基于无下采样Contourlet变换的图像融合  

Image Fusion Based on nonsubsampled contourlet transform

作　　者：吴一全[1] 陈飒[1] 罗子娟[1] 

机构地区：[1]南京航空航天大学信息科学与技术学院,南京210016

出　　处：《中国体视学与图像分析》2008年第2期111-115,共5页Chinese Journal of Stereology and Image Analysis

摘　　要：无下采样Contourlet变换(nonsubsampled contourlet transform,NSCT)是基于Contourlet变换的一种扩展,可以对图像进行灵活的多尺度、多方向和平移不变性分解。本文提出了一种基于NSCT的图像融合算法。首先对图像进行NSCT变换;然后对变换得到的低频和高频分量系数分别采用三种不同的融合规则进行融合处理;最后重构图像得到融合结果。文中给出了实验结果,通过对结果的分析比较,表明该方法的性能较传统的基于小波变换或基于Contourlet变换的融合方法有了一定的提高,融合效果更好。Nonsubsampled contourlet transform （NSCT） is a new extension to contourlet transform. It is a flexible multistage, multidirectional and shift invariant. In this paper, an image fusion method based on NSCT is proposed, itial images are decomposed by NSCT. Then the NSCT coefficients are fused by choosing different rules which correspond to high and low frequency bands. Finally, the fused coefficients are reconstructed to obtain fusion results. The experimental results are given, and compared with those of the methods based on wavelet transform or contourlet transform. Besides, the experiments show that this approach can achieve better results than the methods based on wavelet transform or contourlet transform.

关 键 词：图像融合 无下采样Contourlet变换(NSCT) CONTOURLET变换 小波变换 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]