计算机应用 ›› 2013, Vol. 33 ›› Issue (10): 2931-2934.

• 多媒体技术 • 上一篇    下一篇

基于可变指数及L1保真项的图像去噪算法

耿海,何小卫,樊骏笠   

  1. 浙江师范大学 数理与信息工程学院,浙江 金华 321004
  • 收稿日期:2013-04-26 修回日期:2013-06-17 出版日期:2013-10-01 发布日期:2013-11-01
  • 通讯作者: 何小卫
  • 作者简介: 
    耿海(1988-),男,江苏南京人,硕士研究生,主要研究方向:图像处理、图像修复;何小卫(1968-),男,浙江金华人,副教授,主要研究方向:图像处理、人工智能;樊骏笠(1990-),男,浙江绍兴人,硕士研究生,主要研究方向:图像处理、图像修复。

Image denoising algorithm with variable exponent regularization and L1 fidelity

GENG Hai,HE Xiaowei,FAN Junli   

  1. College of Mathematics, Physics and Information Engineering, Zhejiang Normal University, Jinhua Zhejiang 321004, China
  • Received:2013-04-26 Revised:2013-06-17 Online:2013-11-01 Published:2013-10-01
  • Contact: HE Xiaowei

摘要: 全变分(TV)模型采用了梯度的1范数作为正则化约束, 它能够沿着梯度方向较好地保护图像的边缘信息,但在图像较均匀区域,容易产生“阶梯”效应。利用梯度的可变指数函数作为正则化项,提出TV模型的改进模型, 该模型既保持TV模型保护图像边缘信息的优点,又可以明显地减少非边界区域“阶梯”效应的产生,同时把〖WTHX〗u-〖WTHX〗f的1范数作为数据保真项增强了模型修复图像破损部分的能力

关键词: 图像去噪, TV模型, 可变指数函数, 正则化, L1范数

Abstract: The L1 norm of gradient is used as the regularization term in the Total Variation (TV) model which can preserve the edges of the image well. However, it has the staircasing effect in the relatively smooth regions. Using the variable exponent function as the regularization term, the modified model can not only preserve the edges of image as well as the TV model but also decrease the staircasing effect obviously. Simultaneously, the L1 norm of 〖WTHX〗u-〖WTHX〗f was regarded as the fidelity term of the model, which can enhance the ability of image denoising.

Key words: image denoising, total variation(TV) model, variable exponent functional, regularization term, L1 norm

中图分类号: