航空学报 > 2025, Vol. 46 Issue (3): 30848-030848   doi: 10.7527/S1000-6893.2024.30848

基于深度学习的无人机航拍图像小目标检测研究进展

吴一全(), 童康   

  1. 南京航空航天大学 电子信息工程学院,南京 211106
  • 收稿日期:2024-06-20 修回日期:2024-08-16 接受日期:2024-09-07 出版日期:2024-09-24 发布日期:2024-09-20
  • 通讯作者: 吴一全 E-mail:nuaaimage@163.com
  • 基金资助:
    国家自然科学基金(61573183)

Research advances on deep learning-based small object detection in UAV aerial images

Yiquan WU(), Kang TONG   

  1. College of Electronic and Information Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China
  • Received:2024-06-20 Revised:2024-08-16 Accepted:2024-09-07 Online:2024-09-24 Published:2024-09-20
  • Contact: Yiquan WU E-mail:nuaaimage@163.com
  • Supported by:
    National Natural Science Foundation of China(61573183)

摘要:

基于深度学习的无人机航拍图像小目标检测在军事情报侦察、战场监视和评估、军事目标捕获与验证、智能交通治理、基础设施检查和维护、灾害防治、搜索和救援、农作物管理与分析、生态保护和监测等领域具有广泛应用,近年来已成为当下的研究热点,故对近5年基于深度学习的无人机航拍图像小目标检测展开全面深入的调查。首先介绍无人机航拍图像小目标检测定义与面临的挑战。其次重点从判别性特征学习、超分辨率技术、实时轻量化检测、其他改进思路这4个方面详细阐述了无人机航拍图像小目标检测方法。然后系统总结无人机航拍图像小目标检测数据集,并基于VisDrone挑战赛深入分析不同算法的性能。最后全面呈现无人机航拍图像小目标检测在军事和民生领域的具体应用,讨论其未来潜在的发展方向,并指出了无人机航拍的一些担忧。期望该综述可以启发相关研究人员,进一步推动基于深度学习的无人机航拍图像小目标检测的发展。

关键词: 小目标检测, 无人机(UAV), 航拍图像, 深度学习, 性能评估, 小目标检测应用

Abstract:

Small object detection in UAV aerial images based on deep learning has a wide range of applications in military intelligence reconnaissance, battlefield surveillance and assessment, military object capture and verification, intelligent traffic management, infrastructure inspection and maintenance, disaster prevention and control, search and rescue, crop management and analysis, ecological protection and monitoring and other fields, and has become a current research hotspot in recent years. This review article gives a comprehensive and in-depth investigation on small object detection in UAV aerial images based on deep learning in the past five years. First of all, the definition and challenges of small object detection in UAV aerial images are introduced. Secondly, small object detection methods in drone aerial images are summarized in terms of discriminative feature learning, super-resolution technology, real-time lightweight detection, and other improvement ideas. Then, small object detection datasets of UAV aerial images are systematically summarized, and the performances of different algorithms are analyzed based on the VisDrone Challenge. Finally, the specific applications of small object detection in UAV aerial images in the military and civilian fields are comprehensively presented, and the potential future development directions of small object detection in UAV aerial images and some concerns about UAV aerial photography are also discussed. It is expected that this review would inspire relevant researchers to further promote the development of small object detection in UAV aerial images based on deep learning.

Key words: small object detection, Unmanned Aerial Vehicle (UAV), aerial image, deep learning, performance evaluation, small object detection application

中图分类号: