Broiler detection system for multi-tier broiler house using a cloud platform.
- 1Department of Biosystems Engineering, Seoul National University, Seoul, Korea.
- 2Convergence Major in Global Smart Farm, Seoul National University, Seoul, Korea.
- 3Research Institute of Agriculture and Life Sciences, Seoul National University, Seoul, Korea.
Abstract
Broilers are bred at a higher density than large livestock, and their relatively small size makes monitoring challenging, requiring a reliable, cost-effective automatic monitoring system for precision farming. Particularly in the case of managing large-scale broiler breeding systems, such as multi-tier broiler houses, it is important to reduce investment in computing facilities and to improve productivity through automatic monitoring technology. However, financial ability of farmers to equip and manage the high-specification computing equipment required by these technologies is limited, necessitating development of affordable monitoring systems. We there propose a lightweight broiler-detection model that augments YOLOv5 with two efficiency modules: Bi-directional Feature Pyramid Network (BiFPN), which fuses multi-scale features with learnable weights to sharpen small-object and GhostNet, which replaces many convolutions with inexpensive "ghost" operations to cut parameters. Performance was benchmarked against the original YOLOv5 on a high-spec local GPU machine and CPU based cloud platform. The proposed model attained 90.5% [email protected], marginally below the baseline's 92.8%, yet inference time dropped by 30% on CPU (50.9 ms) and 50% on GPU (1.8 ms). By implementing lightweight models, farmers can utilize monitoring systems from anywhere via mobile devices, leveraging cloud-based technology. This eliminates the necessity for expensive hardware, offering an affordable and practical solution for improving farming management.
Related
- The practicality of implementing a net energy system in the broiler industry.
- Risk Characterization for a Multi‐Pathway Quantitative Risk Assessment Model: A Case Study on Extended‐Spectrum Beta‐Lactamas Producing E. coli Spread From Broiler Production
- Alternative management strategy influences litter microbiome dynamics in a commercial broiler rearing system
- Legume substitution alone is not enough: A multi-criteria LCA of low-soybean meal broiler feed systems
- Integrated multi-omics reveals genistein-mediated protection of broiler intestine against heat stress via metabolic-microbial-barrier crosstalk
BibTeX
@article{yoo2026broiler,
title = {{Broiler detection system for multi-tier broiler house using a cloud platform.}},
author = {Jung-Sang Yoo and Chang-Hyup Lee and Jung-Sun Gloria Kim and JoonYong Kim and Ghiseok Kim and Taehyeong Kim and Joong Yong Rhee},
journal = {Journal of animal science and technology},
year = {2026},
month = sep,
doi = {10.5187/jast.2400235},
url = {https://europepmc.org/article/MED/42898595},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - Broiler detection system for multi-tier broiler house using a cloud platform.
AU - Jung-Sang Yoo
AU - Chang-Hyup Lee
AU - Jung-Sun Gloria Kim
AU - JoonYong Kim
AU - Ghiseok Kim
AU - Taehyeong Kim
AU - Joong Yong Rhee
PY - 2026
DA - 2026/09/30
JO - Journal of animal science and technology
DO - 10.5187/jast.2400235
UR - https://europepmc.org/article/MED/42898595
AB - Broilers are bred at a higher density than large livestock, and their relatively small size makes monitoring challenging, requiring a reliable, cost-effective automatic monitoring system for precision farming. Particularly in the case of managing large-scale broiler breeding systems, such as multi-tier broiler houses, it is important to reduce investment in computing facilities and to improve productivity through automatic monitoring technology. However, financial ability of farmers to equip and manage the high-specification computing equipment required by these technologies is limited, necessitating development of affordable monitoring systems. We there propose a lightweight broiler-detection model that augments YOLOv5 with two efficiency modules: Bi-directional Feature Pyramid Network (BiFPN), which fuses multi-scale features with learnable weights to sharpen small-object and GhostNet, which replaces many convolutions with inexpensive "ghost" operations to cut parameters. Performance was benchmarked against the original YOLOv5 on a high-spec local GPU machine and CPU based cloud platform. The proposed model attained 90.5% [email protected], marginally below the baseline's 92.8%, yet inference time dropped by 30% on CPU (50.9 ms) and 50% on GPU (1.8 ms). By implementing lightweight models, farmers can utilize monitoring systems from anywhere via mobile devices, leveraging cloud-based technology. This eliminates the necessity for expensive hardware, offering an affordable and practical solution for improving farming management.
SN - 2055-0391
ER -