EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SAQ-YOLO: An Efficient Small Object Detection Model for Unmanned Aerial Vehicle in Maritime Search and Rescue

Sichen Li, Hao Yi, Shengyi Chen, Xinmin Chen, Mao Xu, Feifan Yu

Applied Sciences · 22.12.2025 · 16 (1) · 131

DOI: 10.3390/app16010131 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 22.12.2025. Verfasst von Sichen Li und 5 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Kp Talent Technology Project in Ningbo City, Ningbo Science and Technology Innovation Yongjiang Key R&D, Top Talent Technology Project in Ningbo City.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In Search and Rescue (SAR) missions, UAVs must be capable of detecting small objects from complex and noise-prone maritime images. Existing small object detection methods typically rely on super-resolution techniques or complex structural designs, which often demand significant computational resources and fail to meet the real-time requirements for small mobile devices in SAR tasks. To address this challenge, we propose SAQ-YOLO, an efficient small object detection model based on the YOLO framework. We design a Small Object Auxiliary Query branch, which uses deep semantic information to guide the fusion of shallow features, thereby improving small object capture efficiency. Additionally, SAQ-YOLO incorporates a series of lightweight channel, spatial, and group (large kernel) gated attention mechanisms to suppress background clutter in complex maritime environments, enhancing feature extraction at a low computational cost. Experiments on the SeaDronesSee dataset demonstrate that, compared to YOLOv11s, SAQ-YOLO reduces the number of parameters by approximately 70% while increasing mAP@50 by 2.1 percentage points. Compared to YOLOv11n, SAQ-YOLO improves mAP@50 by 8.7 percentage points. When deployed on embedded platforms, SAQ-YOLO achieves an inference latency of only 35 milliseconds per frame, meeting the real-time requirements of maritime SAR applications. These results suggest that SAQ-YOLO provides an efficient and deployable solution for UAV SAR operations in vast and highly dynamic marine environments. Future work will focus on enhancing the robustness of the detection model.

Förderung

  • Kp Talent Technology Project in Ningbo City
  • Ningbo Science and Technology Innovation Yongjiang Key R&D
  • Top Talent Technology Project in Ningbo City

Zitieren

APA

Li, S., Yi, H., Chen, S., Chen, X., Xu, M., & Yu, F. (2025). SAQ-YOLO: An Efficient Small Object Detection Model for Unmanned Aerial Vehicle in Maritime Search and Rescue. Applied Sciences, 16(1), 131. https://doi.org/10.3390/app16010131

Vancouver

Li S, Yi H, Chen S, Chen X, Xu M, Yu F. SAQ-YOLO: An Efficient Small Object Detection Model for Unmanned Aerial Vehicle in Maritime Search and Rescue. Applied Sciences. 2025;16(1):131. doi:10.3390/app16010131

BibTeX
@article{Li2025,
  author = {Li, Sichen and Yi, Hao and Chen, Shengyi and Chen, Xinmin and Xu, Mao and Yu, Feifan},
  title = {SAQ-YOLO: An Efficient Small Object Detection Model for Unmanned Aerial Vehicle in Maritime Search and Rescue},
  journal = {Applied Sciences},
  year = {2025},
  volume = {16},
  number = {1},
  pages = {131},
  doi = {10.3390/app16010131}
}
Kostenlos starten →