A Shadow Detection Method from Urban High Resolution Remote Sensing Image Based on Color Features of Shadow

While high resolution remote sensing images have clearer landscape detail and richer spatial information than low and moderate resolution remote sensing images, high-resolution imagery contains more shadows in the urban areas. First, four color features of shadows in HIS color model and RGB color model are analyzed. Second, an Otsu segmentation method is applied in these four color features' images to acquire four binary candidate shadow images. A final shadow detected image is acquired after the intersection of these four binary candidate shadow images and the process of morphological operation. The results show the effectiveness of the proposed approach on shadow detection.