Alpha稳态噪声下基于Meridian范数的全变分图像去噪算法

发布时间：2018-05-12 03:21

  本文选题：图像处理 + Meridian范数　； 参考：《电子与信息学报》2017年05期

【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。
[Abstract]:In practical application, noise is inevitable, so image denoising has been the focus of image processing research, and has been more and more popular in recent years. Based on the statistical properties of Meridian distribution and Total Variational TVs, a total variational model is proposed to restore noisy images in alpha stationary noise environment. In addition, in order to ensure the uniqueness of the solution of the model, a quadratic penalty term is added to the proposed total variational model, and a strictly convex total variational model is obtained. Then, the primal-dual algorithm is used to solve the proposed total variational model. The convergence of the algorithm is proved. Finally, the simulation experiment is carried out, and the experimental results are analyzed. The experimental results verify the feasibility and validity of the proposed model.
【作者单位】： 南京邮电大学视觉认知计算与应用研究中心;南京邮电大学宽带无线通信与传感网技术教育部重点实验室;
【基金】：国家自然科学基金(61501251,61271335,61271240) 江苏省自然科学基金项目(BK20140891) 南京邮电大学引进人才科研启动基金资助项目(NY214191)~~
【分类号】：TP391.41
，

本文编号：1876899