Abstract: A case-based reasoning(CBR) methodology is studied for the classification of multi-temporal SAR images with the ancillary information.Multi-temporal Radarsat-1 fine-mode SAR images ranged from 2000 to 2004 are used in the experiments.The study site is located in the south of the city zone of Beijing,China;and the area can be primarily characterized by six categories of land covers,they are plow land,woodland,pasture,residential areas,water body and other land.The results of the method show the classification accuracy can achieve up to 85%.