Remote Sensing Experts Classification System Applying in the Land Use Classification in Guangzhou City

Abstract

With an increasingly rich of the remote sensing data, how to extract meaningful surface information from the mass of remote sensing data has become a problem pressed for solution. Using the experts classification method, digital elevation model and the generated slope and aspect information as expert-assisted classification parameters, land use types of Guangzhou city are classified. The results show that the participation of more variable information such as elevation and slope can prevent the frequent occurrence phenomenon of same object with different spectra and different objects with same spectrum in the remote sensing classification effectively eliminate the impact of terrain. Consequently, effectively improve the accuracy of remote sensing image classification. Both accuracy of types and precision of the area, experts classification has the higher precision than the maximum likelihood supervised classification, provides an effective way to the land use classification for the complex terrain region.