SEGMENTATION OF SEQUENCES WITH GAMMA DISTRIBUTION

Abstract

In images obtained with coherent lighting such as synthetic aperture radar (SAR), a multiplicative noise known as speckle appears. The speckle makes radiometric homogeneous zones respond with Gamma type distributions. The detection of edges in this type of images is of interest and usefulness in many applications. In this paper three methods of analysis by segmentation based on the calculation of the Jensen Shannon Divergence that can be applied to SAR images are proposed. The approximation of the probability distributions was carried out using the density Kernel method.