电子学报 ›› 2012, Vol. 40 ›› Issue (3): 477-482.DOI: 10.3969/j.issn.0372-2112.2012.03.011

• 学术论文 • 上一篇    下一篇

去除乘性噪声的小波变分模型

李亚峰1,2, 冯象初1   

  1. 1. 西安电子科技大学应用数学系,陕西西安 710071;2. 宝鸡文理学院计算机科学系,陕西宝鸡 721016
  • 收稿日期:2010-06-19 修回日期:2011-09-12 出版日期:2012-03-25
    • 基金资助:
    • 国家自然科学基金 (No.61105011,No.11101292,No.61001156); 教育部科学技术研究重点项目 (No.211155); 宝鸡文理学院2010年院级科研重点项目 (No.ZK10171)

Wavelet Based Variational Model to Remove Multiplicative Noise

LI Ya-feng1,2, FENG Xiang-chu1   

  1. 1. Department of Applied Mathematics,Xidian University,Xi'an,Shaanxi 721016,China;2. Department of Computer Science,Baoji University of Arts and Science,Baoji,Shaanxi 721016,China
  • Received:2010-06-19 Revised:2011-09-12 Online:2012-03-25 Published:2012-03-25

摘要: 基于一个凸的全变差变分能量泛函,本文提出一个去除乘性噪声的小波变分模型.使用凸优化工具,给出该模型的交替分裂Bregman算法.相比于目前基于全变差去除乘性噪声的变分模型和分裂算法,新模型和算法形式简单,推广性强.一系列实验结果验证了本文算法及推广算法的有效性.

关键词: 乘性噪声, Besov空间, 小波, 曲线波, 变分模型

Abstract: Based on a convex total variation (TV) variational energy function,this paper proposes a wavelet based variational model to remove multiplicative noise.Using convex optimization tool,we give an alternating split Bregman algorithm to solve the proposed model.Compared with existing TV based method,the proposed algorithm is simple and can be extended easily.Numerical experiments are presented to demonstrate the efficient performance of the proposed algorithm and its extensions.

Key words: multiplicative noise, besov space, wavelet, curvelet, variational model

中图分类号: