基于导向滤波和聚焦区域局部标准差的多聚焦图像融合  

Multi-focus Image Fusion Based on Guided Filtering and Local Standard Deviation of Focus Region

作　　者：谢明 曲怀敬[1] 张志升 王纪委 张汉元 XIE Ming;QU Huaijing;ZHANG Zhisheng;WANG Jiwei;ZHANG Hanyuan(School of Information&Electric Engineering,Shandong Jianzhu University,Jinan 250101)

机构地区：[1]山东建筑大学信息与电气工程学院,济南250101

出　　处：《计算机与数字工程》2023年第2期348-354,共7页Computer & Digital Engineering

基　　金：国家自然科学基金项目(编号:62003191);山东省自然科学基金项目(编号:ZR2014FM016)资助。

摘　　要：针对目前多聚焦图像融合存在着在融合图像边缘处容易出现伪影和光环等问题,提出了一种新的基于导向滤波和聚焦区域局部标准差的多聚焦图像融合方法。首先,设计了一种新的聚焦区域检测策略,它将导向滤波后的源图像与均值滤波后的源图像相减得到绝对差值图像,并以此作为高频信息而生成粗略聚焦图;粗略聚焦图进一步经过导向滤波增强其高频细节信息,从而得到对应源图像的聚焦区域。然后,根据聚焦区域的局部标准差,利用像素级最大化规则得到初始决策图。最后,通过小区域去除策略和导向滤波优化生成最终决策图,并据此利用像素加权平均规则得到融合后的图像。实验结果表明,论文提出的多聚焦图像融合方法在主观视觉效果和客观评价指标方面的综合性能都优于现有的典型融合方法,且其计算效率也较高。In order to solve the problems of artifact and halo at the edge of the fusion image,a novel multi-focus image fusion method based on guided filtering and local standard deviation of the focus region is proposed.Firstly,a new focus region detection strategy is designed,by which the guided filtered source image subtracts the mean filtered source image to obtain an absolute difference image,which is used as high-frequency information to generate a rough focus map.The high-frequency detail information of the rough focus map is further enhanced by guided filtering,so as to obtain the focus region corresponding to the source image.Then,based on the local standard deviation of the focus region,the pixel-level maximization rule is used to obtain the initial decision map.Finally,the final decision map is generated through the small area removal strategy and guided filtering optimization,and the fused image is obtained by using the pixel weighted average rule.The experimental results show that the comprehensive performance of the multi-focus image fusion method proposed in this paper is superior to the existing typical fusion methods in both subjective visual effects and objective evaluation metrics,and its computational efficiency is also higher.

关 键 词：多聚焦图像融合 导向滤波 局部标准差 差值图像 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]