Time filter

Source Type

Chen S.-W.,National University of Defense Technology | Li Y.-Z.,National University of Defense Technology | Wang X.-S.,Chinese Institute of Electronics | Xiao S.-P.,National University of Defense Technology | Sato M.,Geoscience and Remote Sensing Society GRSS
IEEE Signal Processing Magazine | Year: 2014

Polarimetric target decomposition is a powerful technique to interpret scattering mechanisms in polarimetric synthetic aperture radar (PolSAR) data. Eigenvalue-?eigenvector-based and model-based methods are two main categories within the incoherent decomposition techniques. Eigenvalue-eigenvector-based decomposition becomes relatively mature since it has a clearer mathematical background and has only one decomposition solution. In contrast, model-based decompositions can obtain different decomposition solutions in terms of various scattering models. Meanwhile, conventional methods with models or assumptions that do not fit the observations may induce deficiencies. Thereby, the development of effective model-based decompositions has received considerable attention and many advances have been reported. This article aims to provide a review for these notable advances, mainly including the incorporation of orientation compensation processing, nonnegative eigenvalue constraint, generalized scattering models, complete information utilization, full-parameter inversion schemes, and fusion of polarimetry and interferometry. Airborne Pi-SAR data sets are used for demonstration. Besides, natural disaster damage evaluation using model-based decomposition is carried out based on advanced land-observing satellite/phased array type L-band synthetic aperture radar (ALOS/PALSAR) data. Finally, further development perspectives are presented and discussed. © 2014 IEEE.

Loading Chinese Institute of Electronics collaborators
Loading Chinese Institute of Electronics collaborators