EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HCF-YOLO: High-Resolution Center-Guided Feature Fusion for Dense Cotton Seedling Detection in UAV Imagery

Yantao Li, Hongbo Su, Wenbin Zhang, Yunjie Zhao iD, Yuxiang Wang

Sensors · 29.10.2026 · 26 (21) · 6842

DOI: 10.3390/s26216842 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 29.10.2026. Verfasst von Yantao Li und 4 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer and Information Engineering, Xinjiang Agricultural University, Urumqi 830052, China. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Xinjiang Talent Development Fund Project, the Major Science and Technology Special Project of the Xinjiang Uygur Autonomous Region.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate estimates of cotton seedling abundance and spatial distribution support emergence assessment and field management. However, seedlings occupy few pixels in unmanned aerial vehicle (UAV) imagery, occur at high densities, and frequently overlap. This study proposes a High-Resolution Center-Guided Feature Fusion model (HCF-YOLO) that combines a P2 detection layer, Lightweight Multi-Scale Feature Fusion (LMSFF), and training-only Gaussian Center Guidance (GCG). The dataset contained 1040 images and 101,824 annotated cotton seedlings and was divided into 832 training, 104 validation, and 104 test images. On the main test split, HCF-YOLO11n achieved an mAP50 of 0.9354 and an mAP50-95 of 0.4431, improvements of 5.10 and 3.74 percentage points over YOLO11n, while reducing the parameter count and FLOPs by 14.18% and 14.63%. The performance ranking was consistent across three random seeds and was retained in the external-source evaluation. Under the pruning protocol, the 5%-pruned HCF-YOLO11n achieved an mAP50-95 of 0.4578 on the same 104-image test split while reducing parameters and FLOPs by 11.19% and 6.87%, respectively. TensorRT FP16 reduced network inference latency to 1.453 ms, and Raspberry Pi 5 deployment reached 6.67 FPS. Overall, HCF-YOLO balances accuracy, model size, and edge efficiency for UAV-assisted cotton seedling recognition and image-plane localization, although onboard validation remains necessary.

Schlagwörter

Stichwörter der Autor:innen

UAV imagerycenter guidancecotton seedling detectionfeature fusionmodel compressionsmall-object detection

Autor:innen und Institutionen

  • College of Computer and Information Engineering, Xinjiang Agricultural University, Urumqi 830052, China.
  • Xinjiang Jinfengyuan Seed Industry Co., Ltd., Wensu County, Aksu Prefecture 843100, China.
  • Engineering Research Center for Intelligent Agriculture, Ministry of Education, Urumqi 830052, China.
  • Xinjiang Agricultural Informatization Engineering Technology Research Center, Xinjiang Agricultural University, Urumqi 830052, China.
  • College of Agriculture, Xinjiang Agricultural University, Urumqi 830052, China.

ORCID-Profile: Yunjie Zhao

Förderung

  • National Natural Science Foundation of China
  • Xinjiang Talent Development Fund Project
  • the Major Science and Technology Special Project of the Xinjiang Uygur Autonomous Region
  • Major Science and Technology Special Project of the Xinjiang Uygur Autonomous Region

Zitieren

APA

Li, Y., Su, H., Zhang, W., Zhao, Y., & Wang, Y. (2026). HCF-YOLO: High-Resolution Center-Guided Feature Fusion for Dense Cotton Seedling Detection in UAV Imagery. Sensors, 26(21), 6842. https://doi.org/10.3390/s26216842

Vancouver

Li Y, Su H, Zhang W, Zhao Y, Wang Y. HCF-YOLO: High-Resolution Center-Guided Feature Fusion for Dense Cotton Seedling Detection in UAV Imagery. Sensors. 2026;26(21):6842. doi:10.3390/s26216842

BibTeX
@article{Li2026,
  author = {Li, Yantao and Su, Hongbo and Zhang, Wenbin and Zhao, Yunjie and Wang, Yuxiang},
  title = {HCF-YOLO: High-Resolution Center-Guided Feature Fusion for Dense Cotton Seedling Detection in UAV Imagery},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {21},
  pages = {6842},
  doi = {10.3390/s26216842}
}
Kostenlos starten →