一种基于二维EMD的图像融合方法  

A method of image fusion based on the two-dimensional empirical mode decomposition

作　　者：赵春晖[1] 任龙涛[1] 万建[1] 

机构地区：[1]哈尔滨工程大学信息与通信工程学院,黑龙江哈尔滨150001

出　　处：《应用科技》2009年第9期15-19,共5页Applied Science and Technology

基　　金：国家自然科学基金资助项目(60672034);高等学校博士学科点基金资助项目(20060217021);黑龙江省自然科学基金资助项目(ZJG0606-01)

摘　　要：图像融合是图像处理中的一个重要内容,常用的小波图像融合主要是基于像素级融合,这种方法容易失去局部特征相关性较强的特性,融合后会出现局部斑点现象.该文应用二维EMD方法在像素级也进行了讨论,并且针对EMD分解的特殊性,提出了一种用二维EMD(two-dimensional EMD)进行图像融合的方法,采用对IMF(intrinsic model function)分量在对应频率段上进行线性加权融合,提出了几种常用计算加权系数的方法.通过实验分析和性能评价表明,基于二维EMD的图像融合较小波图像融合效果较好,更能提取图像细节.Image fusion is an important part of the image processing. The wavelet image fusion is based on pixel, which misses local correlations easily and leads to local speckle. That method mentioned above is also presented with the two-dimensional EMD. According to the characteristic of two-dimensional EMD, a linear weighted method is proposed. And several methods of computing the weighted coefficients are presented. The experimental result shows that the image fusion based on the two-dimensional EMD has some advantages over the wavelet image fusion, which can extract image details better.

关 键 词：二维EMD IMF 图像融合 线性加权 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]