Partially correlated K-distribution for multi-look polarimetric SAR imagery

A partially correlated polarimetric K-distribution model is presented to characterize statistical properties of multilook polarimetric SAR data from urban areas. The model was derived using a more general matrix form of the product model that uses separate gamma distributed texture variables for different polarization channels and assumes that these texture variables be partially correlated. Experimental assessment of the model was conducted using NASA/JPL polarimetric SAR data from an urban area, showing improvement in better characterizing urban polarimetric backscattering.