Information Extraction from Satellite-Based Polarimetric SAR Data Using Simulated Annealing and SIRT Methods and GPU Processing

The main goal of this research was to propose a new method polarimetric SAR data decomposition that will extract additional information from the Synthetic Aperture Radar (SAR) images compared other existing methods. Most current methods are based on scattering, covariance or coherence matrices describing radar wave-scattering phenomenon represented in single pixel an image. A lot different have been proposed up now, but problem is still open since it has no unique solution. In research, signature matrices. Such may be used reveal hidden about image target. Since signatures (size 18 × 9) much larger than scattering 2 2), 3 4 4) matrices, essential use appropriate computational tools calculate results within acceptable time frame. order estimate effectiveness presented method, obtained were with outcomes another (Arii decomposition). conducted showed solution, Arii decomposition, does not overestimate volume-scattering component built-up areas and clearly separates objects mixed-up areas, where both building, vegetation surfaces occur.