测绘科学技术学报2025，Vol.41Issue(2)：204-211,8.DOI:10.3969/j.issn.1673-6338.2025.02.013
一种基于YOLOv10s改进的无人机目标检测及识别方法
An Improved UAV Target Detection and Recognition Method Based on YOLOv10s
摘要
Abstract
UAV targets play an increasingly important role in modern warfare and have quickly won the attention and operational use of various military organizations and combat units due to their difficulty in detection,difficulty in countermeasure,and ultra-high cost-effectiveness ratio.And anti-UAV operations have increasingly become a necessary combat capability for all combat units.In order to realize the rapid discovery,detection and recognition of UAV targets on the battlefield,an improved UAV target detection and recognition algorithm based on YOLOv10s is proposed in this paper.Firstly,the C2fCIB module in the backbone network of the YOLOv10s model is replaced with the self-calibrated convolutional SCConv module to improve the network performance.Secondly,the light-weight upsampling operator DySample is used to replace the standard sampling operator Upsample in YOLOv10s in the neck network to improve the efficiency and quality of image processing.Finally,three groups of MLCA modules are added before the head network to improve the network expression ability.The experimental results show that the accuracy of the training model of the improved YOLOv10s algorithm in 12 types of UAV target recognition is in-creased by 2.24%,the recall rate is increased by 11.59%,the average accuracy value mAP50 is increased by 4.01%,the average accuracy value mAP50-95 is increased by 8.06%,and the inference time is reduced by 6.56%respectively compared with those of YOLOv10s.It can provide a high-precision and high-efficiency solution for UAV target detection.关键词
无人机目标/识别/YOLOv10s算法/自校准卷积/动态上采样算子/混合局部通道注意力Key words
UAV targets/recognition/YOLOv10s algorithm/self-calibrated convolutions/dynamic sample/mixed local channel attention分类
天文与地球科学引用本文复制引用
李鹏,李鹏飞,殷瑞杰,刘志青,廉博,张锴,张二威..一种基于YOLOv10s改进的无人机目标检测及识别方法[J].测绘科学技术学报,2025,41(2):204-211,8.基金项目
陆军炮兵防空兵学院青年基金项目(PFXY220314003). （PFXY220314003）
