ABSTRACT
Terrain and land-use classification is arguably the most important application of polarimetric synthetic aperture radar (PolSAR). Many algorithms have been developed for supervised and unsupervised terrain classification. In supervised classification, training sets for each class are selected, based on ground truth maps or scattering contrast differences in PolSAR images. For each pixel, the PolSAR response is embedded in three real and three complex parameters—a total of nine parameters. When ground truth maps are not available, the high dimensionality of PolSAR data may make the selection of training sets difficult. Unsupervised classi-fication on the other hand, classifies the image automatically by finding clusters based on a certain criterion. However, the final class identification may have to be inferred manually.
