Land use and land cover classification using radarsat-2 polarimetric SAR image

Traditional pixel-based classification methods yield poor results when applied to synthetic aperture radar (SAR) imagery because of the presence of the speckle and limited spectral information in SAR data. A novel classification method, integrating polarimetric target decomposition, object-oriented image analysis, and decision tree algorithms, is proposed for land use and land cover (LULC) classification using RADARSAT-2 polarimetric SAR (PolSAR) data. The new method makes use of polarimetric information of PolSAR data, and takes advantage of object-oriented analysis and decision tree algorithms. The polarimetric target decomposition is aimed at extracting physical information from the observed scattering of microwaves by surface and volume for the classification of scattering data. The main purposes of the object-oriented image analysis are delineating objects and extracting various features, such as tone, shape, texture, area, and context. Decision tree algorithms provide an effective way to select features and create a decision tree for classification. The comparison between the proposed method and the Wishart supervised classification was made to test the performance of the proposed method. The overall accuracies of this proposed method and the Wishart supervised classification were 89.34% and 79.36%, respectively. The results show that the proposed method outperforms the Wishart supervised classification, and is an appropriate method for LULC classification of RADARSAT-2 PolSAR data.