Land cover classification is one of the extensive used applications in the field of remote sensing. Recently, Synthetic Aperture Radar (SAR) data has become an increasing popular data source because its capability to penetrate through clouds, haze, and smoke. This study showed on an alternative method for land cover classification of ALOS-PALSAR data using Support Vector Machine (SVM) classifier. SVM discriminates two classes by fitting an optimal separating hyperplane to the training data in a multidimensional feature space, by using only the closest training samples. In order to minimize the presence of outliers in the training samples and to increase inter-class separabilities, prior to classification, a training sample selection and evaluation technique by identifying its position in a horizontal vertical–vertical horizontal polarization (HV-HH) feature space was applied. The effectiveness of our method was demonstrated using ALOS PALSAR data (25 m mosaic, dual polarization) acquired in Jambi and South Sumatra, Indonesia. There were nine different classes discriminated: forest, rubber plantation, mangrove & shrubs with trees, oilpalm & coconut, shrubs, cropland, bare soil, settlement, and water. Overall accuracy of 87.79% was obtained, with producer’s accuracies for forest, rubber plantation, mangrove & shrubs with trees, cropland, and water class were greater than 92%.
LAND COVER CLASSIFICATION OF ALOS PALSAR DATA USING SUPPORT VECTOR MACHINE
