摘要
针对现有的红外与可见光图像融合方法所存在的融合后图像中细节信息量不足问题,本文提出一种基于区域特性量测的小波变换图像融合算法。首先,对源图像进行2层的小波分解,获得6个高频子图像和1个低频子图像。然后,对高频子图像采用基于区域特性量测的融合规则,对低频子图像采用取最大值方法得到融合图像的多分辨率分析。最后,通过小波逆变换获得融合图像。实验结果表明,采用本文算法所得到的融合图像,其细节信息进一步加强,客观融合评价指标均优于现有经典算法。
Aiming at the problem of insufficient detail information in the fusion image of existing infrared and visible light image fusion methods,a regional characteristic measurement and wavelet transform based image fusion algorithm is proposed.First,the source images are decomposed by the discrete wavelet transform into two levels,and obtain six high-frequency sub-images and one low-frequency sub-image.Then,we use the regional characteristic measurement on the high-frequency sub-images to find construct the fused high-frequency coefficients.The low-frequency coefficient of the fused image is obtained by the maximum selection rule.Last,the fused image is obtained by the inverse wavelet transform.The experimental results indicate that the proposed algorithm works better than the traditional methods in both objective and subjective evaluation.
作者
郭立强
范志祥
GUO Li-qiang;FAN Zhi-xiang(School of Computer Science and Technology Huaiyin Normal University,Huai'an Jiangsu 223300,China)
出处
《长春师范大学学报》
2019年第2期15-18,共4页
Journal of Changchun Normal University
基金
国家自然科学基金项目"可交换Clifford代数彩色图像矩不变量研究"(61243242)
江苏省大学生实践创新训练计划项目"红外与可见光图像融合算法"(201810323004Z)
关键词
图像融合
小波变换
区域特性量测
image fusion
wavelet transform
regional characteristic measurement
作者简介
郭立强,男,副教授,博士,从事图像处理与模式识别研究。
