Parameter estimation with correlated clutter textures

Many natural clutter textures can be shown to be well represented by correlated K-distributed noise. Such noise processes can be completely described in terms of their average two-point statistics (the autocorrelation function ACF). The inverse problem to be solved in these cases has been shown to be the determination of the parameters of the surface cross section noise process from the measured intensity ACF. In a previous paper the author derived the relevant statistical properties of the intensity ACF estimate and compared two different texture discriminant measures, based on the ACF. The author extends the analysis to consider the determination of the parameter values themselves.