基于非抽样Contourlet变换的医学图像融合算法  

Fusion of medical images based on Nonsubsampled Contourlet Transform

作　　者：何国栋[1] 于梅[1] 殷兵[1] 梁栋[1] 

机构地区：[1]安徽大学计算智能与信号处理教育部重点实验室,合肥230039

出　　处：《计算机工程与应用》2008年第26期186-187,220,共3页Computer Engineering and Applications

基　　金：国家自然科学基金No.60772121;安徽省自然科学基金No.070412065;安徽省高等学校自然科学研究项目(No.2005KJ005ZD);安徽大学211工程学术创新团队资助~~

摘　　要：分析了非抽样Contourlet变换(Nonsubsampled Contourlet Transform,NSCT)的原理,提出了一种新的基于NSCT的医学图像融合算法,应用NSCT对CT和MRI图像进行多尺度、多方向分解,低频子带采取区域能量加权法融合,带通子带采取模最大融合,最后将融合的系数进行NSCT逆变换得到融合图像。实验表明,与其它融合算法比较,该算法融合图像效果较好。The principle of Nonsubsampled Contourlet Transform（NSCT） is studied.A novel fusion algorithm is proposed based on NSCT.Firstly,the CT and MRI images are decomposed at different scales and directions,the low frequency subband coefficients and the the bandpass subband coefficients are obtained.Secondly,different fusion rules are applied in different subband coefficients.The regional energy fusion rule is applied in low frequency subband coefficients and the maximum module rule is applied in the bandpass subband coefficients.Finally,the fused image is reconstructed by the inverse NSCT.The experimental results show that the qualitative of the novel algorithm is better than other fusion algorithms.

关 键 词：图像融合 CONTOURLET变换 非抽样CONTOURLET变换 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]