Abstract

D. Amarsaikhan*, M. Ganzorig, H. H. Blotevogel, B. Nergui and R. Gantuya

The aim of this research is to propose an integrated methodology for the extraction of information from high and very high resolution satellite images and also to demonstrate how the extracted information can be used for urban area studies. For the information extraction, three different approaches such as interpretation, automatic and artificial intelligence methods are used. As the interpretation method a visual interpretation is used, whereas as the automatic method a refined maximum likelihood classification that uses spatial thresholds defined from the local knowledge is applied. For the artificial intelligence method, a rule-based algorithm which consists of a set of rules, containing an initial image segmentation procedure as well as the constraints on spectral parameters and spatial thresholds is constructed. Overall, the research indicates that the high and very high resolution satellite images can be successfully used for general and detailed urban studies and the proposed methods can be efficiently used for the extraction of accurate thematic information.