期刊文献+

一种改进的全变分(TV)修补模型 被引量:1

An improved TV inpainting model
下载PDF
导出
摘要 目的提出一种全变分(TV)修补模型的改进方案,而且具有良好的边缘特性,并弥补原有的TV修补模型的不能满足连通性原则的缺陷。方法定义在图像修补区定义一种"加权全变分,即沿修补区边界法线方向的梯度分量对全变分的贡献,远大于沿切线方向的梯度分量的贡献。结果通过对相同的受损图像,采用原有模型和改进模型作比对实验表明,文中的改进模型可以完全满足图像修补的连通性原则。结论改进的TV模型较原始TV模型更适合于非纹理图像修复。 Aim To propose an improved TV inpainting model. The improved one has favorable marginal property in order to make up the limitation of the TV inpainting model which can not meet the connection principle. Methods A weighted total variation is defined in the inpainting domain, which means the contribution of gradient along the tangent direction of the inpainting domain is siginificantly larger than that of the gradient along the norm direction. Results The experiments which compare the results of adopting the original and improved model to the same damaged Image show that the improved model can meet the connection principle. Conclusion The improved model is more suitable to the non texture image inpainting.
出处 《西北大学学报(自然科学版)》 CAS CSCD 北大核心 2009年第6期948-951,959,共5页 Journal of Northwest University(Natural Science Edition)
基金 教育部新世纪优秀人才支持基金资助项目(NCET-07-0693)
关键词 图像修补 全变分(TV)模型 连通性原则 权因子 image inpainting total variation model connection principle weighted factor
  • 相关文献

参考文献6

  • 1BERTALMIO M, SAPIRO G, CASELLES V, et al. Image inpainting[ M ]//In : Proceedings of International Conference on Computer Graphics and Interactive Techniques. Us, New Orleans : Louis Iana,2000:417-424.
  • 2CHAN T F, SHEN J. Non-texture inpainting by curcaturedriven diffusion ( CDD ) [ J ]. J of Visual Communication and Image Representation,2001,12 (4) :436-449.
  • 3CHANT F, SHEN J. Mathematical model for local non-texture inpaintings [ J ]. SIAM J Appl Math,2002,62 (3) : 1 019-10 43.
  • 4RUDIN L, OSHER S, FATEMI E. Nonlinear total variation based noise removal algorithms [ J ]. Physica D, 1992, 60:259-268.
  • 5侯榆青,张欢,史晶,张玲艳.一种改进的全变分自适应图像去噪模型[J].西北大学学报(自然科学版),2008,38(3):371-373. 被引量:10
  • 6冈萨雷斯.数字图像处理[M].2版.北京:电子工业出版社,2003.27-30,54,59-81,117-132,276-295,209.

二级参考文献7

  • 1赵健,周明全,谢端,卜起荣.SAR图像小波域消噪方法[J].西北大学学报(自然科学版),2005,35(1):17-20. 被引量:3
  • 2张红英,彭启琮.全变分自适应图像去噪模型[J].光电工程,2006,33(3):50-53. 被引量:45
  • 3RUDIN L, OSHER S, FATEMI E. Nonlinear total variation based noise removal algorithm [ J ]. Physica D, 1992, 60:259-268.
  • 4RUDIN L, OSHER S. Total Variation Based Image Restoration with Free Local Constraints [ C ]. Austin: Proc 1 st IEEE Int'l Conf on Image Processing, 1994, ( 1 ) :31-35.
  • 5BLOMGREN P V. Color TV: Total Variation Method for Restoration of Vector-Valued Images [ D ]. Angeles: UCLA i 1998.
  • 6YUNMEI C, LEVINE S, MURALI R. Variable exponent, linear growth functionals in image restoration [ J ]. SIAM Applied Mathematics, 2006, 66(4) : 1 383-1 406.
  • 7VOGEL C R, OMAN M E, Iterative methods for total variation denoising [ J ], SIAM Sci Comput, 1996, 17 : 227- 238

共引文献30

同被引文献6

引证文献1

二级引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部