Land Use Information Extraction from Remote Sensing Data Based on Decision Tree Tool

Information extraction of land use from remote sensing image is one of the important ways for land use data acquisition.The spatial distribution of the land resources,the methods of the land-use,the resolution of the remote sensing data and number of the bands have a great impact on the information extraction results.The theory of decision tree methods are discussed.And the remote sensing data and other relevant data are used for the classification of land use information.By the statistic of the spectral characteristics,the characteristics of the normalized index of some type of the land-use,such as arable land,and the DEM model,the paper get a land-use decision tree branch for the information extraction,which can get a better solution for distinction the water bodies,construction shadows and other confused land-use type.The results proved that this method can achieve the initial classification and information quickly and effectively,which can make a feasible approach for land use information extraction by the comprehensive analysis of remote sensing data.