基于多尺度变换的直觉模糊推理图像融合算法  

Image Fusion Algorithm Based on Intuitionistic Fuzzy Inference of Multiscale Transform

作　　者：那彦[1] 刘波[1] 

机构地区：[1]西安电子科技大学电子工程学院,陕西西安710071

出　　处：《电子科技》2015年第8期112-114,118,共4页Electronic Science and Technology

摘　　要：提出了基于多尺度变换的直觉模糊推理医学图像融合方法,针对4种多尺度变换对图像融合的影响进行了分析。文中融合算法是利用4种多尺度变换对待融合图像进行分解,并对得到的低频分量和高频分量采用直觉模糊推理融合规则进行处理,最后将融合后的低频分量和高频分量经逆变换得到融合图像。实验结果表明本文算法优于传统的模糊推理图像融合算法,并且在变换域中,Contourlet变换下的融合结果要优于其它3种变换。This paper presents the multi-scale transform of intuitionistic fuzzy fusion method based on medical image, and the effect of four kinds of multi-scale transforms on the proposed algorithm is as following： first, performing the transform and Shearlet transform; then fusing the low frequency image fusion are analyzed. The brief procedures of Contourlet transform, NSCT transform, Waveletlet component and high frequency component by intu- itionistic fuzzy fusion method; and finally performing converse transform to obtain the finally fusion image. The ex- perimental results show that the proposed algorithm is superior to the traditional fuzzy fusion algorithm, and that in the transform domain, the fusion results of Contourlet transform are better than those of the other three transforms.

关 键 词：多尺度变换 直觉模糊推理系统 图像融合 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]