Gaussian-Hermite moments segmentation of SAR images using the Pearson system distributions

Inhalt:
This paper addresses the problem of unsupervised segmentation of SAR images by modeling and identifying the appropriate distribution for each type of field. Here we propose the idea of using different marginal distributions to improve the fitness of the Pearson’s system. Then, the mixture of distributions characterizes the statistic feature of the images and is estimated by the SEM algorithm and segmentation of SAR images is performed based on Gaussian-Hermite moments (GHM) feature with the system of Pearson distribution. Then, the algorithm is applied to the segmentation of real SAR scene.