Machine Learning Methods and Visual Observations to Categorize Behavior of Grazing Cattle Using Accelerometer Signals
Sensors · 2024
DOI: 10.3390/s24103171 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, 2024. Verfasst von Ira Lloyd Parsons und 5 weiteren Autor:innen. Beteiligte Einrichtung: Quantitative Ecology and Spatial Technologies Laboratory, Department of Wildlife, Fisheries and Aquaculture, Mississippi State University, Starkville…. Hauptthemen laut MEDLINE: Machine Learning, Accelerometry, Behavior, Animal. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: United States Department of Agriculture, Thinking Like a Mountain to Improve Animal Production Systems Ecology, Energy Budgets, and Mechanistic Models, USDA National Institute of Food and Agriculture.
Abstract (Originalsprache)
Accelerometers worn by animals produce distinct behavioral signatures, which can be classified accurately using machine learning methods such as random forest decision trees. The objective of this study was to identify accelerometer signal separation among parsimonious behaviors. We achieved this objective by (1) describing functional differences in accelerometer signals among discrete behaviors, (2) identifying the optimal window size for signal pre-processing, and (3) demonstrating the number of observations required to achieve the desired level of model accuracy,. Crossbred steers (Bos taurus indicus; n = 10) were fitted with GPS collars containing a video camera and tri-axial accelerometers (read-rate = 40 Hz). Distinct behaviors from accelerometer signals, particularly for grazing, were apparent because of the head-down posture. Increasing the smoothing window size to 10 s improved classification accuracy (p < 0.05), but reducing the number of observations below 50% resulted in a decrease in accuracy for all behaviors (p < 0.05). In-pasture observation increased accuracy and precision (0.05 and 0.08 percent, respectively) compared with animal-borne collar video observations.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Quantitative Ecology and Spatial Technologies Laboratory, Department of Wildlife, Fisheries and Aquaculture, Mississippi State University, Starkville, MS 39762, USA.
- West River Research and Extension Center, Department of Animal Science, South Dakota State University, Rapid City, SD 57703, USA.
- Department of Animal and Dairy Sciences, Mississippi State University, Starkville, MS 39762, USA.
- Texas A&M Natural Resources Institute and Department of Rangeland, Wildlife, and Fisheries Management, Texas A&M University, College Station, TX 77843, USA.
ORCID-Profile: Brandi B Karisch, Stephen L Webb
Förderung
- United States Department of Agriculture
- Thinking Like a Mountain to Improve Animal Production Systems Ecology, Energy Budgets, and Mechanistic Models
- USDA National Institute of Food and Agriculture
Zitieren
APA
Parsons, I. L., Karisch, B. B., Stone, A. E., Webb, S. L., Norman, D. A., & Street, G. M. (2024). Machine Learning Methods and Visual Observations to Categorize Behavior of Grazing Cattle Using Accelerometer Signals. Sensors. https://doi.org/10.3390/s24103171
Vancouver
Parsons IL, Karisch BB, Stone AE, Webb SL, Norman DA, Street GM. Machine Learning Methods and Visual Observations to Categorize Behavior of Grazing Cattle Using Accelerometer Signals. Sensors. 2024. doi:10.3390/s24103171
BibTeX
@article{Parsons2024,
author = {Parsons, Ira Lloyd and Karisch, Brandi B and Stone, Amanda E and Webb, Stephen L and Norman, Durham A and Street, Garrett M},
title = {Machine Learning Methods and Visual Observations to Categorize Behavior of Grazing Cattle Using Accelerometer Signals},
journal = {Sensors},
year = {2024},
doi = {10.3390/s24103171}
}
EUVIMED