航空兵器2024，Vol.31Issue(6)：23-35,13.DOI:10.12132/ISSN.1673-5048.2024.0148
基于深度学习的无人机可见光目标检测研究综述
A Survey of UAV Visible-Light Object Detection Based on Deep Learning
摘要
Abstract
With the rapid development of artificial intelligence,visible-light object detection,as an important part of computer vision technology,has been widely used in the unmanned aerial vehicle(UAV)reconnaissance field.U-sing deep learning technology to deeply explore object features in complex battlefield environments and low-quality ima-ges can effectively solve the difficulties and challenges of visible-light object detection in UAV reconnaissance scenario,and further improve the accuracy of visible-light object detection.Therefore,a comprehensive survey is conducted on UAV visible-light object detection methods based on deep learning.First,various challenges of UAV visible-light ob-ject detection are introduced,such as small scale,arbitrary orientation,high camouflage,and motion blur.Second,main public datasets for visible-light object detection and image restoration are described.Then,combined with various challenges faced by UAV visible-light object detection,the application,advantages and disadvantages of deep learning methods in UAV visible-light object detection are summarized.Finally,the future possible research direction for UAV visible light object detection is discussed.关键词
无人机/复杂战场环境/低质量图像/深度学习/可见光目标检测Key words
UAV/complex battlefield environment/low-quality image/deep learning/visible-light object detec-tion分类
军事科技引用本文复制引用
刘克顺,左晓桐,张玉华,王长龙,杨森..基于深度学习的无人机可见光目标检测研究综述[J].航空兵器,2024,31(6):23-35,13.基金项目
基础加强计划技术领域基金项目(2019-JCJQ-JJ-015) （2019-JCJQ-JJ-015）
