EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MCP: Multi-Chicken Pose Estimation Based on Transfer Learning

Cheng Fang iD, Zhenlong Wu, Haikun Zheng iD, Jikang Yang, Chuang Ma, Tiemin Zhang

Animals · 12.06.2024 · 14 (12) · 1774

DOI: 10.3390/ani14121774 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Animals, online erschienen am 12.06.2024. Verfasst von Cheng Fang und 5 weiteren Autor:innen. Beteiligte Einrichtung: College of Engineering, South China Agricultural University, 483 Wushan Road, Guangzhou 510642, China. Die Arbeit wurde bisher 23-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Plan, Guangdong Basic and Applied Basic Research Foundation, Guangdong Province Special Fund for Modern Agricultural Industry Common Key Technology R&D Innovation Team.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Poultry managers can better understand the state of poultry through poultry behavior analysis. As one of the key steps in behavior analysis, the accurate estimation of poultry posture is the focus of this research. This study mainly analyzes a top-down pose estimation method of multiple chickens. Therefore, we propose the “multi-chicken pose” (MCP), a pose estimation system for multiple chickens through deep learning. Firstly, we find the position of each chicken from the image via the chicken detector; then, an estimate of the pose of each chicken is made using a pose estimation network, which is based on transfer learning. On this basis, the pixel error (PE), root mean square error (RMSE), and image quantity distribution of key points are analyzed according to the improved chicken keypoint similarity (CKS). The experimental results show that the algorithm scores in different evaluation metrics are a mean average precision (mAP) of 0.652, a mean average recall (mAR) of 0.742, a percentage of correct keypoints (PCKs) of 0.789, and an RMSE of 17.30 pixels. To the best of our knowledge, this is the first time that transfer learning has been used for the pose estimation of multiple chickens as objects. The method can provide a new path for future poultry behavior analysis

Schlagwörter

Stichwörter der Autor:innen

chickenmulti-objectivepose estimationtop-downtransfer learning

Autor:innen und Institutionen

  • College of Engineering, South China Agricultural University, 483 Wushan Road, Guangzhou 510642, China.
  • National Engineering Research Center for Breeding Swine Industry, Guangzhou 510642, China.
  • Guangdong Laboratory for Lingnan Modern Agriculture, Guangzhou 510642, China.

ORCID-Profile: Cheng Fang, Haikun Zheng

Förderung

  • National Key Research and Development Plan
  • Guangdong Basic and Applied Basic Research Foundation
  • Guangdong Province Special Fund for Modern Agricultural Industry Common Key Technology R&D Innovation Team
  • Project of State Key Laboratory of Swine and Poultry Breeding Industry
  • Guangzhou Science and Technology Planning Project
  • Chaozhou Science and Technology Planning Project
  • Guangdong Province Special Fund for Modern Agricultural Industry Common Key Technology R&D Innovation Team

Zitieren

APA

Fang, C., Wu, Z., Zheng, H., Yang, J., Ma, C., & Zhang, T. (2024). MCP: Multi-Chicken Pose Estimation Based on Transfer Learning. Animals, 14(12), 1774. https://doi.org/10.3390/ani14121774

Vancouver

Fang C, Wu Z, Zheng H, Yang J, Ma C, Zhang T. MCP: Multi-Chicken Pose Estimation Based on Transfer Learning. Animals. 2024;14(12):1774. doi:10.3390/ani14121774

BibTeX
@article{Fang2024,
  author = {Fang, Cheng and Wu, Zhenlong and Zheng, Haikun and Yang, Jikang and Ma, Chuang and Zhang, Tiemin},
  title = {MCP: Multi-Chicken Pose Estimation Based on Transfer Learning},
  journal = {Animals},
  year = {2024},
  volume = {14},
  number = {12},
  pages = {1774},
  doi = {10.3390/ani14121774}
}
Kostenlos starten →