EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AeroLight: A Lightweight Architecture with Dynamic Feature Fusion for High-Fidelity Small-Target Detection in Aerial Imagery

Hao Qiu iD, Xiaoyan Meng, Yunjie Zhao iD, Liang Yu, Shuai Yin

Sensors · 30.08.2025 · 25 (17) · 5369

DOI: 10.3390/s25175369 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 30.08.2025. Verfasst von Hao Qiu und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer and Information Engineering, Xinjiang Agricultural University, Urumqi 830052, China. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the National Key R&D Program of China, the Provincial Key S&T Program of Xinjiang, National Key R&D Program of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Small-target detection in Unmanned Aerial Vehicle (UAV) aerial images remains a significant and unresolved challenge in aerial image analysis, hampered by low target resolution, dense object clustering, and complex, cluttered backgrounds. In order to cope with these problems, we present AeroLight, a novel and efficient detection architecture that achieves high-fidelity performance in resource-constrained environments. AeroLight is built upon three key innovations. First, we have optimized the feature pyramid at the architectural level by integrating a high-resolution head specifically designed for minute object detection. This design enhances sensitivity to fine-grained spatial details while streamlining redundant and computationally expensive network layers. Second, a Dynamic Feature Fusion (DFF) module is proposed to adaptively recalibrate and merge multi-scale feature maps, mitigating information loss during integration and strengthening object representation across diverse scales. Finally, we enhance the localization precision of irregular-shaped objects by refining bounding box regression using a Shape-IoU loss function. AeroLight is shown to improve mAP50 and mAP50-95 by 7.5% and 3.3%, respectively, on the VisDrone2019 dataset, while reducing the parameter count by 28.8% when compared with the baseline model. Further validation on the RSOD dataset and Huaxing Farm Drone dataset confirms its superior performance and generalization capabilities. AeroLight provides a powerful and efficient solution for real-world UAV applications, setting a new standard for lightweight, high-precision object recognition in aerial imaging scenarios.

Schlagwörter

Stichwörter der Autor:innen

AeroLightUAV aerial imagesdynamic feature fusionlightweightsmall-target detection

Autor:innen und Institutionen

  • School of Computer and Information Engineering, Xinjiang Agricultural University, Urumqi 830052, China.
  • Ministry of Education, Engineering Research Center for Intelligent Agriculture, Urumqi 830052, China.
  • Xinjiang Agricultural Informatization Engineering Technology Research Center, Urumqi 830052, China.

ORCID-Profile: Hao Qiu, Yunjie Zhao

Förderung

  • the National Key R&D Program of China
  • the Provincial Key S&T Program of Xinjiang
  • National Key R&D Program of China
  • Provincial Key S&T Program of Xinjiang

Zitieren

APA

Qiu, H., Meng, X., Zhao, Y., Yu, L., & Yin, S. (2025). AeroLight: A Lightweight Architecture with Dynamic Feature Fusion for High-Fidelity Small-Target Detection in Aerial Imagery. Sensors, 25(17), 5369. https://doi.org/10.3390/s25175369

Vancouver

Qiu H, Meng X, Zhao Y, Yu L, Yin S. AeroLight: A Lightweight Architecture with Dynamic Feature Fusion for High-Fidelity Small-Target Detection in Aerial Imagery. Sensors. 2025;25(17):5369. doi:10.3390/s25175369

BibTeX
@article{Qiu2025,
  author = {Qiu, Hao and Meng, Xiaoyan and Zhao, Yunjie and Yu, Liang and Yin, Shuai},
  title = {AeroLight: A Lightweight Architecture with Dynamic Feature Fusion for High-Fidelity Small-Target Detection in Aerial Imagery},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {17},
  pages = {5369},
  doi = {10.3390/s25175369}
}
Kostenlos starten →