On Texture Feature Extraction Based on Local Walsh Transform
Atr La · Signal Processing · 2005
In the paper, a new texture features extracting method using Local Walsh Transform (LWT) is presented. The definition of LWT is firstly given out. Then the statistic property of LWT coefficients is analyzed. The texture discrimination performance of the central moments of LWT coefficients are examined. And the 2nd,4th,6th order moments are selected as texture features. Comparing with other methods lick Gray Level Cooccurence Matrix (GLCM), Texture Spectrum (TS), Local Fourier Transform (LFT), Markov Random Field (MRF), Gabor Filter Banks, texture features based on LWT have the best texture discrimination performance, and is easily computed.