基于广义伽马分布的SAR图像相干斑抑制

SAR Image Despeckling Based on Generalized Gamma Distributio

引入广义伽马分布对SAR图像杂波的统计特性进行建模,在此基础上给出了一种自适应SAR图像相干斑抑制方法。该方法在贝叶斯理论框架下利用最大后验概率准则对地物的真实后向散射强度进行估计。分布模型的参数采用以梅林变换为基础的对数累积量方法进行估计。基于SAR实测数据的仿真实验结果表明,提出的方法能够有效地去除SAR图像中的相干斑噪声。

In this paper,the recently introduced generalized Gamma distribution is utilized to model the statistical properties of synthetic aperture radar（SAR） imagery,and a new adaptive despeckling algorithm is accordingly proposed.The backscattering cross-section of distributed targets is estimated using maximum a posteriori criterion in the framework of Bayesian theory.The unknown model parameters are estimated by means of method of log-cumulants relying on the Mellin transform.Experimental results show that the proposed despeckling algorithm can efficiently removes speckle noise from SAR images.