Taking the labour out of labour detection – a computer vision approach
Publication Date
December 10, 2025
Creator
Abstract
Difficulty during parturition has negative consequences for both animal welfare and farm profitability on cattle farms. Early intervention has been shown to be a key requirement for minimising losses caused by calving problems, however, this requires high levels of monitoring to ensure that problems are promptly detected and resolved. Current methods of detecting calving either rely on labour intensive manual observations or cow-based sensors which are often expensive and require the placement of physical sensors on each animal. In this study, we explore the potential of computer vision technology to accurately predict calving in dairy cattle and highlighted the advantages of both object and pose detection approaches in comparison to current methods.
Item Type
ethesis
Thesis Type
MRes
Supervisors
Subjects (LC)
Associated Schools / Departments
School of Veterinary Medicine and Science
eprints ID
78569
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
MilesBenjamin20133624Corrections.pdf
Type
Full-text
Description
Examined. Second submission following corrections
Size
2.71 MB
Format
Adobe PDF
Checksum (MD5)
bfa06aecba3333254f8268d650f9a541