An automatic shadowdetection and compensation method for remote sensed color images
Juan Su, Xingang Lin, Daizhi Liu · 2006
An effective and robust method for shadow detection and compensation in remote sensed color images is presented. Based on particular properties of shadows in HIS color space, thresholds are chosen automatically to detect shadows, and a homogenous operator is carried out on the intensity component to preserve shadow shape. Afterwards, the segmented shadows are compensated directly in HIS color space, and then converted back to RGB color space, avoiding the compensation results seem unnatural. Experimental results demonstrate the effectiveness and feasibility of the presented method