Statistical Modelling of Polarimetric SAR Data

iAbstract This thesis discusses a statistical modelling technique to analyse polarimet-ric synthetic aperture radar (PolSAR) data. Polarimetric SAR data consists of four complex scattering coefficients at each image location, and must therefore be treated with multivariate modelling techniques. The work focuses on a simple class of multivariate distributions, the scale mixture of Gaussians, and in partic-ular three parametric model families of that class (the multivariate Laplacian, K and normal inverse Gaussian distributions). They are chosen because closed form expressions have been derived for these models and because their general charac-teristics, i.e., sparse symmetric non-Gaussian distributions, have been observed in experimental SAR data. The commonly used multivariate Gaussian distribution is included as a reference. The primary aim is to investigate whether any particular model has advantages over the others, or the Gaussian, with a secondary aim of using the produced parametric features for image classification. The models are characterised and parameter estimation methods, includin