Comparison of architectures and training strategies for convolutional neural networks intended for location-specific counting of slaughter pigs
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Pen fouling is an undesired behaviour seen in growing pigs, where they start resting in the excretion area and excrete in the designated resting area. It is reasonable to assume that automatic monitoring of the location of the pigs within the pen could be used for early warnings of imminent pen fouling events. We intend to provide such automatic monitoring using convolutional neural networks (CNN) applied to images captured above the pens. In this preliminary study, we compared 12 different combinations of CNN architectures and training strategies for this purpose. The best performing strategy yielded an overall mean absolute error of 0.35 pigs and a coefficient of determination of 96% between the predicted and observed number of pigs in a given area of the pen.
| Originalsprog | Engelsk |
|---|---|
| Titel | Precision Livestock Farming 2019 : Papers Presented at the 9th European Conference on Precision Livestock Farming, ECPLF 2019 |
| Redaktører | Bernadette O'Brien, Deirdre Hennessy, Laurence Shalloo |
| Antal sider | 8 |
| Publikationsdato | 2019 |
| Sider | 476-483 |
| ISBN (Elektronisk) | 9781841706542 |
| Status | Udgivet - 2019 |
| Begivenhed | 9th European Conference on Precision Livestock Farming, ECPLF 2019 - Cork, Irland Varighed: 26 aug. 2019 → 29 aug. 2019 |
Konference
| Konference | 9th European Conference on Precision Livestock Farming, ECPLF 2019 |
|---|---|
| Land | Irland |
| By | Cork |
| Periode | 26/08/2019 → 29/08/2019 |
| Sponsor | Agriculture and Food Development Authority (Teagasc), An Roinn Talmhaiochta, Bia agus Mara, Department of Agriculture, Food and the Marine, Dairymaster, et al., SoundTalks, Zoetis |
| Navn | Precision Livestock Farming 2019 - Papers Presented at the 9th European Conference on Precision Livestock Farming, ECPLF 2019 |
|---|
Bibliografisk note
Funding Information:
This research was done with support from The Danish Council for Strategic Research (The PigIT Project, Grant number 11-116191) and from the Green Development and Demonstration Programme under the Ministry of Food, Agriculture and Fisheries, Denmark (project IntactTails j. nr. 34009-13-0743; project StraWell j.nr. 34009-13-0736). We further wish to thank the stockmen at the pig facilities at Aarhus University, Foulum. Lastly, a special thanks to student worker Sally Veronika Hansen for manually and diligently counting the pigs in the several thousand images used in this study.
Publisher Copyright:
© Precision Livestock Farming 2019 - Papers Presented at the 9th European Conference on Precision Livestock Farming, ECPLF 2019. All rights reserved.
ID: 292229330