航空学报 > 2026, Vol. 47 Issue (6): 332601-332601   doi: 10.7527/S1000-6893.2024.32601

机载光电目标识别算法综述

黄俊, 张菁, 翁世倩()   

  1. 复杂航空系统仿真重点实验室,北京 100076
  • 收稿日期:2025-07-18 修回日期:2025-08-11 接受日期:2025-10-14 出版日期:2025-11-11 发布日期:2025-11-11
  • 通讯作者: 翁世倩 E-mail:68074638@qq.com

Airborne electro-optical target recognition algorithms

Jun HUANG, Jing ZHANG, Shiqian WENG()   

  1. National Key Laboratory of Complex Aviation System Simulation,Beijing 100076,China
  • Received:2025-07-18 Revised:2025-08-11 Accepted:2025-10-14 Online:2025-11-11 Published:2025-11-11
  • Contact: Shiqian WENG E-mail:68074638@qq.com

摘要:

机载光电目标识别技术在现代军事领域中具有重要的应用价值,但其在复杂战场环境下的性能仍面临诸多挑战。对基于可见光和红外数据的机载光电目标识别算法进行了系统性综述,重点分析了多光谱数据融合、深度学习应用、轻量化设计以及环境适应性等关键技术的研究现状与发展趋势。通过对现有算法分类应用研究,总结了各类方法的优势与局限性,并探讨了未来研究中亟待解决的问题,如复杂环境下的鲁棒性提升、实时性优化以及小样本学习等。旨在为机载光电目标识别领域的研究者提供全面的技术参考,并为进一步的研究方向提供理论支持。

关键词: 目标识别, 图像处理, 深度学习, 机载, 可见光与红外

Abstract:

Airborne electro-optical target recognition technology holds significant application value in modern military fields, yet its performance in complex battlefield environments still faces numerous challenges. This paper provides a systematic review of airborne electro-optical target recognition algorithms based on visible and infrared data, with a focus on key technologies such as multispectral data fusion, deep learning applications, lightweight design, and environmental adaptability. Through the classification and comparison of existing algorithms, this paper summarizes the advantages and limitations of various methods and explores critical issues to be addressed in future research, such as robustness enhancement in complex environments, real-time optimization, and few-shot learning. The aim of this paper is to offer a comprehensive technical reference for researchers in the field of airborne electro-optical target recognition and to provide theoretical support for future research directions.

Key words: target recognition, image processing, deep learning, airborne, visible and infrared

中图分类号: