Clustering is traditionally viewed as an unsupervised method for data analysis. The primary objective of cluster analysis is to partition a given data set into homogeneous clusters. In this paper, we present a novel algorithm for performing texture based clustering using com matrix and spatial information’s of pixels. In this work the co-occurrence features energy and entropy which can easily differentiate non homogeneous region from homogeneous region are considered. Run length features are based on computation of continuous probability of the length and gray level of the primitive in the texture. The final parameters are obtained by using the Expectation and Maximization algorithm. The segmentation is determined by Maximum Likelihood function. It is observed that the proposed method is computationally efficient allowing the segmentation of large images and performs much superior to the earlier image segmentation methods.