吉林大学学报(理学版)

• 计算机科学 • 上一篇    下一篇

一种基于全变差正则化与小波包变换的图像去噪算法

左平1, 王洋2,3, 申延成1   

  1. 1. 空军航空大学 基础部, 长春 130022; 2. 吉林大学 计算机科学与技术学院, 长春 130012;3. 吉林大学 公共计算机教学与研究中心, 长春 130012
  • 收稿日期:2013-01-19 出版日期:2014-01-26 发布日期:2014-03-05
  • 通讯作者: 王洋 E-mail:wyangjlu@mail.jlu.edu.cn

Image Denoising Algorithm Based on Wavelet PacketTransform and Total Variation Model

ZUO Ping1, WANG Yang2,3, SHEN Yan-cheng1   

  1. 1. Department of Foundation, Aviation University of Air Force, Changchun 130022, China;2. College of Computer Science and Technology, Jilin University, Changchun 130012, China;3. Public Computer Teaching and Research Center, Jilin University, Changchun 130012, China
  • Received:2013-01-19 Online:2014-01-26 Published:2014-03-05
  • Contact: WANG Yang E-mail:wyangjlu@mail.jlu.edu.cn

摘要:

提出一种基于全变差(TV)模型和小波包变换的图像去噪算法, 并给出了针对该模型的一种改进正则化参数选取方法, 改善了全变差模型去噪中出现的块效应问题, 同时保留了图像中的边缘信息. 数值实验表明, 用所给算法去噪可得到较高的峰值信噪比和较好的视觉效果.

关键词: 图像处理, 图像去噪, 全变差模型, 小波包变换

Abstract:

Authors proposed an effcient image denoising method based on the combination of wavelet packet transform with  total variation model and presented
how to select the regularization parameter in this model. The combination of wavelet packet transform with total variation model helps to alleviate staircase effect efficiently and  preserve sharp discontinuities in images as well. The numerical experimental results show that the new method is effective in removing Gaussian noise and keep the detail of the image  well.

Key words:  image processing, image denoising, total variation model, wavelet packet transform

中图分类号: 

  • TP391