.png){kind=icon}
返回
PROXIMAL LINEARIZED ALTERNATING DIRECTION METHOD FOR MULTIPLICATIVE DENOISING
.png){kind=icon}
PRE
AI
DOI：10.1137/11083811X.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Multiplicative noise naturally appears in coherent imaging systems, such as synthetic aperture radar. Due to the edge preserving feature of total variation (TV), variational models with TV regularization show good performance for multiplicative denoising. Recently proposed augmented Lagrangian frameworks with the Gauss-Seidel method are efficient for solving multiplicative denoising variational models with TV. But recent algorithms based on the augmented Lagrangian require inner iterations or an inverse involving the Laplacian operator at each iteration. In this paper, we propose efficient algorithms for TV regularized convex variational models. Our proposed algorithms do not require any inner iteration or the inverse involving the Laplacian operator by using a linearization scheme and a proximal function. We establish global convergence under a Lipschitz continuous assumption on the gradient of the fidelity term when a correction step is used. Numerical results show that our proposed algorithms overall outperform the augmented Lagrangian based state-of-the-art algorithms for multiplicative denoising problems.
Keyword:
multiplicative noise
speckle
synthetic aperture radar
denoising
alternating minimization
proximal
linearization
total variation
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
2.6
论文数:
5.1K
被引数:
1.8W
机构
引用论文
.png){kind=icon}
First Electrical Characterization of Prototype 600 A HTS Twisted-pair Cables at Different Temperatures首次对不同温度下原型600安培高温超导扭绞对电缆的电学特性表征
OAAI
OAAI分享收藏
查看更多内容