Semi-automatic building extraction in dense urban settlement areas from high-resolution satellite images

Recent availability of high-resolution satellite images provides a new data source for geospatial data acquisition. This high-resolution data source has made it possible to extract man-made features such as roads and buildings, which are required for varieties of applications including urban planning, creation of GIS databases, and environmental monitoring. In this paper, we have developed a new method to extract buildings in urban informal settlement areas using high-spatial resolution panchromatic imagery. The proposed method uses radial casting algorithm to initialize snakes contours, and the fine measurement of building outlines is automatically carried out using snakes models. The building extraction results are satisfactory with an extraction rate of 94 percent as demonstrated by examples over a variety of selected test areas. The potential and limitations of applying this method to extract buildings is also discussed.