Abstract: To solve the shortcomings of the traditional hierarchical clustering algorithm for remotesensing image segmentation,in this paper we propose a new algorithm based on a cloud model and EMalgorithm.In the proposed algorithm,we extract the bottom concepts from the remote sensing imageusing apeak method for cloud transform,and get higher cloud concepts through the EM algorithm,and finally segment the image with Great Criterion.The results show that clustering concepts can beclassified quickly as a specified number by the EM algorithm for estimatation of the mathematical fea-tures of the higher clouds.The proposed algorithm is superior to the traditional methods.