Classification of Terrain Using the Correlation Coefficient in the Circular Polarization Basis for X-band POLSAR data

Classification of terrain is one of the most important applications of Polarimetric Synthetic Aperture Radar (POLSAR) sensing. This paper presents a simple method to classify terrain by the correlation coefficients in the circular polarization basis and the total power of scattering matrix in the X-band. It is known that the circular polarization correlation coefficient contains essential polarimetric indexes and plays the best performance in classification and detection of targets among various polarization bases. A slight modification of the correlation coefficient taking into account of reflection symmetry condition enables us to detect forest area and man-made targets effectively. A combination of the coefficient, its modification, and the total power information, leads us to classify terrain very effectively. This paper presents classified results using the X-band Pi-SAR dataset and comparison with photo ground truth data.