Rotation Invariant Color Texture Classification in Perceptually Uniform Color Spaces

* * * b a L is a perceptually uniform color space and HSV approximately perceptually uniform space for representing color. But commonly used color space is RGB, which is not perceptually uniform. In this work we compare these three-color spaces for rotation invariant color texture classification. The inclusion of color aspects of texture in image processing is increasing rapidly. Gabor wavelets are used to obtain features. We compare the color spaces by using two strategies. One is based on quadratic Bayesian classifer. The other is classifier independent based on Bhattacharyya distance figure of merit. Perceptually uniform color spaces gave better results.