Daily eating activity of dairy cows from 3D accelerometer data and RFID signals: prediction by random forests and detection of sick cows

Leslie Foldager; Lars Bilde Gildbjerg; Heidi Voss; Philipp Trénel; Lene Munksgaard; Peter T. Thomsen

Daily eating activity of dairy cows from 3D accelerometer data and RFID signals: prediction by random forests and detection of sick cows

Research output: Contribution to book/anthology/report/proceeding › Article in proceedings › Research

Abstract

Feed intake is very important for dairy cows and deviation from normal eating behaviour may predict a cow that needs treatment. We used video recordings of dairy cows at the Danish Cattle Research Centre (DKC) combined with data from a neck-collar mounted 3D accelerometer and RFID device from Lyngsoe Systems (Aars, Denmark) to develop a random forests model for predicting daily eating activity. We investigated performance by internal cross-validation and the results indicate that we obtain accurate predictions of daily eating time by the algorithm. Technical challenges are delaying the planned tests on commercial farms. We are therefore currently utilising historical data from DKC to examine the potential of using changes in daily eating time for detection of sick cows.

Original language	English
Title of host publication	Symposium i Anvendt Statistik
Editors	Peter Linde
Publication date	Jan 2017
Pages	109-122
ISBN (Print)	978-87-501-2267-8
Publication status	Published - Jan 2017
Event	Symposium i Anvendt Statistik - Syddansk Universitet, Odense, Denmark Duration: 23 Jan 2017 → 24 Jan 2017 Conference number: 39

Conference

Conference	Symposium i Anvendt Statistik
Number	39
Location	Syddansk Universitet
Country/Territory	Denmark
City	Odense
Period	23/01/2017 → 24/01/2017

Library access?

Check availability with the Royal Danish Library

COWTrack: GUDP COWTrack
Sørensen, J. T. (Project manager), Thomsen, P. T. (Project manager), Munksgaard, L. (Participant), Oliveira, V. H. S. D. (Participant) & Voss, H. (Participant)
Grønt Udviklings- og Demonstrationsprogram (GUDP)
01/04/2013 → 31/03/2016
Project: Research

Symposium i Anvendt Statistik
Foldager, L. (Speaker)
23 Jan 2017 → 24 Jan 2017
Activity: Participating in or organising an event types › Participation in or organisation af a conference

Cite this

@inproceedings{9f097cb83e194535b97fb4cc75948351,

title = "Daily eating activity of dairy cows from 3D accelerometer data and RFID signals: prediction by random forests and detection of sick cows",

abstract = "Feed intake is very important for dairy cows and deviation from normal eating behaviour may predict a cow that needs treatment. We used video recordings of dairy cows at the Danish Cattle Research Centre (DKC) combined with data from a neck-collar mounted 3D accelerometer and RFID device from Lyngsoe Systems (Aars, Denmark) to develop a random forests model for predicting daily eating activity. We investigated performance by internal cross-validation and the results indicate that we obtain accurate predictions of daily eating time by the algorithm. Technical challenges are delaying the planned tests on commercial farms. We are therefore currently utilising historical data from DKC to examine the potential of using changes in daily eating time for detection of sick cows.",

author = "Leslie Foldager and Gildbjerg, \{Lars Bilde\} and Heidi Voss and Philipp Tr{\'e}nel and Lene Munksgaard and Thomsen, \{Peter T.\}",

year = "2017",

month = jan,

language = "English",

isbn = "978-87-501-2267-8",

pages = "109--122",

editor = "Peter Linde",

booktitle = "Symposium i Anvendt Statistik",

note = "Symposium i Anvendt Statistik ; Conference date: 23-01-2017 Through 24-01-2017",

}

Daily eating activity of dairy cows from 3D accelerometer data and RFID signals: prediction by random forests and detection of sick cows. / Foldager, Leslie ; Gildbjerg, Lars Bilde; Voss, Heidi et al.
Symposium i Anvendt Statistik. ed. / Peter Linde. 2017. p. 109-122.

Research output: Contribution to book/anthology/report/proceeding › Article in proceedings › Research

TY - GEN

T1 - Daily eating activity of dairy cows from 3D accelerometer data and RFID signals: prediction by random forests and detection of sick cows

AU - Foldager, Leslie

AU - Gildbjerg, Lars Bilde

AU - Voss, Heidi

AU - Trénel, Philipp

AU - Munksgaard, Lene

AU - Thomsen, Peter T.

N1 - Conference code: 39

PY - 2017/1

Y1 - 2017/1

N2 - Feed intake is very important for dairy cows and deviation from normal eating behaviour may predict a cow that needs treatment. We used video recordings of dairy cows at the Danish Cattle Research Centre (DKC) combined with data from a neck-collar mounted 3D accelerometer and RFID device from Lyngsoe Systems (Aars, Denmark) to develop a random forests model for predicting daily eating activity. We investigated performance by internal cross-validation and the results indicate that we obtain accurate predictions of daily eating time by the algorithm. Technical challenges are delaying the planned tests on commercial farms. We are therefore currently utilising historical data from DKC to examine the potential of using changes in daily eating time for detection of sick cows.

AB - Feed intake is very important for dairy cows and deviation from normal eating behaviour may predict a cow that needs treatment. We used video recordings of dairy cows at the Danish Cattle Research Centre (DKC) combined with data from a neck-collar mounted 3D accelerometer and RFID device from Lyngsoe Systems (Aars, Denmark) to develop a random forests model for predicting daily eating activity. We investigated performance by internal cross-validation and the results indicate that we obtain accurate predictions of daily eating time by the algorithm. Technical challenges are delaying the planned tests on commercial farms. We are therefore currently utilising historical data from DKC to examine the potential of using changes in daily eating time for detection of sick cows.

M3 - Article in proceedings

SN - 978-87-501-2267-8

SP - 109

EP - 122

BT - Symposium i Anvendt Statistik

A2 - Linde, Peter

T2 - Symposium i Anvendt Statistik

Y2 - 23 January 2017 through 24 January 2017

ER -

Daily eating activity of dairy cows from 3D accelerometer data and RFID signals: prediction by random forests and detection of sick cows

Abstract

Conference

Library access?

Fingerprint

Projects

COWTrack: GUDP COWTrack

Activities

Symposium i Anvendt Statistik

Cite this