Abstract: According to the texture characteristics of residential areas on the remote sensing images,the texture features of residential areas on the remote sensing images are analyzed and described by the methods of variance based,LAWS texture energy based and LBP/C(local binary pattern/contrast) based texture analysis.Then the extend detection based on region growing is performed,and the results of change detection of residential areas are acquired.Experiments are completed on three different remote sensing images with different regions and different distribution of residential areas.The experiment results indicate that the change detection method based on the LBP/C can detect the extend change of the residential areas with high accuracy,and improve detection automation.