RFHS-RTDETR: Multi-Domain Collaborative Network With Hierarchical Feature Integration for UAV-Based Object Detection

To address UAV-based detection challenges including scale variation, high target density, and hardware limitations, we propose RFHS-RTDETR with four innovations: (1) RMConv, a reparameterized lightweight module for efficient multi-scale feature extraction; (2) FSConv fusing Scharr operator and Fouri...

Full description

Saved in:
Bibliographic Details
Main Authors: Songtao Tang, Leiming Zhang, Xudong Liu, Rongfu Lv, Ruyi Qin
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11072410/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850100804094001152
author Songtao Tang
Leiming Zhang
Xudong Liu
Rongfu Lv
Ruyi Qin
author_facet Songtao Tang
Leiming Zhang
Xudong Liu
Rongfu Lv
Ruyi Qin
author_sort Songtao Tang
collection DOAJ
description To address UAV-based detection challenges including scale variation, high target density, and hardware limitations, we propose RFHS-RTDETR with four innovations: (1) RMConv, a reparameterized lightweight module for efficient multi-scale feature extraction; (2) FSConv fusing Scharr operator and Fourier transform to enhance edge preservation and robustness; (3) AH module combining HiLo attention for dense target recognition; (4) SOPS Feature Pyramid with hierarchical feature integration of P2 features and dynamic upsampling for small objects. On VisDrone2019, RFHS-RTDETR reduces FLOPs and parameters by 16.2% and 34.3% versus RT-DETR-R18, while improving precision by 2.2%, mAP50 by 2%, and FPS by 17.1%.These advancements demonstrate its practicality for resource-constrained aerial scenarios.
format Article
id doaj-art-fe67b296726b4e0eacc035ed9b5c5795
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-fe67b296726b4e0eacc035ed9b5c57952025-08-20T02:40:11ZengIEEEIEEE Access2169-35362025-01-011312168612170310.1109/ACCESS.2025.358664911072410RFHS-RTDETR: Multi-Domain Collaborative Network With Hierarchical Feature Integration for UAV-Based Object DetectionSongtao Tang0Leiming Zhang1https://orcid.org/0009-0003-7889-610XXudong Liu2Rongfu Lv3Ruyi Qin4College of Information Engineering and Artificial Intelligence, Henan University of Science and Technology, Luoyang, ChinaCollege of Information Engineering and Artificial Intelligence, Henan University of Science and Technology, Luoyang, ChinaLuoyang CloudTek Optoelectronics Technology Company, Luoyang, ChinaCollege of Information Engineering and Artificial Intelligence, Henan University of Science and Technology, Luoyang, ChinaCollege of Information Engineering and Artificial Intelligence, Henan University of Science and Technology, Luoyang, ChinaTo address UAV-based detection challenges including scale variation, high target density, and hardware limitations, we propose RFHS-RTDETR with four innovations: (1) RMConv, a reparameterized lightweight module for efficient multi-scale feature extraction; (2) FSConv fusing Scharr operator and Fourier transform to enhance edge preservation and robustness; (3) AH module combining HiLo attention for dense target recognition; (4) SOPS Feature Pyramid with hierarchical feature integration of P2 features and dynamic upsampling for small objects. On VisDrone2019, RFHS-RTDETR reduces FLOPs and parameters by 16.2% and 34.3% versus RT-DETR-R18, while improving precision by 2.2%, mAP50 by 2%, and FPS by 17.1%.These advancements demonstrate its practicality for resource-constrained aerial scenarios.https://ieeexplore.ieee.org/document/11072410/RT-DETRUAVsmall target detectionSOPS feature pyramidlightweight
spellingShingle Songtao Tang
Leiming Zhang
Xudong Liu
Rongfu Lv
Ruyi Qin
RFHS-RTDETR: Multi-Domain Collaborative Network With Hierarchical Feature Integration for UAV-Based Object Detection
IEEE Access
RT-DETR
UAV
small target detection
SOPS feature pyramid
lightweight
title RFHS-RTDETR: Multi-Domain Collaborative Network With Hierarchical Feature Integration for UAV-Based Object Detection
title_full RFHS-RTDETR: Multi-Domain Collaborative Network With Hierarchical Feature Integration for UAV-Based Object Detection
title_fullStr RFHS-RTDETR: Multi-Domain Collaborative Network With Hierarchical Feature Integration for UAV-Based Object Detection
title_full_unstemmed RFHS-RTDETR: Multi-Domain Collaborative Network With Hierarchical Feature Integration for UAV-Based Object Detection
title_short RFHS-RTDETR: Multi-Domain Collaborative Network With Hierarchical Feature Integration for UAV-Based Object Detection
title_sort rfhs rtdetr multi domain collaborative network with hierarchical feature integration for uav based object detection
topic RT-DETR
UAV
small target detection
SOPS feature pyramid
lightweight
url https://ieeexplore.ieee.org/document/11072410/
work_keys_str_mv AT songtaotang rfhsrtdetrmultidomaincollaborativenetworkwithhierarchicalfeatureintegrationforuavbasedobjectdetection
AT leimingzhang rfhsrtdetrmultidomaincollaborativenetworkwithhierarchicalfeatureintegrationforuavbasedobjectdetection
AT xudongliu rfhsrtdetrmultidomaincollaborativenetworkwithhierarchicalfeatureintegrationforuavbasedobjectdetection
AT rongfulv rfhsrtdetrmultidomaincollaborativenetworkwithhierarchicalfeatureintegrationforuavbasedobjectdetection
AT ruyiqin rfhsrtdetrmultidomaincollaborativenetworkwithhierarchicalfeatureintegrationforuavbasedobjectdetection