A LSMA-based Comparison of the Performances in Retrieving Impervious Surface Between Landsat ETM+ and EO-1 ALI

The extraction of impervious surfaces from satellite imagery has been a hot topic in the remote sensing field over the past decade.Nevertheless,whether the impervious surface information extracted from different sensor images is comparable is still unknown.This paper implemented a complementary study based on a comparison of the retrieved impervious surface information from Landsat ETM+ and EO-1 ALI sensor data.Impervious surface features were derived from a date-coincident image pair of the two sensors by using linear spectral mixture analysis(LSMA).The accuracy of retrieved impervious surface information of the two sensors was assessed and compared.The results show that the ALI image has higher accuracy than ETM+,as suggested by its higher overall accuracy and Kappa coefficient and lower root mean square error and systematic error(in absolute value).The differences in spectral resolution and radiometric resolution between the two sensors are believed to be the main factors causing these differences when retrieving impervious surfaces.An increase in spectral information in ALI sensor can be of help when distinguishing differences between land cover types,while the enhancement in radiometric resolution in the ALI sensor can make the sensor more sensitivite when detecting ground surface features.