Analytical review on shadow detection and removal in images and videos

Shadows are the part of images which are mostly responsible for reducing the reliability and quality of many computer vision algorithms which include segmentation, object detection, scene analysis etc. For improving performance and increasing the reliability of such vision tasks shadow detection and removal is an important pre-processing technique. Many techniques have been proposed over the years to decompose a single image into shadow images and shadow-free images but it is still a challenging problem. Thus our goal is to analyse various shadow detection and removal techniques. A systematic study of various shadow detection and removal techniques is done in this paper and based on this study a comparative analysis is presented.