【摘要】 稀疏表示和非局部相似性在图像去噪领域扮演着越来越重要的角色,并且取得了很好的去噪效果.解决了高斯噪声和椒盐噪声混合的图像去噪问题.在去噪过程中,如何更好地保留图像中原有的边缘信息是一个很重要的问题.为此提出了在稀疏表示和非局部相似性的基础上,引入Sobel算子的算法.实验结果表明,该算法去噪效果突出,并且能够在去噪的同时保留图像的边缘信息,为去除图像中的混合噪声提供了一种有效的方法.更多还原

【Abstract】 Sparse representation and nonlocal self-similarity play an important role and show better results in image denoising. This paper solved the mixed noise which included Gauss noise and salt-and-pepper noise. How to better retain the edge information is a very important issue in the process of denoising. And proposed the algorithm which combines sparse representation,nonlocal self-similarity and Sobel operator. The experimental results showed that the proposed algorithm was effective and could preserve the edge information. It provided an effective method to remove the mixed noise in the image.更多还原