EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Broiler Behavior Detection and Tracking Method Based on Lightweight Transformer

Haixia Qi, Zihong Chen, Guangsheng Liang, Riyao Chen, Jinzhuo Jiang, Xiwen Luo

Applied Sciences · 18.03.2025 · 15 (6) · 3333

DOI: 10.3390/app15063333 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 18.03.2025. Verfasst von Haixia Qi und 5 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University Special Discipline Construction Project, Guangdong University Characteristic Innovation Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Detecting the daily behavior of broiler chickens allows early detection of irregular activity patterns and, thus, problems in the flock. In an attempt to resolve the problems of the slow detection speed, low accuracy, and poor generalization ability of traditional detection models in the actual breeding environment, we propose a chicken behavior detection method called FCBD-DETR (Faster Chicken Behavior Detection Transformer). The FasterNet network based on partial convolution (PConv) was used to replace the Resnet18 backbone network to reduce the computational complexity of the model and to improve the speed of model detection. In addition, we propose a new cross-scale feature fusion network to optimize the neck network of the original model. These improvements led to a 78% decrease in the number of parameters and a 68% decrease in GFLOPs. The experimental results show that the proposed model is superior to the traditional network in the speed, accuracy and generalization ability of broiler behavior detection. (1) The detection speed is improved from 49.5 frames per second to 68.5 frames per second, which is 22.6 frames and 10.9 frames higher than Yolov7 and Yolov8, respectively. (2) mAP0.5 reaches 99.4%, and MAP0.5:0.95 increases from 84.9 to 88.4%. (3) Combined with the multi-target tracking algorithm, the chicken flock counting, behavior recognition, and individual tracking tasks are successfully realized.

Förderung

  • University Special Discipline Construction Project
  • Guangdong University Characteristic Innovation Project

Zitieren

APA

Qi, H., Chen, Z., Liang, G., Chen, R., Jiang, J., & Luo, X. (2025). Broiler Behavior Detection and Tracking Method Based on Lightweight Transformer. Applied Sciences, 15(6), 3333. https://doi.org/10.3390/app15063333

Vancouver

Qi H, Chen Z, Liang G, Chen R, Jiang J, Luo X. Broiler Behavior Detection and Tracking Method Based on Lightweight Transformer. Applied Sciences. 2025;15(6):3333. doi:10.3390/app15063333

BibTeX
@article{Qi2025,
  author = {Qi, Haixia and Chen, Zihong and Liang, Guangsheng and Chen, Riyao and Jiang, Jinzhuo and Luo, Xiwen},
  title = {Broiler Behavior Detection and Tracking Method Based on Lightweight Transformer},
  journal = {Applied Sciences},
  year = {2025},
  volume = {15},
  number = {6},
  pages = {3333},
  doi = {10.3390/app15063333}
}
Kostenlos starten →