Abstract: Based on the data of ALOS image of Yiliang County,Yunnan Province,this paper discusses the use of the maximum likelihood method,support vector machine method and object-oriented support vector machine(SVM).The results show that maximum like-lihood classification accuracy is 79.33%,SVM classification accuracy 82.25%,oriented object based support vector machine classification accuracy 86.13%,and oriented-object based support vector machine classification method has better classification results.The results can provide a reference for the study of high-resolution remote sensing image classification。

Key words: ALOS, maximum likelihood method, SVM, object-oriented