EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

UAV-DETR: An Enhanced RT-DETR Architecture for Efficient Small Object Detection in UAV Imagery

Yu Zhou iD, Yan Wei

Sensors · 24.07.2025 · 25 (15) · 4582

DOI: 10.3390/s25154582 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 24.07.2025. Verfasst von Yu Zhou und 1 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer and Information Science, Chongqing Normal University, Chongqing 401331, China. Die Arbeit wurde bisher 59-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To mitigate the technical challenges associated with small-object detection, feature degradation, and spatial-contextual misalignment in UAV-acquired imagery, this paper proposes UAV-DETR, an enhanced Transformer-based object detection model designed for aerial scenarios. Specifically, UAV imagery often suffers from feature degradation due to low resolution and complex backgrounds and from semantic-spatial misalignment caused by dynamic shooting conditions. This work addresses these challenges by enhancing feature perception, semantic representation, and spatial alignment. Architecturally extending the RT-DETR framework, UAV-DETR incorporates three novel modules: the Channel-Aware Sensing Module (CAS), the Scale-Optimized Enhancement Pyramid Module (SOEP), and the newly designed Context-Spatial Alignment Module (CSAM), which integrates the functionalities of contextual and spatial calibration. These components collaboratively strengthen multi-scale feature extraction, semantic representation, and spatial-contextual alignment. The CAS module refines the backbone to improve multi-scale feature perception, while SOEP enhances semantic richness in shallow layers through lightweight channel-weighted fusion. CSAM further optimizes the hybrid encoder by simultaneously correcting contextual inconsistencies and spatial misalignments during feature fusion, enabling more precise cross-scale integration. Comprehensive comparisons with mainstream detectors, including Faster R-CNN and YOLOv5, demonstrate that UAV-DETR achieves superior small-object detection performance in complex aerial scenarios. The performance is thoroughly evaluated in terms of mAP@0.5, parameter count, and computational complexity (GFLOPs). Experiments on the VisDrone2019 dataset benchmark demonstrate that UAV-DETR achieves an mAP@0.5 of 51.6%, surpassing RT-DETR by 3.5% while reducing the number of model parameters from 19.8 million to 16.8 million.

Schlagwörter

Stichwörter der Autor:innen

RT-DETRUAV imageryfeature fusionsmall object detection

Autor:innen und Institutionen

  • College of Computer and Information Science, Chongqing Normal University, Chongqing 401331, China.

ORCID-Profile: Yu Zhou

Zitieren

APA

Zhou, Y., & Wei, Y. (2025). UAV-DETR: An Enhanced RT-DETR Architecture for Efficient Small Object Detection in UAV Imagery. Sensors, 25(15), 4582. https://doi.org/10.3390/s25154582

Vancouver

Zhou Y, Wei Y. UAV-DETR: An Enhanced RT-DETR Architecture for Efficient Small Object Detection in UAV Imagery. Sensors. 2025;25(15):4582. doi:10.3390/s25154582

BibTeX
@article{Zhou2025,
  author = {Zhou, Yu and Wei, Yan},
  title = {UAV-DETR: An Enhanced RT-DETR Architecture for Efficient Small Object Detection in UAV Imagery},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {15},
  pages = {4582},
  doi = {10.3390/s25154582}
}
Kostenlos starten →