Since almost fully automated techniques as well as expert systems in the geocoding of SAR images obtain more and more importance, the solution for the problem of the identification of homologue points between the SAR image and reference system moves into the fields of pattern recognition and feature matching. Shadow regions, which appear as dark regions in the SAR image, are independent from the backscattering of the imaged terrain. Due to the multiplicity of signals, layover regions appear brighter than the surrounding regions. Hence, the phenomena of layover and shadow are stressed in this paper. The reasons for their occurrence are studied through the digital elevation model representing the Earth's surface. An algorithm to define layover and shadow regions directly in the geometry of the digital elevation model is presented in this paper. The results are given in a so-called “layover and shadow map.”. © 1990 IEEE
