Classification trees for improving the accuracy of land use urban data from remotely sensed images

摘要

This paper summarizes the results of a joint research carried out between the Informatics Institute of IT, University of Nottingham in the UK and the Information Technology Institute (ITI) in Egypt. The research focuses on achieving higher classification accuracy to extract urban planning data from remotely sensed images. The research aims at providing an affordable system, which is capable of operating with limited data of produce the best possible classification accuracy. A Landsat TM scene of Greater Cario Metropolitan Area (GCMA) is used to test the valuidity and reliability of the proposed classification tree.