计算机工程与应用 ›› 2024, Vol. 60 ›› Issue (18): 17-31.DOI: 10.3778/j.issn.1002-8331.2403-0157

• 热点与综述 • 上一篇    下一篇

特征级红外与可见光图像融合方法综述

张宏钢,杨海涛,郑逢杰,王晋宇,周玺璇,王浩宇,徐一帆   

  1. 1.航天工程大学 研究生院,北京 101416
    2.航天工程大学,北京 101416
  • 出版日期:2024-09-15 发布日期:2024-09-13

Review of Feature-Level Infrared and Visible Image Fusion

ZHANG Honggang, YANG Haitao, ZHENG Fengjie, WANG Jinyu, ZHOU Xixuan, WANG Haoyu, XU Yifan   

  1. 1.Graduate School, Space Engineering University, Beijing 101416, China
    2.Space Engineering University, Beijing 101416, China
  • Online:2024-09-15 Published:2024-09-13

摘要: 红外与可见光图像融合是一门重要的图像处理技术,因其实用性被广泛应用。红外与可见光图像融合(infrared and visible image fusion,IVIF)是多模态图像融合技术中的一个重要分支,在对国内外的红外与可见光融合方法研究基础上,阐述了红外与可见光融合的基本理论,归纳了IVIF技术现状,分析了传统方法和深度学习融合方法的优缺点;重点对IVIF深度学习方法进行了详细的总结和分析,并梳理了现有的数据集和融合性能评价指标。最后,结合实际应用探讨了当前IVIF面临的挑战问题,并对未来该领域的发展方向进行了展望。

关键词: 红外与可见光, 图像融合, 融合方法

Abstract: Infrared and visible image fusion is an important image processing technique that is widely used because of its practicality. Infrared and visible image fusion (IVIF) is an important branch of multimodal image fusion technology. Based on the research of infrared and visible fusion methods at home and abroad, the basic theory of infrared and visible fusion is described, the current status of IVIF technology is summarized, and the advantages and disadvantages of the traditional method and the deep learning fusion method are analyzed. The detailed summary and analysis are conducted on the IVIF deep learning method, and the existing datasets and fusion performance evaluation indexes are sorted out. Finally, the current challenging issues facing IVIF are discussed in the context of practical applications, and the future development direction of the field is envisioned.

Key words: infrared and visible, image fusion, fusion method