Iterative classification of polarimetric SAR image based on the freeman decomposition and scattering entropy

Abstract

In this paper, a new scheme for iterative classification of polarimetric SAR image based on the Freeman decomposition and scattering entropy is proposed. This technique extracts three decomposition coefficients of three scattering mechanism components through the Freeman decomposition and the scattering entropy through the H/Alpha decomposition first; then classifies the polarimetric SAR image into nine initial classes using the four parameters, and obtains the final result by iterative classification with Wishart classifier. This method, utilizing polarimetric scattering information of terrain, can acquire good effect of classification and requires less computation. The effectiveness of this method is demonstrated by the experimental results of polarimetric SAR data acquired practically.