Cast Shadow Recognition in Color Images

Shadows are often integral parts of natural scenes and their identification is an important task in image analysis. In this paper, we propose an algorithm for recognition of shadows cast by objects on the scene's background. The proposed approach is based on the use of color information by means of photometric invariant color transformations. The method is divided into two levels: first, object and shadow edges are extracted from the RGB components and only object edges from the invariant features; then, edges are filled and the obtained masks are combined to extract shadow regions. Simulation results show that the proposed algorithm is robust and efficient in detecting shadows within a set of assumptions on the scene that makes the method's applicability wider than that of state-of-the-art methods.