A simple and effective texture classification of invariant scale and rotation was proposed．In this classifica-tionfirstlythe texture images to be classified are translated into Log-polar images via Log-polar transformation．Thenthe feature vectors of texture images are constituted with the normalized row project datum vector of the translated images. Finallythe support vector machines are used to the texture classification of invariant scale and rotation．Experimental results show that the proposed method is of an accuracy of more than90％ when it is used to the texture classification of invariant scale and rotation．