干旱区生态环境知识资源中心

An approach based on decision rules algorithm and multifeatures was proposed for desertification monitoring in Turpan Oasis using SPOT images. At first, the common methods such as principal component transformation, tasseled cap transformation and minimum noise fraction transformation were used to extract spectral feature, and vegetation index as well as wetness index were also calculated. Elevation and slope were generated from DEM. Then inter-class separability method was applied to choose the optimum features. Based on supervised classification results and each feature, a decision tree model was built to extract the land use/cover information especially desertification distribution. The results showed that the accuracy of decision tree classification is 88% and the kappa coefficient is 0.76, which had increased by 7%, 10% and 11% than maximum likelihood method, markov distance method and the minimum distance method. The whole process may provide reference for land use/cover real-time monitoring in oasis of arid areas.