Model-Based Nine-Component Scattering Matrix Power Decomposition

This letter aims to establish a physical interpretation of nine-component scattering power decomposition (9SD) using all coherency matrix elements/parameters for fully polarimetric SAR data analysis. It has been known that complete scattering mechanisms can be characterized by using all nine parameters of coherency matrix. We try to decompose the coherency matrix data in a physical scattering manner, as previously reported in a geometrical way by Huynen. New physical scattering models of real and imaginary parts of
$T_{12}$
are introduced, which represent dipole scattering power and quarter-wave plate scattering power, respectively. These models are added to the existing scattering models for ideal case of surface scattering, double-bounce scattering, and volume scattering. A quantitative analysis of the oriented urban patch of Mumbai in the L-band dataset result reveals a 9.7% decrease in volume scattering, which avoids misinterpretation between vegetation and the oriented urban area, and the 22.5% and 13.5% contributions come from dipole and quarter-wave scattering powers, respectively. It is confirmed that the proposed method produces better results and interpretations when compared to those by the existing decomposition methods.