Comparison of the Efficiency of Different Types of Decision Tree and Maximum probability for produce Land Use maps in Arid Areas

Land use mapping is the basic tools for administrators and land planners. Different methods have been proposed for land-use mapping. The latest and most important methods is using remotey sensed data for Land-use mapping. The aim of the present study was performance evaluation of classification decision tree and maximum probability methods using Landsat 8 image of 2013 for land-use mapping of Yazad- ardacan plant. Different land use classes were difined using training samples comperison of classification. results of four different methods of, Gini decision tree, entropy, Cta and maximum probability respectively thus, Show that Kappa coefficient of 85.78, 88.95, 76.78 and 91.15 the maximum probability than decision tree methods has a higher accuracy. Map area defined by the different methods of classification, are similar in sandy lands and rocky lands. The greatest differences were observed in area of medium sand dunes and minimum differences were related to the rocky lands. Therefore, the present study proves the efficiency and feasibility of thed maximum probability method in the better classification of remote sensing images.