EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Dual-Modal Adaptive Pyramid Transformer Algorithm for UAV Cross-Modal Object Detection

Qiqin Li, Ming Yang iD, Xiaoqiang Zhang, Nannan Wang iD, Xiaoguang Tu, Xijun Liu, Xinyu Zhu

Sensors · 11.12.2025 · 25 (24) · 7541

DOI: 10.3390/s25247541 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 11.12.2025. Verfasst von Qiqin Li und 6 weiteren Autor:innen. Beteiligte Einrichtung: College of Aviation Electronics and Electrical, Civil Aviation Flight University of China, Guanghan 618307, China. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Basic Scientific Research Expenses of Central Universities, the Key Laboratory of Flight Techniques and Flight Safety, the Sichuan Provincial Youth Science and Technology Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unmanned Aerial Vehicles (UAVs) play vital roles in traffic surveillance, disaster management, and border security, highlighting the importance of reliable infrared–visible image detection under complex illumination conditions. However, UAV-based infrared–visible detection still faces challenges in multi-scale target recognition, robustness to lighting variations, and efficient cross-modal information utilization. To address these issues, this study proposes a lightweight Dual-modality Adaptive Pyramid Transformer (DAP) module integrated into the YOLOv8 framework. The DAP module employs a hierarchical self-attention mechanism and a residual fusion structure to achieve adaptive multi-scale representation and cross-modal semantic alignment while preserving modality-specific features. This design enables effective feature fusion with reduced computational cost, enhancing detection accuracy in complex environments. Experiments on the DroneVehicle and LLVIP datasets demonstrate that the proposed DAP-based YOLOv8 achieves mAP50:95 scores of 61.2% and 62.1%, respectively, outperforming conventional methods. The results validate the capability of the DAP module to optimize cross-modal feature interaction and improve UAV real-time infrared–visible target detection, offering a practical and efficient solution for UAV applications such as traffic monitoring and disaster response.

Schlagwörter

Stichwörter der Autor:innen

dual-modality fusioninfrared-visible image detectiontarget detectiontransformerunmanned aerial vehicle

Autor:innen und Institutionen

  • College of Aviation Electronics and Electrical, Civil Aviation Flight University of China, Guanghan 618307, China.
  • School of Optoelectronic Science and Engineering, University of Electronic Science and Technology of China, Chengdu 610054, China.
  • Sichuan Province Engineering Technology Research Center of General Aircraft Maintenance, Civil Aviation Flight University of China, Guanghan 618307, China.
  • Luoyang College, Civil Aviation Flight University of China, Luoyang 471000, China.

ORCID-Profile: Ming Yang, Nannan Wang

Förderung

  • the Basic Scientific Research Expenses of Central Universities
  • the Key Laboratory of Flight Techniques and Flight Safety
  • the Sichuan Provincial Youth Science and Technology Foundation

Zitieren

APA

Li, Q., Yang, M., Zhang, X., Wang, N., Tu, X., Liu, X., & Zhu, X. (2025). A Dual-Modal Adaptive Pyramid Transformer Algorithm for UAV Cross-Modal Object Detection. Sensors, 25(24), 7541. https://doi.org/10.3390/s25247541

Vancouver

Li Q, Yang M, Zhang X, Wang N, Tu X, Liu X, et al. A Dual-Modal Adaptive Pyramid Transformer Algorithm for UAV Cross-Modal Object Detection. Sensors. 2025;25(24):7541. doi:10.3390/s25247541

BibTeX
@article{Li2025,
  author = {Li, Qiqin and Yang, Ming and Zhang, Xiaoqiang and Wang, Nannan and Tu, Xiaoguang and Liu, Xijun and Zhu, Xinyu},
  title = {A Dual-Modal Adaptive Pyramid Transformer Algorithm for UAV Cross-Modal Object Detection},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {24},
  pages = {7541},
  doi = {10.3390/s25247541}
}
Kostenlos starten →