基于分块奇异值分解的两级图像去噪算法

为了更有效地进行图像去噪, 提出了一种基于分块奇异值分解(Singular value decomposition, SVD) 的两级图像去噪方法, 该方法首先将含噪图像中具有相似结构的图像块组织成具有很强相关性的图像块组; 然后, 利用二维奇异值分解去除图像块组中每个相似块的内部相关性, 利用一维奇异值分解去除相似图像块组之间的冗余; 最后, 通过硬阈值方法收缩变换系数实现图像与噪声的有效分离. 为了进一步提高去噪效果, 对含噪图像再次进行上述操作. 不同的是, 在第二级去噪过程中,相似图像块组根据第一级估计出的图像计算获得且相似图像块间的相关性通过离散余弦变换去除. 仿真实验表明, 提出的两级图像去噪算法不仅可以较大程度地去除图像噪声, 还能有效保留图像细节, 取得了良好的去噪效果.

Two-stage Image Denoising Using Patch-based Singular Value Decomposition

This paper presents an effcient patch-based image denoising scheme by using singular value decomposition (SVD). In this scheme, similar image patches from a noisy image are simply grouped together. For a better sparse representation of these similar patches, firstly, the 2-D SVD is utilized to reveal the essential features of each individual patch, and then the 1-D SVD is utilized to exploit the correlation between similar patches. By doing so, the image features can be well preserved when attenuating the noise by the shrinkage of transform coeffcients. To further improve the denoising performance, the proposed scheme is employed once again. But the similar patch grouping is performed from the first-stage estimated image and a fixed orthogonal transform instead of 1-D SVD is adopted to remove the redundancy shared by similar patches. Experimental results show that the proposed two-stage denoising scheme achieves more competitive performance than the state-of-the-art denoising algorithms, especially in preserving image details and introducing very few artifacts.