EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Research on a CEF-YOLOv8n-Based Method for Small Object Detection in UAV Aerial Imagery

Yamei Zhang iD, Keding Yan iD

Sensors · 29.07.2026 · 26 (15) · 4818

DOI: 10.3390/s26154818 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 29.07.2026. Verfasst von Yamei Zhang und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Intelligent Science and Engineering, Xi'an Peihua University, Xi'an 710125, China. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Education Department of Shaanxi Province, Shaanxi Provincial Department of Education.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To address the recognition challenges caused by the high proportion, low resolution, and significant multi-scale variations of small objects in UAV small-object detection tasks, a UAV small-object detection and recognition algorithm based on CEF-YOLOv8n is proposed. The proposed algorithm uses YOLOv8n as the baseline network and introduces a Partial Convolution-based Cross Partial Feature (CPF) module into the backbone network to enhance the local feature extraction capability for low-resolution small objects. In the neck network, the concept of feature focusing and diffusion is adopted to construct a Focusing Generalized Feature Pyramid Network (FGFPN). A Feature Semantic Fusion Module (FSFM) based on a cross-attention mechanism is designed to complementarily fuse shallow detail features with deep semantic features, thereby enhancing information interaction among objects at different scales. In addition, a Lightweight Weight-Sharing Detection Head (LWSD) is proposed to improve the computational efficiency and real-time performance of the model while maintaining detection accuracy. Publicly available datasets are used for network training and detection evaluation, and comparative experiments are conducted with other algorithms. The results show that the proposed detection and recognition algorithm achieves 37.6% and 22.6% in terms of mAP50 and mAP50-95, respectively, representing improvements of 3.4 and 2.6 percentage points over the original YOLOv8n. Meanwhile, the number of parameters and FLOPs are reduced from 3.2 M and 8.7 G to 2.5 M and 6.9 G, respectively.

Schlagwörter

Stichwörter der Autor:innen

UAVYOLOv8nfeature fusionmulti-scaletarget recognition

Autor:innen und Institutionen

  • School of Intelligent Science and Engineering, Xi'an Peihua University, Xi'an 710125, China.
  • School of Electronic and Information Engineering, Xi'an Technological University, Xi'an 710021, China.

ORCID-Profile: Yamei Zhang, Keding Yan

Förderung

  • Education Department of Shaanxi Province
  • Shaanxi Provincial Department of Education

Zitieren

APA

Zhang, Y., & Yan, K. (2026). Research on a CEF-YOLOv8n-Based Method for Small Object Detection in UAV Aerial Imagery. Sensors, 26(15), 4818. https://doi.org/10.3390/s26154818

Vancouver

Zhang Y, Yan K. Research on a CEF-YOLOv8n-Based Method for Small Object Detection in UAV Aerial Imagery. Sensors. 2026;26(15):4818. doi:10.3390/s26154818

BibTeX
@article{Zhang2026,
  author = {Zhang, Yamei and Yan, Keding},
  title = {Research on a CEF-YOLOv8n-Based Method for Small Object Detection in UAV Aerial Imagery},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {15},
  pages = {4818},
  doi = {10.3390/s26154818}
}
Kostenlos starten →