43.pdf

The aim of this study is to demonstrate the validation of the original polarization components of polarimetric ALOS PALSAR data for land cover mapping using standard supervised and unsupervised methods. For this purpose, the original and decomposed polarization components of the ALOS PALSAR data are used and the results are compared. As the methods for the land cover discrimination, the statistical maximum likelihood decision rule and isodata clustering are selected. Overall, the research indicated that the original ALOS PALSAR polarization components can be successfully used for separation of different land cover types without taking polarization decomposition.