计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (7): 192-194.DOI: 10.3778/j.issn.1002-8331.2009.07.058

• 图形、图像、模式识别 • 上一篇    下一篇

改进TV复原模型的彩色-灰度图像变换方法

郭彦伶,彭进业,王大凯   

  1. 西北大学 信息科学与技术学院,西安 710127
  • 收稿日期:2008-08-25 修回日期:2008-10-29 出版日期:2009-03-01 发布日期:2009-03-01
  • 通讯作者: 郭彦伶

Transformation of color image to gray image based on improved total variation restoration model

GUO Yan-ling,PENG Jin-ye,WANG Da-kai   

  1. School of Information Science and Technology,Northwest University,Xi’an 710127,China
  • Received:2008-08-25 Revised:2008-10-29 Online:2009-03-01 Published:2009-03-01
  • Contact: GUO Yan-ling

摘要: 针对彩色图像到灰度图像的变换这一问题,提出了一种新的基于TV(全变分)复原模型的新方法。利用Sapiro等所提出的矢量图像的水平集概念,结合全变分复原,建立新的变换模型,实现算法。实验表明,所提方法实现的彩色-灰度图像变换,不仅很好地解决了原来传统变换不能保持形状的问题,并且能解决Sapiro等所提出方法的边缘模糊化缺点,使变换后图像的边缘锐利程度得到了较大的提高。

Abstract: An improved method for shape preserving transformation of color image to gray image is proposed,based on combination of the concept of level-lines of vector-valued images originated by Sapiro and the total variation restoration model originated by Rudin,Osher and Fatemi.The theoretical analysis for the proposal is given.Experimental results show that the proposed method produces gray image with edge sharpness much better than that of the original Sapiro’s method.