基于人眼视觉特性与自适应PCNN的医学图像融合算法

针对多尺度变换的图像特征,提出了一种基于人眼 视觉特性与自适应脉冲耦合神经网络(PCNN)的医学图像融合新方法。首先,对经配准的源图 像进行非下采样Contourlet变换(NSCT), 得到低频、高频子带系数；然后,考虑到低频子带系数中保留了绝大部分源图像能量和图像 轮廓特征,提出 区域能量(RE)和梯度奇异值度量(GSVM)相结合的方法；考虑到图像全局特 征,将PCNN用于高频子带系数中,提出区域视觉对比度(SLVC )模拟人眼视觉特性作为PCNN的 外部刺激输入,设定PCNN的链接强度随视觉对比敏感度(VCS) 自适应变化,同时考虑到PCNN的迭 代次数,利用Sigmoid函数计算其点火输出幅值的显著性度量；最后,对获得的融合系数进 行逆NSCT得到融合图像。通过实验对比分析表明,本文算法不仅可以保留源图像信息的同时 ,还得到较好的客观评价指标和视觉效果。

Medical image fusion algorithm based on human visual features and adaptive PCNN

According to the characterist ics of multi-scale transform,a novel medical image fusion algorithm based on human visual features and adaptive pulse coupled neural network (PCNN) is pro posed.Firstly,source images after registration are decomposed into low and high frequency sub-bands by nonsubsampled contourlet t ransform (NSCT).Secondly,majority ener gy and characteristics of the source image is retained in the low frequency sub-bands ,a fusion rule based on region energy (RE) combined with gradient singular value measurement (GSVM) is adopted.Moreover, considering the problem of global image feature,PCNN is utilized to fuse the high frequency sub-bands,the sum of local visual contrast (SLVC) to simulate the human visual feature is used as the external stimulus input to PCNN,the strength connection of PCNN is set to change with the visual contrast sensitivity (VCS),and considering th e iterations of PCNN,the Sigmoid function is used to compute the significant measurement of ignition output ampli tude of PCNN.Finally,the fused image is obtained by performing the inverse NSCT on the combined coefficients.T he comparison and analysis of experimental results show that the proposed approach can preserve in formation of source images effectively and improve the objective evaluation index and visual quality .