Parameter estimation of log-normal distribution of RCS model for SAR images.

In order to estimate the parameters of the log-normal distribution of radar cross section (RCS ) model of synthetic aperture radar (SAR) images from the observed image, this paper proposed the parameter estimator for intensity image and amplitude image. First, it derived the first two log-cumulants of speckle. Second, it derived the first two log-cumulants of log-normal distribution. Last, it obtained the parameter estimator of the log-normal distribution of RCS model based on the second-kind statistics and the multiplicative model of speckle. The proposed estimator was easy to implement due to the analytical expressions. Parameter estimation experiments on Monte Carlo simulations and SAR images demonstrate that the proposed estimator can effectively estimate the parameters of the log-normal distribution of RCS model from the observed image. [ABSTRACT FROM AUTHOR]