Texture Classification using Combined Multiresolution Transformations and Support Vector Machines

In this paper, the texture classification problem is investigated with hybrid multiresolution features. The studies on stationary wavelet transform (SWT) and nonsubsampled contourlet transform (NSCT) show that there are some complementary characteristics between them. In order to take their advantages simultaneously, a hybrid multiresolution method is proposed by combining the SWT with the NSCT to perform texture classification. Support vector machines are used as classifiers. Experimental results demonstrate the combination of two feature sets always outperforms each method individually.