计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (20): 198-201.DOI: 10.3778/j.issn.1002-8331.2008.20.060

• 图形、图像、模式识别 • 上一篇    下一篇

基于偏微分方程的图像去噪综合模型

杜宏伟   

  1. 云南机电职业技术学院,昆明 650203
  • 收稿日期:2007-10-10 修回日期:2008-01-17 出版日期:2008-07-11 发布日期:2008-07-11
  • 通讯作者: 杜宏伟

Combined model for image denoising based on partial differential equations

DU Hong-wei   

  1. Yunnan Vocational College of Mechanical & Electrical Technology,Kunming 650203,China
  • Received:2007-10-10 Revised:2008-01-17 Online:2008-07-11 Published:2008-07-11
  • Contact: DU Hong-wei

摘要: 介绍了基于偏微分方程(PDE)的两种去噪模型,即ROF模型和LLT模型。根据对这两种模型的比较,提出了应用权函数来合并ROF模型和LLT模型的综合模型。实验表明,综合模型既能克服ROF模型和LLT模型的缺点,又能融合它们的优点,在去噪、保护平滑区域、保护边缘和纹理细节方面都有较好的表现。

关键词: 图像去噪, 偏微分方程, ROF模型, LLT模型, 综合模型

Abstract: The paper introduces two kinds of model for denoising based on partial differential equations,called the ROF model and the LLT model.According to the comparison of the two models,this paper proposes a combined model which combines ROF model and LLT model by weight function.The experiment shows that the combined model not only can overcome the shortcomings of ROF model and LLT model,but also can integrate their advantages.It has better performance in denoising,protecting smoothing region,protecting the edge and texture details.

Key words: image denoising, partial differential equation, ROF model, LLT model, combined model