Aarhus Universitets segl

Kaitlin Elizabeth Wurtz

Assessment of open-source programs for automated tracking of individual pigs within a group

Publikation: Bidrag til bog/antologi/rapport/proceedingBidrag til bog/antologiForskningpeer review

As researchers search for tools to automate individual animal tracking, particularly in the livestock industry, they come across papers that make strong claims about the ability of such programs to accurately track individual animals within group-housed laboratory settings and imply that this ability will translate to other situations and species. We selected four of these programs (idTracker, ToxTrac, BioTracker, and Ctrax) and tested their ability to track single, pairs, and groups of pigs in an indoor pen typical of commercial farms and a pen modified to improve lighting, contrast, and camera location. We concluded that at present, these four tracking programs do not perform robustly enough to be adapted for use processing videos from commercial-style pig farms, including smaller research farms. It is hoped that identifying limitations of detection programs can help make systems more robust and user-friendly for future use.
OriginalsprogEngelsk
TitelPractical Precision Livestock Farming : Hands-on experiences with PLF technologies in commercial and R&D settings
RedaktørerT. Banhazi, V. Halas, F. Maroto-Molina
UdgivelsesstedWageningen
ForlagWageningen Academic Publicers
Udgivelsesår9 sep. 2022
Sider213-230
Kapitel13
ISBN (trykt)978-90-8686-382-2
ISBN (Elektronisk)978-90-8686-934-3
DOI
StatusUdgivet - 9 sep. 2022

Se relationer på Aarhus Universitet Citationsformater

ID: 302200552