NSCT域内基于自适应PCNN的图像融合新方法

论文结合非下采样contourlet变换（NSCT）的平移不变性、多尺度、多方向特性和脉冲耦合神经网络(PCNN)的同步脉冲发放、捕获特性，提出在NSCT域中基于PCNN的图像融合框架。对于低频子带，利用改进拉普拉斯能量和作为特征激励PCNN；对于高频方向子带，采用改进的空间频率作为PCNN的外部激励；同时利用各子带图像的平均梯度自适应调节PCNN的链接强度，最后，选取具有较大点火次数的系数作为融合图像的系数，经逆NSCT变换重构融合图像。实验结果表明本文方法无论在主观视觉还是客观评价标准上都要优于传统的基于小波变换、contourlet变换、PCNN的图像融合方法。 

A Novel Image Fusion Algorithm Using Adaptive PCNN in NSCT Domain

In this paper, a novel image fusion algorithm based on non-subsampled contourlet transform(NSCT) and pulse coupled neural network(PCNN) was proposed. NSCT provides flexible multiresolution, anisotropy and directional expansion for images. PCNN is a visual cortex-inspired neural network and characterized by the global coupling and pulse synchronization of neurons. For low-frequency sub-images, using sum-modified-Laplacian as the external stimuli of the PCNN, for the high-frequency directional sub-images, the modified spatial frequency are used as the external stimuli of the PCNN. At the same time, the average gradient of each sub-band image is used to adjust the liking strength adaptively. Coefficients with large firing times were selected as the coefficients of the fused image. Experimental results show that the proposed scheme can significantly improve image fusion performance and outperforms the conventional algorithms such as wavelet transform, contourlet transform and PCNN in term of objective criteria and visual appearance.