基于Contourlet域主成分分析的SAR图像去噪

SAR Image Denoising Based on Contourlet Domain PCA Analysis

相干斑噪声是合成孔径雷达图像所固有的,并且严重降低了图像的可编译性,影响了后续图像分割,特征提取,目标分类和识别等工作。因此,SAR图像的相干斑去除问题一直是SAR图像应用研究的重要问题之一。针对SAR图像噪声去除问题,提出了一种基于Contourlet多尺度分解域主成分分析的SAR图像去噪新方法,并且简要归纳了已有的SAR图像去噪方法。方法首先对源图像进行Contourlet分解,在不同频段的子带图像中,利用主成分分析方法进行能量保持,用重构图像来进行子带去噪,最后通过Contourlet逆变换得到去噪之后的图像。在SAR图像上的实验结果表明,方法不仅较好地保持了图像的纹理和细节特征,且信噪比也较高。

Speckle noise is generated by the coherent processing of synthetic aperture radar （SAR） signals. It severely affects the image understanding. The next processing steps such as image segmentation, feature extraction, object classification and recognition are influenced severely. Therefore, speckle denoising is always one of the important problems in SAR image application study. A new method based on principal component analysis （PCA） in Contourlet multi - scale decomposition domain is proposed in order to solve SAR image denoising problem. Furthermore, existing speckle denoising methods are introduced briefly. Firstly, Contourlet transformation is applied to source image with this method. Secondly, image denoising is executed for different frequency sub - images with reconstructed images by PCA method. Finally, denoised image is attained with inverse Contourlet transform. The experimental results of SAR images demonstrate that the proposed method can not only reserve the texture and detail characteristics of the image, but also improve SNR.