NSST域内结合UDWT与PCNN医学图像融合算法

为了进一步突出医学融合图像的细节信息,提升 清晰度,本文提出NSST域内结合UDWT与PCNN医学图 像融合算法。首先,将两幅源图像分别通过NSST 进行分解,获得相应的低频和高频子带。 在低频融合规则中, 采用UDWT将低频子带进一步分解为能量子带与细节子带,进一步利用PCNN融合这两幅源图 像对应的低频能量 子带；利用区域能量和融合这两幅源图像的低频细节子带,再应用逆UDWT融合低频细节子 带和能量子带。其次, 在高频融合规则中,采取UDWT分别将A和B两幅源图像对应的高频子带进一步分解为高频 能量子带与高频细节 子带,再根据拉普拉斯能量和与区域能量和的组合,获取融合后的高频子带。最后,利用逆 NSST获取融合图像。 实验证明,本文提出的算法与现有主流算法相比,实验结果在视觉效果和客观指标方面均具 有较大优势。

Medical image fusion algorithm based on UDWT and PCNN in NSST domain

In order to further highlight the details of medical image fusion and improve the clarity,this paper proposes a medical image fusion algorithm based on UDWT and PCNN in NSST domain.In this alg orithm,the source image is decomposed by NSST to obtain low-frequency and high-frequency subband coeffici ents.In the low frequency fusion rules, the low frequency sub-band is further decomposed into energy sub-band and deta il sub-band by UDWT,and the low frequency energy sub-band corresponding to the two source images is fused by PC NN; the low frequency detail sub-band of the two source images is fused by region energy and the low frequency detail sub -band of the fusion source image,and then the inverse UDWT is applied to fuse the low frequency detail sub-band and energ y quantum band.Secondly,in the rules of high-frequency fusion,the high-frequency sub-band corresponding to A and B s ource images is further decomposed into high-frequency energy sub-band and high-frequency detail sub-band by UDWT,a nd then the fused high-frequency sub-band is obtained according to the combination of Laplace energy and regiona l energy.Finally,the fusion sub-band is obtained by inverse NSST.Experimental results show that the proposed algorithm has greater advantages in visual effect and objective indicators compared with the existing mainstream algorithms.