计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (21): 106-108.DOI: 10.3778/j.issn.1002-8331.2009.21.031

• 图形、图像、模式识别 • 上一篇    下一篇

小波的循环阈值去噪方法

余楚才1,陈怀新2,李 雷1   

  1. 1.电子科技大学 电子工程学院,成都 610054
    2.中国电子科技集团公司 第十研究所(中国西南电子技术研究所),成都 610036
  • 收稿日期:2008-04-24 修回日期:2008-07-23 出版日期:2009-07-21 发布日期:2009-07-21
  • 通讯作者: 余楚才

Circulating threshold denoising method based on wavelet

YU Chu-cai1,CHEN Huai-xin2,LI Lei1   

  1. 1.University of Electronic Science and Technology of China,Chengdu 610054,China
    2.Southwest China Institute of Electronic Technology,Chengdu 610036,China
  • Received:2008-04-24 Revised:2008-07-23 Online:2009-07-21 Published:2009-07-21
  • Contact: YU Chu-cai

摘要: 在D.L.Donoho提出的多分辨分析小波阈值去噪方法的基础上,针对SAR图像的相干斑噪声现象,提出一种小波域SAR图像的循环阈值去噪法,通过估计的噪声标准差和滤除信号后的噪声标准差之间的循环迭代,得到最优阈值。并以峰值信噪比和平滑指数等作为评价指标,与传统的软阈值、硬阈值以及几种空域滤波方法进行了比较。仿真结果表明,提出的新的阈值去噪算法无论是峰值信噪比,还是主观质量都优于软阈值、硬阈值去噪算法和各种空域滤波方法。

关键词: 小波变换, 软阈值函数, 硬阈值函数, 斑点噪声, 峰值信噪比

Abstract: Based on the multi-analysis wavelet threshold denoising method which is put forward by D.L.Donoho,a new wavelet threshold denoising method for SAR image is proposed aiming at speckle noise.The paper makes a comparison study among the traditional soft threshold,hard threshold and several kinds of spacial denoising methods by the performance evaluation criteria of PSNR and smooth index.Simulation results show that the proposed denoising algorithm owns better visual effect and PSNR performance than the traditional soft threshold,hard threshold and several kinds of spacial denoising methods.

Key words: wavelet transform, soft threshold function, hard threshold function, speckle noise, Peak Signal to Noise Ratio(PSNR)