改进多尺度结构化融合的红外与可见光图像融合
Infrared and visible images fusion based on improved multi-scale structural fusion
龙志亮 1邓月明 1谢竞 1王润民1
作者信息
- 1. 湖南师范大学 信息科学与工程学院,湖南 长沙 410081
- 折叠
摘要
为解决弱光环境下的红外与可见光图像的融合结果存在对比度低、细节纹理不足、融合耗时长的问题,提出了一种改进多尺度结构化融合的方法.在图像融合前,采用动态范围压缩算法对弱光下的可见光图像进行增强,再通过多尺度结构化分解将增强后的可见光图像和红外图像分解成低频和高频信息;在融合过程中,提出一种基于均方根误差系数的方法对低频信息进行融合,提出一种基于信息熵自适应调整权重的策略对初步融合的高频信息进行二次优化融合,再通过多尺度结构化分解的逆变换重构出融合图像;最后,提出一种基于灰度分类的区域像素增强算法来提高融合后图像的对比度.将提出方法与9种常用的融合方法进行了定性和定量的分析比较,在TNO和CVC-14数据集上的实验结果表明,该方法在平均梯度、交叉熵、边缘强度、标准差和空间频率指标上取得了更好的客观评价结果,整体视觉效果也要优于对比方法.本方法的融合结果具有丰富的细节纹理、较高的清晰度和对比度,且融合耗时短.
Abstract
Under low-light conditions,the fusion of infrared and visible images often results in images with poor contrast,lacking in detail,and requiring a lengthy processing time.To address these issues,this pa-per introduces an enhanced multi-scale structural fusion approach.Initially,it improves the low-light visi-ble image using a dynamic range compression enhancement algorithm.Subsequently,through a multi-scale structural image decomposition method,it separates the enhanced visible and infrared images into their low-frequency and high-frequency components.For image fusion,the low-frequency components of both image types are merged using a technique based on the root mean square error coefficient.The high-fre-quency components are initially fused in a straightforward manner,followed by an optimized fusion using a self-adaptive weight adjustment based on image information entropy.Afterward,by reversing the multi-scale structural decomposition,the fused low and high-frequency components are combined to form a com-plete image.To further enhance the image contrast,a regional pixel enhancement algorithm based on gray level classification is introduced.The effectiveness of this method is compared with nine conventional infra-red and visible image fusion techniques,both qualitatively and quantitatively,using TNO and CVC-14 da-tasets.The proposed method demonstrates superior performance in terms of average gradient,cross entro-py,edge intensity,standard deviation,and spatial frequency,along with an improved overall visual quali-ty.This confirms that the images produced by the proposed method exhibit enhanced detail,clarity,con-trast,and are processed more quickly.
关键词
图像处理/多尺度结构化融合/动态范围压缩/均方根误差/信息熵/对比度Key words
image processing/multi-scale structural fusion/dynamic range compression/root mean square error/information entropy/contrast引用本文复制引用
基金项目
国家自然科学基金资助项目(62072175)
湖南省教育厅科学研究资助项目(21C0008)
湖南省教育厅重点基金资助项目(21A0052)
长沙市重点研发计划资助项目(kq2004050)
出版年
2024
国家科技期刊平台
NSTL
万方数据