Polarization signatures of land cover classes using L-band polarimetric SAR data

Conventional imaging radars operating with a single polarization antenna measures only the amplitude of the returned signal. Therefore, accurate discrimination between similar scattering returns is difficult and, depending on the system's polarization configuration, many features go undetected. Recent developments in radar technology have led to the development of imaging radar polarimeters, that is, radars that are capable of imaging the Earth's surface at any and all possible polarizations through antenna synthesis techniques. The capability enables the complete measurement of a target's polarization properties, thus permitting a much more detailed understanding of the electromagnetic scattering process. The work presented in this paper describes the concept of polarimetric signatures, polarimetric scattering matrix and simple theoretical signatures from various models. The Stokes' matrix for individual class of pixels was generated using the amplitude and relative phase of the scattered wave. Based on the Stokes' matrices of the full polarimetric L-band airborne SAR data, polarization signatures were extracted and analyzed for four land cover classes: urban area, forest, agriculture fallow and water body. These signatures were compared with the theoretical polarization signatures and the scattering mechanisms were studied. The scattering mechanism of these land cover classes was also analysed based on the images generated by Pauli as well as Freeman - Durden target decomposition.