航空学报2025，Vol.46Issue(22)：255-271,17.DOI:10.7527/S1000-6893.2025.31987
MCS-RETR:改进RT-DETR的无人机航拍图像目标检测方法
MCS-RETR:Improved RT-DETR object detection method for UAV aerial images
摘要
Abstract
To address the challenges in Unmanned Aerial Vehicle(UAV)aerial imagery,such as blurred object edge information,complex background interference,and the low resolution and difficulty in identification of small objects,this paper proposes a novel improved RT-DETR-based object detection method for UAV aerial imagery,named MCS-DETR.Initially,a multi-scale edge information enhancement module is designed within the backbone network.By constructing an edge information enhancement mechanism and integrating it with deep convolutional operations and feature fusion strategies,the module aims to extract feature information at different scales and enhance the model's perception of image edge information.Subsequently,a convolutional additive token mixer is incorporated into the intra-scale feature interaction mechanism to optimize the feature interaction process and improve the model's capability of capturing global contextual key information.Finally,based on the original feature fusion method,a small object en-hancement pyramid network is proposed to enhance the model's ability to extract detailed features of small targets.Experimental results indicate that,compared to the RT-DETR model,the MCS-DETR algorithm reduces the param-eter size by 20.7%,while increasing accuracy,recall,mAP50,and mAP50∶95 of 2.4%,3.1%,2.8%,and 2.0%,re-spectively on the Visdrone2019-DET-Test dataset.This method effectively migrates missed and erroneous detections in complex scenes for UAV aerial image object detection.关键词
目标检测/无人机航拍图像/RT-DETR/边缘信息增强/卷积加法标记混合器/小目标增强金字塔Key words
object detection/UAV aerial imagery/RT-DETR/multi-scale edge information enhancement/convolu-tional additive token mixer/small object enhancement pyramid分类
航空航天引用本文复制引用
钟帅,王丽萍..MCS-RETR:改进RT-DETR的无人机航拍图像目标检测方法[J].航空学报,2025,46(22):255-271,17.基金项目
中国人民公安大学刑事科学技术双一流创新研究专项(2023SYL06) （2023SYL06）
国家自然科学基金(62271491) Double First-Class Innovative Research Special Fund for Criminal Science and Technology of People's Public Security University of China(2023SYL06) （62271491）
National Natural Science Foundation of China(62473044) （62473044）
