Semi-Split Bregman Iteration Algorithm for Image Denoising
首发时间:2010-12-31
Abstract:The split Bregman iteration has been demonstrated to be an efficient tool for solving total variation regularized minimization problems. In denosing case, it can remove noise efficiently, but it can not preserve textures well. In this paper, we analyze the split Bregman method from the perspective of function matching, and reveal the reason why it can not preserve textures well. Based on this analysis, we develop a new method called the semi-split Bregman iteration algorithm for image denoising. The numerical results show that the semi-split Bregman iteration algorithm can preserve the textures and improve the peak signal to noise ratio efficiently in the processing of denoising.
keywords: image denoising texture preserving split Bregman iteration
点击查看论文中文信息
图像去噪的半分裂Bregman迭代法
摘要:分裂Bregman迭代法是求解基于总变分正则化极小化问题的一种有效算法。在图像去噪中,这种方法能有效地去除噪声,然而该方法对于纹理的保持并不好。本文从函数匹配的角度分析了分裂Bregman迭代,找出了该方法不能较好保持纹理的原因,并且提出了一种半分裂Bregman迭代算法。数值试验表明,半分裂Bregman迭代算法可以在图像去噪过程中能较好地保持纹理并有效提高图像的峰值信噪比。
关键词: 图像去噪 纹理保持 分裂Bregman迭代
基金:
论文图表:
引用
No.4401265509979129****
同行评议
共计0人参与
勘误表
图像去噪的半分裂Bregman迭代法
评论
全部评论0/1000