基于FCM算法的SAR图像相干斑噪声滤波算法研究

针对FCM分割对SAR图像斑点噪声敏感的特点，本文提出了一种新的去噪方法。所提出的方法主要分三个步骤。首先，分别对获得的两幅同源不同时相的SAR图像进行小波和lee滤波结合去噪。然后，通过对数比值检测方法获得变化信息。利用双边滤波器处理变化信息，以获得能保留图像丰富细节信息的图像降噪效果。最后，利用FCM方法把变化信息分成两类，这样就可以获得变化检测结果。Ottawa 地区的部分图像作为检测算法的性能的数据库。将去噪方法相互比较，结果表明,本文提出的算法分类正确率达到98.29%，优于其他去噪方法。

A COHERENT NOISE FILTERING ALGORITHM IN SAR IMAGES BASED ON FCM ALGORITHM

In order to improve FCM segmentation, this paper proposed a new image denoising method. The traditional FCM method was sensitive to SAR image speckle noise. This proposed method worked in three steps. First of all, this paper used the wavelet transform method combined with lee filtering method to denoise two SAR images respectively. The two images were of homologous phase from same area. Then, the logarithmic ratio detection method achieved the change information. With the bilateral filter, the denoising result can keep ample details of image. Finally, the FCM method divided the change information into two classes, and the change detection result was achieved. This method took the parts of image of Ottawa area as data set for the performance evaluation. Compared with other denoising methods, the results show that the classification accuracy of the proposed algorithm reaches 98.29%, so it is better than other algorithms.