基于改进引导滤波和双通道脉冲发放皮层模型的红外与可见光图像融合算法  

Infrared and Visible Image Fusion Algorithm Based on Improved Guided Filtering and Dual-Channel Spiking Cortical Model

作　　者：江泽涛[1] 吴辉[1] 周哓玲 

机构地区：[1]桂林电子科技大学广西高校图像图形智能处理重点实验室,广西桂林541004

出　　处：《光学学报》2018年第2期112-120,共9页Acta Optica Sinica

基　　金：国家自然科学基金(61572147);广西科技计划项目(AC16380108;2015BC19022);桂林电子科技大学图像图形智能处理重点实验项目(GIIP201501;GIIP201601);广西可信软件重点实验室项目(kx201502);广西研究生教育创新计划资助项目(YJCXS201536;2016YJCX71)

摘　　要：针对传统红外与可见光图像融合算法中存在的目标不够突出、背景缺失、边缘信息保留不够充分等问题,提出了一种基于改进引导滤波和双通道脉冲发放皮层模型(DCSCM)的红外与可见光图像融合算法。首先,对源图像进行非降采样Shearlet变换(NSST),得到相应的低频和高频分量。然后,分别采用改进的引导滤波算法和DCSCM模型对低频、高频分量进行融合。最后,对融合得到的高低频分量进行NSST逆变换得到最终的融合图像。与其他几种方法进行比较,实验结果表明,本文算法的融合图像目标突出,背景信息丰富,且在图像清晰度、对比度、信息熵等方面均有优势。We propose a novel multi-source image fusion algorithm based on the improved guided image filtering and dual-channel spiking cortical model（DCSCM）,in order to solve the problems that targets are not prominent,background information is easy to lose and margin information is not adequately retained.Firstly,we obtain the low frequency and high frequency sub-bands by using the nonsubsampled shearlet transform（NSST）to the two original images.Then we use the improved guided image filtering algorithm and the DCSCM model to fuse the low frequency and high frequency components respectively.Finally,we perform the inverse NSST transform on the fused high and low frequency components to obtain the final fused image.Experimental results show that the fused images of the proposed algorithm have prominent targets and rich background information.The proposed algorithm has advantages in terms of image clarity,contrast,and information entropy,compared with several other methods.

关 键 词：图像处理 图像融合 非降采样Shearlet变换 引导滤波 双通道脉冲发放皮层模型 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]