A relative radiometric normalization (RRN) based on an Automatic Scattergram-Controlled Regression (ASCR) has been developed to create radiometrically comparable multispectral data sets, compensating for radiometric divergence present in images acquired under different illumination, atmospheric, or sensor conditions. The ASCR procedure locates the statistical centers for stable land and stable water data clusters using the near-infrared date 1 versus date 2 scattergrams to establish an initial regression line. Thresholds are placed about the initial line to select a no-change pixel set, which is used in the regression analysis of each band to derive gains and off sets for the radiometric normalization. The ASCR procedure was designed for preparing large numbers of multitemporal Landsat data sets for digital detection of landcover change.