EUVIMEDEUVIMED entdecken
Open Access · CC BY 3.0

Enhanced YOLOv8-Based Real-Time Navel Orange Detection for UAV Aerial Imaging in Complex Orchard Environments

Haiping Li, Xingzhen Tao

International Journal of Cognitive Informatics and Natural Intelligence · 10.04.2026 · 20 (1) · 1-13

DOI: 10.4018/ijcini.406955 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Cognitive Informatics and Natural Intelligence, erschienen am 10.04.2026. Verfasst von Haiping Li und 1 weiteren Autor:innen. Die Arbeit nennt 16 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 3.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To address low accuracy and high missed detection in navel orange identification from unmanned aerial vehicle (UAV) images, this study proposes an enhanced You Only Look Once version 8 (YOLOv8) method. A large-scale dataset was constructed using UAV-captured orchard images and public citrus data and enhanced through rotation, brightness adjustment, and mosaic techniques. The YOLOv8n model was optimized by (a) integrating an improved Convolutional Block Attention Module (CBAM) before feature fusion to enhance focus on dense and occluded targets and (b) replacing Complete Intersection over Union with Wise-Intersection over Union (Wise-IoU) loss to accelerate convergence by 15% through a dynamic non-monotonic focusing mechanism. Experimental results demonstrate significant improvements: precision of 76.38%, recall of 58.36%, F1-score of 66.16%, and mean average precision (mAP@50) of 66.28%, outperforming baseline models. Ablation studies confirm the complementary benefits of CBAM (+2.26% mAP@50) and Wise-IoU (+4.44% mAP@50). Deployment on UAV platforms achieved 28 frames per second, meeting real-time requirements. This work provides an effective solution for intelligent orchard management with substantial precision agriculture.

Zitieren

APA

Li, H., & Tao, X. (2026). Enhanced YOLOv8-Based Real-Time Navel Orange Detection for UAV Aerial Imaging in Complex Orchard Environments. International Journal of Cognitive Informatics and Natural Intelligence, 20(1), 1-13. https://doi.org/10.4018/ijcini.406955

Vancouver

Li H, Tao X. Enhanced YOLOv8-Based Real-Time Navel Orange Detection for UAV Aerial Imaging in Complex Orchard Environments. International Journal of Cognitive Informatics and Natural Intelligence. 2026;20(1):1-13. doi:10.4018/ijcini.406955

BibTeX
@article{Li2026,
  author = {Li, Haiping and Tao, Xingzhen},
  title = {Enhanced YOLOv8-Based Real-Time Navel Orange Detection for UAV Aerial Imaging in Complex Orchard Environments},
  journal = {International Journal of Cognitive Informatics and Natural Intelligence},
  year = {2026},
  volume = {20},
  number = {1},
  pages = {1--13},
  doi = {10.4018/ijcini.406955}
}
Kostenlos starten →