Alpha稳态噪声下基于Meridian范数的全变分图像去噪算法

摘要

在实际应用中,噪声不可避免,因此,图像去噪一直是图像处理领域研究的重点,并且近年来受到越来越多的研究者的青睐.该文首先基于Meridian分布和全变分(Total Variational,TV)的统计特性,提出一种全变分模型来复原alpha稳态噪声环境下的含噪声图像.此外,为了保证模型解的唯一性,对提出的全变分模型添加了一个二次惩罚项,得到一个严格凸的全变分模型,然后,使用原始-对偶算法对提出的全变分模型进行求解,并证明了该算法的收敛性.最后,进行了仿真实验,并对实验结果进行了分析,实验结果验证了提出模型的可行性与有效性.%In actual applications, noises may inevitably exist, and thus to study the denoising method for images is great significant task in image processing filed that attracts much attention in recent years. In this paper, based on the statistical property of Meridian distributed and the Total Variational (TV), a total variational method is proposed for restoring images degraded by alpha-stable noise. Besides, in order to obtain a strictly convex model, a quadratic penalty term is added, which guarantees the uniqueness of the solution. For solving the novel convex variational model, a primal-dual algorithm is employed to solve the above model, and the convergence of the algorithm is proved. The experimental results demonstrate that the feasibility and effectiveness of the proposed model for the noisy images with alpha-stable noise.