Adaptive Cast Shadow Elimination Algorithm for Surveillance Videos Using t Random Values

Cast shadows produce troublesome effects for video surveillance systems, typically for object tracking from a fixed viewpoint. It yields appearance variations of objects depending on whether they are inside or outside the shadows. To eliminate these cast shadows from video sequences, we propose an algorithm based on the knowledge that, cast shadow points are usually adjacent to object points and are merged in a single blob on the edge of the moving objects and also cast shadow occurs only at run time (as objects move in the scene).This paper presents a novel pixel based statistical approach to detect moving cast shadows and consequently eliminates them from segmented motion objects. The approach calculates t random values (t) using color information of segmented object edges to build t distribution model, t distribution models are constructed and updated for every input frame. This statistical modeling can deal with scenes with complex and time varying illuminations. Results obtained with different indoor and outdoor sequences show the robustness of the approach.