Shadow elimination method for moving object detection

This paper presents a new method for eliminating shadow areas accompanied with pedestrianlike moving objects in visual surveillance systems. In sunny outdoor scenes, shadow areas cast by moving objects such as pedestrians are often detected as a part of the moving objects since shadow areas move in accordance with the movement of objects. When detected areas contain. shadow areas! large errors occur respecting the estimation of the number of objects from their sites, analysis of their shapes, and estimation of their locations. Thus, it is important to separate shadow areas from object ones. The proposed method gets rid of shadow areas by the way of height information since most of the shadow areas accompanying moving objects are assumed to be on the road plane. Two cameras set at any locations so that their common visual fields include the surveillance area are used for this purpose. The image obtained from the one camera is inversely projected to the road plane and the projected image on the road plane is transformed to the view from the other camera. Parameters for this transformation are easily acquired by indicating several corresponding points between images obtained from two cameras. Shadows existing on the road plane occupy the same areas in the transformed image and in the image acquired from the other camera, whereas objects' areas with different heights from the road plane occupy different areas in these images. Therefore, shadow areas can be removed by a subtraction between these images. An experimental result (in detecting pedestrians) on a pedestrian crossing shouted the effectiveness of the proposed method.