一种基于Freeman分解与散射熵的极化SAR图像迭代分类方法

Iterative Classification of Polarimetric SAR Image Based on Freeman Decomposition and Scattering Entropy

该文提出了一种基于Freeman分解与散射熵的极化SAR图像迭代分类新方法。该方法首先通过Freeman分解提取3种散射机理成分的功率,同时通过H/α分解提取地物的散射熵;再利用这4个表征地物特性的参数将极化SAR图像中的地物划分为9个初始类,最后使用Wishart分类器对初始类进行迭代分类得到最终的结果。该方法合理利用了地物的极化散射信息,能够取得较好的分类效果,同时运算量也比较小。实测极化SAR数据的实验结果验证了该方法的有效性。

In this paper, a new iterative classification of polarimetric SAR image based on Freeman decomposition and scattering entropy is proposed. This technique extracts the powers of three scattering mechanism components through Freeman decomposition and scattering entropy through H/a decomposition first; Then using the four parameters which can characterize terrain divides the terrains of polarimetric SAR image into nine initial classes, and the final result is obtained by iterative classification with Wishart classifier. This method utilizes polarimetric scattering information of terrain with reason, can acquire good effect of classification and requires a little operation The effectiveness of this method is demonstrated with the experimental results of polarimetric SAR datas measured practically.