Visual interpretation of FCC image for land use and land cover mapping: an expert system approach

Abstract

Emphasizes importance of visual image interpretation of satellite based standard false color composite (FCC) images for small areas and describes development of a rule based expert system for application in visual image interpretation of IRS-1B based FCC images, for land use and land cover mapping. The main objective is to develop computer based support for a human photo-interpreter and beginners in the area of FCC photo interpretation. The proposed expert system has been developed using a logic programming approach in the Borland's Turbo-Prolog 2.0 environment. The knowledge base of the system is based on the multilevel classification scheme of United States Geological Surveys (USGS) for land use/cover with due variation for Indian context. This system can make classification at the first level of the above scheme, using the standard FCC (images with band combination 2:3:4:B:G:R) of IRS-1B. The knowledge base of the system has been developed in accordance with syntax of Turbo-Prolog. This system can be used to help photo interpreter and to learn the visual image interpretation skill of FCC. The system has been used to identify land cover on FCC images of an area of Vaishali district of Bihar, India. The result has been found to be satisfactory at the coarse level.