Theoretical analysis of an information-based quality measure for image fusion

被引：35

作者：

Chen, Yin ^{[1
]}

Xue, Zhiyun ^{[1
]}

Blum, Rick S. ^{[1
]}

机构：

[1] Lehigh Univ, ECE Dept, Bethlehem, PA 18015 USA

来源：

INFORMATION FUSION | 2008年 / 9卷 / 02期

关键词：

image fusion; Image quality analysis; Image quality measure;

D O I：

10.1016/j.inffus.2007.03.002

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 [模式识别与智能系统]; 0812 [计算机科学与技术]; 0835 [软件工程]; 1405 [智能科学与技术];

摘要：

While recently a few image fusion quality measures have been proposed, analytical studies of these measures have been lacking. Here, we focus on one popular mutual information-based quality measure and weighted averaging image fusion. Based on an image formation model, we obtain a closed-form expression for the quality measure and mathematically analyze its properties under different types of image distortion. Tests with real images are also presented which agree with the conclusions of the analytical results. The results show the quality measure studied does not generally properly characterize increases in the distortion (noise and blurring) of the images which are input into a weighted averaging fusion algorithm. (c) 2007 Elsevier B.V. All rights reserved.

引用

页码：161 / 175

页数：15

共 29 条

[1]

Aguilar M, 2002, PROCEEDINGS OF THE FIFTH INTERNATIONAL CONFERENCE ON INFORMATION FUSION, VOL II, P1206, DOI 10.1109/ICIF.2002.1020950

[2]

On multisensor image fusion performance limits from an estimation theory perspective [J].