Development of an electronic device for automatic and individualized monitoring of enteric methane emissions in dairy cows
Revista Colombiana de Ciencias Pecuarias · 2025
DOI: 10.17533/udea.rccp.e358533 ↗Kurz gefasst
Fachartikel in der Zeitschrift Revista Colombiana de Ciencias Pecuarias, 2025. Verfasst von John-Fredy Ramirez-Agudelo und 3 weiteren Autor:innen. Die Arbeit nennt 23 Literaturangaben.
Abstract (Originalsprache)
Background
Monitoring enteric methane (CH₄) emissions is crucial for identifying animals with lower emissions in selection programs and to measure the effectiveness of emission reduction strategies. Current methods are often expensive and complex, limiting their widespread application.
Objective
This study aimed to develop and test a low-cost, automated [1]system for individualized monitoring of CH₄ emissions in dairy cows.
Methods
The system comprises a CH₄ concentration measurement device based on the MQ-4 gas sensor, complemented by a 2 L/min airflow system, and an animal identification module utilizing artificial intelligence. The CH₄ data were wirelessly transmitted via an ESP8266 module to a laptop for storage. CH₄ concentrations were recorded three times per second, and precise timestamps were used to document cow entry and exit from the milking stall. For the animal identification module, video frames of 26 cows during milking were extracted and organized into individual folders for each cow. Four versions (s, n, m, and l) of the Yolov8 and Yolov10 models were fine-tuned and evaluated using a dataset divided into training, validation, and testing sets. Performance metrics included Precision, Recall, F1-Score, and Accuracy. The CH₄ concentration system was tested with 10 Holstein cows during their milking sessions.
Results
The prototypes successfully measured and recorded CH₄ emissions from individual cows. Continuous recording allowed for detailed time-series graphs, showing fluctuations in emissions. Some cows exhibited highest average CH₄ emission level, demonstrating the device's ability to identify high-emitting individuals. Baseline CH₄ concentrations in the feeder area were stable across cows, ensuring accurate emission measurements. The identification module's comparative analysis highlighted the Yolov8s model as the optimal choice due to its balance between low latency (16.4 ms) and high performance, achieving perfect scores in precision, recall, F1-score, and accuracy.
Conclusions
The developed system effectively monitors CH₄ emissions in dairy cows, offering a practical and economical alternative to traditional methods. The use of low-cost sensors and advanced artificial intelligence enhances its potential for genetic improvement programs and sustainable livestock management practices.
Zitieren
APA
Ramirez-Agudelo, J., Bedoya-Mazo, S., Moreno-Pulgarín, L., & Guarin, J. (2025). Development of an electronic device for automatic and individualized monitoring of enteric methane emissions in dairy cows. Revista Colombiana de Ciencias Pecuarias. https://doi.org/10.17533/udea.rccp.e358533
Vancouver
Ramirez-Agudelo J, Bedoya-Mazo S, Moreno-Pulgarín L, Guarin J. Development of an electronic device for automatic and individualized monitoring of enteric methane emissions in dairy cows. Revista Colombiana de Ciencias Pecuarias. 2025. doi:10.17533/udea.rccp.e358533
BibTeX
@article{RamirezAgudelo2025,
author = {Ramirez-Agudelo, John-Fredy and Bedoya-Mazo, Sebastian and Moreno-Pulgarín, Luisa-Fernanda and Guarin, Jose-Fernando},
title = {Development of an electronic device for automatic and individualized monitoring of enteric methane emissions in dairy cows},
journal = {Revista Colombiana de Ciencias Pecuarias},
year = {2025},
doi = {10.17533/udea.rccp.e358533}
}
EUVIMED