Tracking multiple cows simultaneously in barns using computer vision and deep learning

C. Kamphuis; I. Adriaens; W. Ouweltjes; I. Hulsegge

doi:10.3920/978-90-8686-940-4_139

Tracking multiple cows simultaneously in barns using computer vision and deep learning

C. Kamphuis, I. Adriaens, W. Ouweltjes, I. Hulsegge

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › Academic

Abstract

This study investigated the automated tracking of multiple cows simultaneously using computer vision and deep learning. Video clips were collected in 2019 at Dairy Campus, where cows were housed in small groups (n=16). A systematic approach covering the true variability of barn circumstances eventually resulted in the selection of 159 frames that were annotated by drawing bounding boxes around each cow. These frames were used to retrain and test four You Only Look Once version 5 (YOLOv5) models to automatically detect cows. The weights of the best performing YOLOv5 model were used to parametrize the deep learning algorithm DeepSORT to track multiple cows simultaneously. This algorithm was applied to a 10 min timeframe of a randomly selected video clip and evaluated by computing the multi-object tracking accuracy, which was 92.8%. This outcome is a promising and essential step towards automated monitoring of individual behaviour of group-housed cows.

Original language	English
Title of host publication	Proceedings of 12th World Congress on Genetics Applied to Livestock Production (WCGALP)
Subtitle of host publication	Technical and species orientated innovations in animal breeding, and contribution of genetics to solving societal challenges
Editors	R.F. Veerkamp, Y. de Haas
Place of Publication	Wageningen
Publisher	Wageningen Academic Publishers
Pages	606-609
ISBN (Electronic)	9789086869404
DOIs	https://doi.org/10.3920/978-90-8686-940-4_139
Publication status	Published - 2022
Event	World Congress on Genetics Applied to Livestock Production: WCGALP 2022 - Rotterdam, Netherlands Duration: 3 Jul 2022 → 8 Jul 2022

Conference

Conference	World Congress on Genetics Applied to Livestock Production: WCGALP 2022
Country/Territory	Netherlands
City	Rotterdam
Period	3/07/22 → 8/07/22

Access to Document

10.3920/978-90-8686-940-4_139Licence: CC BY

https://edepot.wur.nl/588202Licence: CC BY

Cite this

Kamphuis, C., Adriaens, I., Ouweltjes, W., & Hulsegge, I. (2022). Tracking multiple cows simultaneously in barns using computer vision and deep learning. In R. F. Veerkamp, & Y. de Haas (Eds.), Proceedings of 12th World Congress on Genetics Applied to Livestock Production (WCGALP): Technical and species orientated innovations in animal breeding, and contribution of genetics to solving societal challenges (pp. 606-609). Article 139 Wageningen Academic Publishers. https://doi.org/10.3920/978-90-8686-940-4_139

Kamphuis, C. ; Adriaens, I. ; Ouweltjes, W. et al. / Tracking multiple cows simultaneously in barns using computer vision and deep learning. Proceedings of 12th World Congress on Genetics Applied to Livestock Production (WCGALP): Technical and species orientated innovations in animal breeding, and contribution of genetics to solving societal challenges. editor / R.F. Veerkamp ; Y. de Haas. Wageningen : Wageningen Academic Publishers, 2022. pp. 606-609

@inproceedings{e3fa8359e85241cb8774d4733d1a0e2c,

title = "Tracking multiple cows simultaneously in barns using computer vision and deep learning",

abstract = "This study investigated the automated tracking of multiple cows simultaneously using computer vision and deep learning. Video clips were collected in 2019 at Dairy Campus, where cows were housed in small groups (n=16). A systematic approach covering the true variability of barn circumstances eventually resulted in the selection of 159 frames that were annotated by drawing bounding boxes around each cow. These frames were used to retrain and test four You Only Look Once version 5 (YOLOv5) models to automatically detect cows. The weights of the best performing YOLOv5 model were used to parametrize the deep learning algorithm DeepSORT to track multiple cows simultaneously. This algorithm was applied to a 10 min timeframe of a randomly selected video clip and evaluated by computing the multi-object tracking accuracy, which was 92.8%. This outcome is a promising and essential step towards automated monitoring of individual behaviour of group-housed cows.",

author = "C. Kamphuis and I. Adriaens and W. Ouweltjes and I. Hulsegge",

year = "2022",

doi = "10.3920/978-90-8686-940-4_139",

language = "English",

pages = "606--609",

editor = "R.F. Veerkamp and {de Haas}, Y.",

booktitle = "Proceedings of 12th World Congress on Genetics Applied to Livestock Production (WCGALP)",

publisher = "Wageningen Academic Publishers",

address = "Netherlands",

note = "World Congress on Genetics Applied to Livestock Production: WCGALP 2022 ; Conference date: 03-07-2022 Through 08-07-2022",

}

Kamphuis, C , Adriaens, I , Ouweltjes, W & Hulsegge, I 2022, Tracking multiple cows simultaneously in barns using computer vision and deep learning. in RF Veerkamp & Y de Haas (eds), Proceedings of 12th World Congress on Genetics Applied to Livestock Production (WCGALP): Technical and species orientated innovations in animal breeding, and contribution of genetics to solving societal challenges., 139, Wageningen Academic Publishers, Wageningen, pp. 606-609, World Congress on Genetics Applied to Livestock Production: WCGALP 2022, Rotterdam, Netherlands, 3/07/22. https://doi.org/10.3920/978-90-8686-940-4_139

Tracking multiple cows simultaneously in barns using computer vision and deep learning. / Kamphuis, C.; Adriaens, I.; Ouweltjes, W. et al.
Proceedings of 12th World Congress on Genetics Applied to Livestock Production (WCGALP): Technical and species orientated innovations in animal breeding, and contribution of genetics to solving societal challenges. ed. / R.F. Veerkamp; Y. de Haas. Wageningen: Wageningen Academic Publishers, 2022. p. 606-609 139.

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › Academic

TY - GEN

T1 - Tracking multiple cows simultaneously in barns using computer vision and deep learning

AU - Kamphuis, C.

AU - Adriaens, I.

AU - Ouweltjes, W.

AU - Hulsegge, I.

PY - 2022

Y1 - 2022

N2 - This study investigated the automated tracking of multiple cows simultaneously using computer vision and deep learning. Video clips were collected in 2019 at Dairy Campus, where cows were housed in small groups (n=16). A systematic approach covering the true variability of barn circumstances eventually resulted in the selection of 159 frames that were annotated by drawing bounding boxes around each cow. These frames were used to retrain and test four You Only Look Once version 5 (YOLOv5) models to automatically detect cows. The weights of the best performing YOLOv5 model were used to parametrize the deep learning algorithm DeepSORT to track multiple cows simultaneously. This algorithm was applied to a 10 min timeframe of a randomly selected video clip and evaluated by computing the multi-object tracking accuracy, which was 92.8%. This outcome is a promising and essential step towards automated monitoring of individual behaviour of group-housed cows.

AB - This study investigated the automated tracking of multiple cows simultaneously using computer vision and deep learning. Video clips were collected in 2019 at Dairy Campus, where cows were housed in small groups (n=16). A systematic approach covering the true variability of barn circumstances eventually resulted in the selection of 159 frames that were annotated by drawing bounding boxes around each cow. These frames were used to retrain and test four You Only Look Once version 5 (YOLOv5) models to automatically detect cows. The weights of the best performing YOLOv5 model were used to parametrize the deep learning algorithm DeepSORT to track multiple cows simultaneously. This algorithm was applied to a 10 min timeframe of a randomly selected video clip and evaluated by computing the multi-object tracking accuracy, which was 92.8%. This outcome is a promising and essential step towards automated monitoring of individual behaviour of group-housed cows.

U2 - 10.3920/978-90-8686-940-4_139

DO - 10.3920/978-90-8686-940-4_139

M3 - Conference paper

SP - 606

EP - 609

BT - Proceedings of 12th World Congress on Genetics Applied to Livestock Production (WCGALP)

A2 - Veerkamp, R.F.

A2 - de Haas, Y.

PB - Wageningen Academic Publishers

CY - Wageningen

T2 - World Congress on Genetics Applied to Livestock Production: WCGALP 2022

Y2 - 3 July 2022 through 8 July 2022

ER -

Kamphuis C , Adriaens I , Ouweltjes W , Hulsegge I. Tracking multiple cows simultaneously in barns using computer vision and deep learning. In Veerkamp RF, de Haas Y, editors, Proceedings of 12th World Congress on Genetics Applied to Livestock Production (WCGALP): Technical and species orientated innovations in animal breeding, and contribution of genetics to solving societal challenges. Wageningen: Wageningen Academic Publishers. 2022. p. 606-609. 139 doi: 10.3920/978-90-8686-940-4_139

Tracking multiple cows simultaneously in barns using computer vision and deep learning

Abstract

Conference

Access to Document

Fingerprint

Cite this