Fast and Robust Topographic Correction Method for Medium Resolution Satellite Imagery Using a Stratified Approach

Several preprocessing steps have to be performed to reliably study mountainous terrains with satellite imagery, and one of the most important is topographic correction. The illumination conditions of these images often vary due to unequal physical properties, such as sun elevation angles and different illumination levels, while the temporal resolution of the imagery has to be accounted for as well. Two digital elevation models, a pre-classification/stratification approach and several correction methods were tested on selected medium resolution sensors. The processed images were selected to encompass different land cover types and temporal variations in solar illumination and a range of topography. It has been demonstrated over several study sites that the empirical-statistical method in combination with a pre-classification/stratification approach provided exceptional results in correcting topographic effects of the satellite imagery using the ASTER Global Digital Elevation Model. The pre-classification/stratification approach was used to split the different land cover types into “strata” which were corrected individually with the selected topographic correction method to achieve better reduction of the terrain effects.