An integrated approach for shadow detection of the building in urban areas

Shadows commonly exist in high resolution image, particularly in urban areas. The presence of shadows can represent a serious obstacle for their full exploitation. Especially, because of the complexity of the urban circumstance, there are some factors effecting the shadow detection. For example, the high reflectivity ground, or the glass wall of the building, makes shadow somewhere brighter so that it may cause their intensities close to the non-shadow area. In order to solve these problems, we combine the zero-crossing detection method with the DBM-based geometric method of computing the zenith angle of the sun and the altitude angle of the sun. Firstly, the proposed method uses the zero-crossing detection method to detect the edge of the whole image. Then it selects the edges of both the buildings and their shadows by matching the coordinates of DBM. Because the building and its shadow are adjacent, we match these shadows with the ideal shadow gotten by the DBM-based geometric method to get the only shadow of the building. Compared with other methods, this method may detect the shadow areas of the buildings accurately.