Shadow removal in remote sensing images using features sample matting

摘要

Remote sensing images often suffer from shadow duo to partially or totally occludes direct light from an illumination source. In this paper, we propose a novel shadow removal algorithm. The trimap could be generated automatically by morphological subtraction method according to the result of shadow detection. Then, the weighted color and texture sample selection image matting method is applied in order to obtain the accurate shadow coefficient. The experiment results are illustrated with practical examples and verify the efficacy of this algorithm.