EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An Enhanced YOLOv8n-Based Approach for Pig Behavior Recognition

Jianjun Guo, Yudian Xu, Lijun Lin, Beibei Zhang, Piao Zhou, Shangwen Luo, Yuhan Zhuo, Jingyu Ji, Zhijie Luo, Guangming Cheng

Computers · 08.04.2026 · 15 (4) · 230

DOI: 10.3390/computers15040230 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computers, online erschienen am 08.04.2026. Verfasst von Jianjun Guo und 9 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Special Projects in Key Fields of Ordinary Universities in Guangdong Province, Guangdong Province Rural Science and Technology Commissioner, Guangdong Province Graduate Education Innovation Program Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Pig behavior statistics can reflect their health status. Conventional approaches depend on manual observation to derive behavioral information from video recordings, a process that demands substantial time and human effort. To overcome these limitations in indoor intensive farming environments, this study introduces an effective approach for recognizing pig behaviors, employing an enhanced YOLOv8n architecture. The approach utilizes advanced object detection algorithms to automatically identify pig behaviors, including stand, lie, eat, fight, and tail-bite, from overhead video footage of the enclosure. First, images of daily pig behaviors are collected using cameras to build a pig behavior dataset. To boost detection accuracy, the SE attention mechanism is embedded within the feature extraction backbone of the YOLOv8n network to enhance its representational capacity, strengthening the model’s capacity to grasp overarching contextual information and improve the expressiveness of extracted features. The GIoU loss function is employed during training to reduce computational cost and accelerate model convergence. Moreover, integrating Ghost convolution into the backbone significantly reduces both computational complexity and the total number of parameters. The experimental findings reveal that the optimized YOLOv8n model contains just 1.71 million parameters, marking a 42.93% reduction relative to the baseline model. Its floating-point operations total 5.0 billion, indicating a 38.27% decrease, while the mean average precision (mAP@50) reaches 96.8%, surpassing the original by 2.6 percentage points. Compared with other widely used YOLO-based object detection frameworks, the proposed approach achieves notably higher accuracy while requiring significantly lower computational resources and model complexity.

Förderung

  • Special Projects in Key Fields of Ordinary Universities in Guangdong Province
  • Guangdong Province Rural Science and Technology Commissioner
  • Guangdong Province Graduate Education Innovation Program Project
  • Innovative projects with distinctive features in ordinary universities in Guangdong Province
  • Guangzhou Rural Science and Technology Commissioner Special Project
  • Yunfu 2023 provincial science and technology innovation strategy and rural revitalization strategy project
  • Yunfu City’s 2025 Provincial Science and Technology Support “Hundred, Thousand, and Million Project”

Zitieren

APA

Guo, J., Xu, Y., Lin, L., Zhang, B., Zhou, P., Luo, S., Zhuo, Y., Ji, J., Luo, Z., & Cheng, G. (2026). An Enhanced YOLOv8n-Based Approach for Pig Behavior Recognition. Computers, 15(4), 230. https://doi.org/10.3390/computers15040230

Vancouver

Guo J, Xu Y, Lin L, Zhang B, Zhou P, Luo S, et al. An Enhanced YOLOv8n-Based Approach for Pig Behavior Recognition. Computers. 2026;15(4):230. doi:10.3390/computers15040230

BibTeX
@article{Guo2026,
  author = {Guo, Jianjun and Xu, Yudian and Lin, Lijun and Zhang, Beibei and Zhou, Piao and Luo, Shangwen and Zhuo, Yuhan and Ji, Jingyu and Luo, Zhijie and Cheng, Guangming},
  title = {An Enhanced YOLOv8n-Based Approach for Pig Behavior Recognition},
  journal = {Computers},
  year = {2026},
  volume = {15},
  number = {4},
  pages = {230},
  doi = {10.3390/computers15040230}
}
Kostenlos starten →