ABSTRACT
Shadow detection is an important preprocessing task and a hot topic in computer vision. There exist numerous applications which vary in their motivations to address shadows in acquired digital images and video. For example, in video surveillance [1], [2], aerial exploitation [3], and traffic monitoring [4] shadows are usually mentioned as harmful effects, because they make it difficult to separate and track moving objects via background subtraction (Figure 11.1). In remote sensing, shadows may reduce the performance of change detection techniques [5]. Similarly, in scene reconstruction it is a fundamental problem to distinguish surface edges from illumination differences [6]. It should also be noted that shadow-free images are commonly required for purely visual purposes [6].
