Evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour

Automated behavioural classification and identification through sensors has the potential to improve health and welfare of the animals. Position of a sensor, sampling frequency and window size of segmented signal data has a major impact on classification accuracy in activity recognition and energy n...

Full description

Bibliographic Details
Main Authors: Walton, Emily, Casey, Christy, Mitsch, Jurgen, Vázquez-Diosdado, Jorge A., Yan, Juan, Dottorini, Tania, Ellis, Keith A., Winterlich, Anthony, Kaler, Jasmeet
Format: Article
Published: Royal Society 2018
Subjects:
Online Access:https://eprints.nottingham.ac.uk/49686/
_version_ 1848798054989168640
author Walton, Emily
Casey, Christy
Mitsch, Jurgen
Vázquez-Diosdado, Jorge A.
Yan, Juan
Dottorini, Tania
Ellis, Keith A.
Winterlich, Anthony
Kaler, Jasmeet
author_facet Walton, Emily
Casey, Christy
Mitsch, Jurgen
Vázquez-Diosdado, Jorge A.
Yan, Juan
Dottorini, Tania
Ellis, Keith A.
Winterlich, Anthony
Kaler, Jasmeet
author_sort Walton, Emily
building Nottingham Research Data Repository
collection Online Access
description Automated behavioural classification and identification through sensors has the potential to improve health and welfare of the animals. Position of a sensor, sampling frequency and window size of segmented signal data has a major impact on classification accuracy in activity recognition and energy needs for the sensor, yet, there are no studies in precision livestock farming that have evaluated the effect of all these factors simultaneously. The aim of this study was to evaluate the effects of position (ear and collar), sampling frequency (8, 16 and 32 Hz) of a triaxial accelerometer and gyroscope sensor and window size (3, 5 and 7 s) on the classification of important behaviours in sheep such as lying, standing and walking. Behaviours were classified using a random forest approach with 44 feature characteristics. The best performance for walking, standing and lying classification in sheep (accuracy 95%, F-score 91%–97%) was obtained using combination of 32 Hz, 7 s and 32 Hz, 5 s for both ear and collar sensors, although, results obtained with 16 Hz and 7 s window were comparable with accuracy of 91%–93% and F-score 88%–95%. Energy efficiency was best at a 7 s window. This suggests that sampling at 16 Hz with 7 s window will offer benefits in a real-time behavioural monitoring system for sheep due to reduced energy needs.
first_indexed 2025-11-14T20:13:41Z
format Article
id nottingham-49686
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T20:13:41Z
publishDate 2018
publisher Royal Society
recordtype eprints
repository_type Digital Repository
spelling nottingham-496862020-05-04T19:31:26Z https://eprints.nottingham.ac.uk/49686/ Evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour Walton, Emily Casey, Christy Mitsch, Jurgen Vázquez-Diosdado, Jorge A. Yan, Juan Dottorini, Tania Ellis, Keith A. Winterlich, Anthony Kaler, Jasmeet Automated behavioural classification and identification through sensors has the potential to improve health and welfare of the animals. Position of a sensor, sampling frequency and window size of segmented signal data has a major impact on classification accuracy in activity recognition and energy needs for the sensor, yet, there are no studies in precision livestock farming that have evaluated the effect of all these factors simultaneously. The aim of this study was to evaluate the effects of position (ear and collar), sampling frequency (8, 16 and 32 Hz) of a triaxial accelerometer and gyroscope sensor and window size (3, 5 and 7 s) on the classification of important behaviours in sheep such as lying, standing and walking. Behaviours were classified using a random forest approach with 44 feature characteristics. The best performance for walking, standing and lying classification in sheep (accuracy 95%, F-score 91%–97%) was obtained using combination of 32 Hz, 7 s and 32 Hz, 5 s for both ear and collar sensors, although, results obtained with 16 Hz and 7 s window were comparable with accuracy of 91%–93% and F-score 88%–95%. Energy efficiency was best at a 7 s window. This suggests that sampling at 16 Hz with 7 s window will offer benefits in a real-time behavioural monitoring system for sheep due to reduced energy needs. Royal Society 2018-02-07 Article PeerReviewed Walton, Emily, Casey, Christy, Mitsch, Jurgen, Vázquez-Diosdado, Jorge A., Yan, Juan, Dottorini, Tania, Ellis, Keith A., Winterlich, Anthony and Kaler, Jasmeet (2018) Evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour. Royal Society Open Science, 5 (2). p. 171442. ISSN 2054-5703 sheep behaviour classification algorithm accelerometer and gyroscope sensor signal processing precision livestock monitoring http://rsos.royalsocietypublishing.org/content/5/2/171442 doi:10.1098/rsos.171442 doi:10.1098/rsos.171442
spellingShingle sheep behaviour
classification algorithm
accelerometer and gyroscope
sensor
signal processing
precision livestock monitoring
Walton, Emily
Casey, Christy
Mitsch, Jurgen
Vázquez-Diosdado, Jorge A.
Yan, Juan
Dottorini, Tania
Ellis, Keith A.
Winterlich, Anthony
Kaler, Jasmeet
Evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour
title Evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour
title_full Evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour
title_fullStr Evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour
title_full_unstemmed Evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour
title_short Evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour
title_sort evaluation of sampling frequency, window size and sensor position for classification of sheep behaviour
topic sheep behaviour
classification algorithm
accelerometer and gyroscope
sensor
signal processing
precision livestock monitoring
url https://eprints.nottingham.ac.uk/49686/
https://eprints.nottingham.ac.uk/49686/
https://eprints.nottingham.ac.uk/49686/