摘要
为了进一步改善医学图像融合质量,提出一种基于NSCT(非下采样Contourlet变换)与区域点火PCNN(脉冲耦合神经网络)的医学图像融合方法。该方法在低频子带系数采用基于区域点火PCNN的融合规则,应用PCNN改进的简化模型,将低频子带系数作为信号激励PCNN网络,利用点火区域强度分析区域点火特性,根据区域点火特性确定低频子带融合系数;在选择带通方向子带系数时,充分利用非下采样Contourlet变换的方向特性,采用改进的拉普拉斯能量作为带通方向子带系数的融合规则。实验结果表明,该方法与传统融合方法相比,能够较好的保留图像的边缘和过渡区域信息,大幅度提高融合图像的质量。
In order to further improve the quality of medical image fusion, a novel method for medical image fusion was proposed based on NSCT （nonsubsampled contourlet transform） and regional fire PCNN （pulse coupled neural network）. A fusion rule based on regional fire PCNN was adopted in low frequency sub-band coefficient. Firstly, the low frequency sub-band coefficient was input into PCNN network as the signal, through the application of the improved PCNN simplied model. Secondly, characteristics of regional fire was analysed by the regional fire intensity. Finally, the low frequency sub-band coefficient was determined by characteristics of regional fire,＂ When choosing the bandpass directional sub-band coefficients, directional characteristics of NSCT was made best use of. A fusion rule based on improved energy of Laplacian was proposed in bandpass directional sub-band cosfficients. The experiment results show that the proposed method can greatly improve the quality of fusion image compared with traditional fusion methods with abundant edge and transition information.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2014年第2期274-278,共5页
Journal of System Simulation
基金
国家自然科学基金资助项目(60962004
61162016)
