基于相关系数之和的图像融合质量评价  

Evaluation of Image Fusion Quality Based on Sum of Correlation Coefficients

作　　者：马文娟 许峰 周继振 MA Wenjuan;XU Feng;ZHOU Jizhen(School of Mathematics and Big Data,Anhui University of Science and Technology,Huainan 232001,China)

机构地区：[1]安徽理工大学数学与大数据学院,安徽淮南232001

出　　处：《淮海工学院学报（自然科学版）》2018年第4期20-24,共5页Journal of Huaihai Institute of Technology:Natural Sciences Edition

基　　金：安徽省自然科学基金资助项目(1608085MA01);安徽省教育厅提升计划项目(13082)

摘　　要：针对现有融合图像质量评价算法的特点,提出了一种新型的、基于差值图像和源图像间相关系数之和(SCD)的图像融合评价方法,而不需要标准图像作为参考依据.该算法首先得到两个源图像和融合图像的差值图像,然后求出源图像和差值图像之间的相关系数,最后把两个相关系数的和作为指标来评价图像融合的质量.结果表明,该算法客观定量的评价和主观目视的评价是一致的,而且优于其他传统的图像融合指标,如互信息量、边缘保持度和联合熵.By analyzing the existent of image fusion quality,a non-reference objective image fusion metric based on the sum of correlation coefficients(SCD)between the source images and corresponding difference images is proposed.Firstly,the difference image between the source image and the fused image is obtained,and then the correlation coefficient between the source image and the difference image is obtained.Finally,the sum of the two correlation coefficients is used as an index to evaluate the quality of image fusion.The experimental results show that the objective and quantitative evaluation of the algorithm is consistent with the subjective and visual evaluation,and it is superior to other traditional image fusion indicators,such as mutual information,edge preservation and joint entropy.

关 键 词：图像融合 相关系数之和 差值图像 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]