计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (12): 189-191.

• 图形图像、模式识别 • 上一篇    下一篇

基于无下采样Contourlet变换的图像融合算法

王连庆,安锦文   

  1. 西北工业大学 自动化学院,西安 710072
  • 收稿日期:2007-08-10 修回日期:2007-11-21 出版日期:2008-04-21 发布日期:2008-04-21
  • 通讯作者: 王连庆

Image fusion based on nonsubsampled Contourlet transform

WANG Lian-qing,AN Jin-wen   

  1. Department of Automatic Control,Northwestern Polytechnical University,Xi’an 710072,China
  • Received:2007-08-10 Revised:2007-11-21 Online:2008-04-21 Published:2008-04-21
  • Contact: WANG Lian-qing

摘要: 针对目前最新发展的无下采样Contourlet变换,同时具有方向性、各向异性和平移不变性的特点,提出了一种新的基于无下采样Contourlet变换图像融合算法。首先对原始图像进行无下采样的Contourlet变换,得到高频和低频图像系数;然后根据它们的系数特性,采用一种新的加权融合规则对其进行融合。实验证明,此方法相对于传统的变换法,取得更优的融合结果。

关键词: 图像融合, 无下采样Contourlet变换, 融合规则

Abstract: A novel method of image fusion based on the nonsubsampled Contourlet transform is proposed,based on the recently developed nonsubsampled Contourlet transform,which can offer a trait of multidirection,flexible multiscale and shift-invariant simultaneity.At first,obtain the high and low frequency-coefficients by decomposing the image with the nonsubsampled Contourlet transformation.Then,according to the coefficient characters between the high and low frequency-coefficients,syncretize them by using new fusion algorithms of weighted average.The experimental results show that the new method makes a better fusion result compared with the traditional transform methods.

Key words: image fusion, nonsubsampled Contourlet transform, fusion algorithm