An Adaptive Decomposition Approach with Dipole Aggregation Model for Polarimetric SAR Data

Polarimetric synthetic aperture radar (PolSAR) has attracted lots of attention from remote sensing scientists because its various advantages, e.g., all-weather, all-time, penetrating capability, and multi-polarimetry. The three-component scattering model proposed by Freeman Durden (FDD) bridged the data observed target with physical model, whose simplicity practicality have advanced applications. However, also some disadvantages, such as negative powers a unfitted to target, which can be improved adaptive methods. In this paper, we propose novel decomposition approach in established dipole aggregation fit every pixel PolSAR image an independent volume mechanism, resulting reduction capability models. Compared existing methods, is fast it does not utilize any time-consuming algorithm iterative optimization, simple complicate original clear for each being fitted explicit meaning, i.e., determined parameter responds mechanism target. simulation results indicated that reduced possibility occurrence powers. experiments on ALOS-2 RADARSAT-2 images showed increasing reflected more effective scatterers aggregating at 45° direction corresponding high cross-polarized property, always appeared oriented buildings. Moreover, random used FDD could expressed equal one forest area.