期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2006
卷号：XXXVI Part 4
出版社：Copernicus Publications
摘要：Classification of land cover using SAR images is an area of considerable current interest and research. A number of methods have been developed to classify land cover from SAR images and these techniques are often grouped into supervised and unsupervised classification algorithms. Supervised methods have yielded better accuracy but suffer from the need of human interaction to determine the classes. In contrast, unsupervised methods determine classes automatically but limitation with the algorithms developed for this method is that they use multi band or multi polarized data. In this paper, we proposed an algorithm to effectively classify a single band, single polarized image having intensity and texture information only. The proposed algorithm provides the user with the required parameters for direct segmentation process with out any trial and error approach. Classification accuracy for water and urban areas are computed by comparing with LISS image and topographic sheet and quite good agreement is obtained. The algorithm is also evaluated by applying the results to the SAR images obtained at different time instants
