Various Segmentation Techniques for Extraction of Buildings Using High Resolution Satellite Images

Abstract

The current generation of high resolution satellite offers high quality and detailed information about the Earth’s surface. However, owing to the heterogeneity of the earth objects, analysis and separation of specific information from the images of these satellites pose a serious impediment. In effect, this limitation results in less accurate information from high quality images! This chapter attempts to overcome the problems through the application of various segmentation techniques including, edge detection, intensities, texture, etc. The main focus of the present work is the segmentation of man-made objects, like various types of buildings from high resolution image. This information can be very much useful for urban planning, disaster management and municipality house tax calculations. These results are further compared with the visually analyzed data of the region. On the basis of the results, a method of extraction of the required features from the high-quality imageries is suggested.