Image segmentation is a hotspot in the field of computer vision.Gray level co-occurrence matrix is a second-order statistical measure of image grayscale which reflects the comprehensive information of image grayscale in the direction

local neighborhood and magnitude of changes

support vector machine（SVM） is used in training and classification of these features

which are energy

contrast

entropy

different variance and different entropy stand for textural image

to achieve the purpose of texture image segmentation.The process of texture image segmentation is explained

and the effect of different parameters on the segmentation precision is analyzed.The experimental results of Brodatz texture library show that this method has better segmentation effect.