid IMM/SVM method is able to achieve superior performance. Experimental results are presented to demonstrate the effectiveness of this proposed method.

Original languageEnglish
Title of host publicationProceedings - Seventh IEEE Workshop on Applications of Computer Vision, WACV 2005
Pages281-286
Number of pages6
DOIs
StatePublished - 2005
Event7th IEEE Workshop on Applications of Computer Vision, WACV 2005 - Breckenridge, CO, United States
Duration: 5 Jan 20057 Jan 2005

Publication series

NameProceedings - Seventh IEEE Workshop on Applications of Computer Vision, WACV 2005

Conference

Conference7th IEEE Workshop on Applications of Computer Vision, WACV 2005
Country/TerritoryUnited States
CityBreckenridge, CO
Period5/01/057/01/05

Fingerprint

Dive into the research topics of 'A hybrid IMM/SVM approach for wavelet-domain probabilistic model based texture classification'. Together they form a unique fingerprint.

Cite this