A Novel Parameter Estimation Method for Generalized Gamma Distribution Towards SAR Data Processing

Statistical modeling of synthetic aperture radar (SAR) data is a crucial step in SAR image processing. In this context, the generalized Gamma (GGamma) distribution, which generalizes many common distributions, has been recognized as a valid model. Parameter estimation remains, however, a challenging step that conditions the quality of model fitting to data and thus, of the required processing. In this paper, we propose a novel parameter estimation method for GGamma distribution in the log-domain, named as the maximum likelihood and logarithmic cumulants (ML-LC) method. The ML-LC method constructs a novel scale-independent shape parameter estimation in the log-domain based on the Mellin transform and maximum likelihood estimation, and proposes a feasible solution technique for the shape parameter based on the multistart local search (MLS), gradient descent (GD) and bisection methods. To assess the performance of our estimation method, we perform the goodness-of-fit test on SAR data. In addition, we apply the ML-LC method in some SAR data processing tasks covering image segmentation and classification. The results obtained confirm the interest of the proposed ML-LC method.