Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (16): 182-184.DOI: 10.3778/j.issn.1002-8331.2010.16.053

• 图形、图像、模式识别 • Previous Articles     Next Articles

Novel fusion method of infrared and color visible images using BEMD

ZHANG Xiu-qiong1,2,CHEN Rong1,YANG Jun1   

  1. 1.Institute of Image & Graph,Computer College,Sichuan University,Chengdu 610064,China
    2.Department of Computer Science and Information Engineering,Leshan Teacher’s College,Leshan,Sichuan 614004,China
  • Received:2008-11-24 Revised:2009-02-11 Online:2010-06-01 Published:2010-06-01
  • Contact: ZHANG Xiu-qiong


张秀琼1,2,陈 蓉1,杨 军1   

  1. 1.四川大学 计算机学院 图像图形研究所,成都 610064
    2.乐山师范学院 计算机科学与信息工程系,四川 乐山 614004
  • 通讯作者: 张秀琼

Abstract: A novel color fusion method is proposed for the infrared and visible images using the Bidimentional Empirical Mode Decomposition(BEMD).Firstly,the source images are decomposed into a set of Intrinsic Mode Functions(IMFs) and the residue by BEMD.Then,the range of pixel value of every IMF of the infrared image is adjusted to meet the range of the visible image.The fused IMFs and the residue are obtained by the weighted averaging and which are reconstructed into a single fused image by BEMD at last.Using this method,the fused image not only is enhanced in detail but also has color information similar to the source color visible image.In the experiment,compared with the DWT,PCA and EMD,CEMD,the results show that the proposed method is the best.

Key words: image fusion, Empirical Mode Decomposition(EMD), Bidimentional Empirical Mode Decomposition(BEMD), Intrinsic Mode Function(IMF)

摘要: 针对红外和彩色可见光图像的融合,提出了一种基于二维经验模式分解(BEMD)的新的融合方法。源图像由BEMD分解成为本征模式函数集(IMFs)和残余;再将红外图像的IMFs和残余分别与对应可见光图像的IMFs和残余进行灰度范围匹配后,用加权平均方法进行融合;最后由BEMD重构成融合图像。用此法所得的融合图像增强了红外图像的细节并具有与可见光图像相似的自然色彩。实验中将此法与传统的小波变换方法和主成分分析方法进行了比较,还与经验模式分解(EMD)和复经验模式分解(CEMD)的方法进行了比较,实验结果都证明了该方法的融合效果最优。

关键词: 图像融合, 经验模式分解, 二维经验模式分解, 本征模式函数

CLC Number: