摘要
针对多模图像融合问题提出了一种基于小波变换的新方法。将小波低频系数乘以加权因子1/R,减少低频部分所占整个图像的信息比例,并采取绝对值取大的融合规则选取小波低频系数;使用由方差和平均梯度构造的新的评价因子——小波邻域信息量作为融合规则选取小波高频系数。实验结果表明,该方法得到的融合图像体现出更强的融合性能。
This paper proposes an image fusion algorithm based on wavelet transform, it is mainly concerned with the CT and MRI. For the image low-frequency part, wavelet low-frequency coefficients will multiply 1/R, to reduce the proportion of low-frequency part of the whole image information; For the image high-frequency part, use the new evaluation factor - wavelet neighborhood information to choose the ultimate wavelet high-frequency coefficients. Compared with several commonly used image fusion methods, it shows that the algorithm this article proposed can give us the fusion image that ownes more image detail information.
出处
《微型机与应用》
2010年第4期62-64,共3页
Microcomputer & Its Applications
关键词
图像融合
多模图像
小波变换
邻域信息量
image fusion
multi-mode image
wavelet transform
neighborhood information