一种基于退化模型的高分辨率SAR去噪算法

为了保持高分辨率合成孔径雷达（SAR）图像中的纹理结构，提出了一种基于高斯．马尔可夫模型（Gauss-Markov Model）的方法来抑制SAR图像的斑点噪声。通过引入贝叶斯分析框架，建立Markov随机场的退化图像恢复模型，从而将图像的恢复问题转化为求解最大后验概率（MAP）问题，并直接从噪声图像中估计随机场模型参数进行有效的噪声抑制。实验结果表明，对所研究的高分辨SAR图像，基于退化模型的去噪算法（RMBD）不论是在噪声的去除上还是在结构信息等细节的保持上均不同程度地优于其他常用斑点去噪方法。

A Restoration Model Based Despeckling Algorithm of High Resolution SAR Images

In order to preserve the textural feature affected by multiplicative speckle especially in high resolution synthetic aperture radar(SAR) images,this paper proposes a despeckling method based on the Gauss-Markov model to suppress the speckle in SAR images.By introducing bayesian analysis framework,restoration model of degradation image of markov random field is built,and then the problem of image restoration is transformed into the problem of solving a maximum a posterior(MAP),random field model parameters can be also estimated directly from noise image,thus speckle is effectively reduced.In this paper,on the basis of discussing the main idea of the restoration model based despeckling(RMBD) algorithm in detail,other commonly used denoising methods are compared with the proposed method,experiments show that the model-based despeckling algorithm achieves better performance not only at speckle reduction but also at preservation of structural detail information than other commonly used speckle filters.