Abstract: A new method, based on wavelet transform and support vector machines (SVM), is proposed for color texture recognition. Convert the RGB color texture image to HSV color model at first, then extract feature on the color texture image on the basis of tree-structured, and then utilizes SVM to classification and recognition the different features respectively. Experiments are conducted on different color texture images. Compared with the other methods, the accurate recognition rate of our method is obviously higher than it.