Abstract: Based on the current-state-of-art of extracting urban area’s variation from remote sensing image,an L-Ratio method based on PCA is constructed taking Kunming as an example.As well,an RS processing model is presented combining such approaches for digital image processing as threshold,mask layer,and so on.At the same time,the corresponding techniques in mathematical morphology are used.This method is suitable for the geographical and economic situations in the South-West areas in China.And this method can extract the urban area’s variation information from TM images.Experimental result shows the performance of this method.