Automatic recognition of feeding and foraging behaviour in pigs using deep learning

Automated recognition of postures and drinking behaviour for the detection of compromised health in pigs

“…We did not quantify pig feeding behaviour, as the feeders were outside the camera field of view. However, we recognise that changes in feeding behaviour are one of the most common indicators of compromised health and welfare 5 , 35 . In order to extrapolate this approach to encompass feeding associated behaviours, we may define new classes of images, such as the feeding and non-nutritive visit (foraging) behaviours 35 .…”

“…However, we recognise that changes in feeding behaviour are one of the most common indicators of compromised health and welfare 5 , 35 . In order to extrapolate this approach to encompass feeding associated behaviours, we may define new classes of images, such as the feeding and non-nutritive visit (foraging) behaviours 35 . Practically, this can be done by either utilising transfer learning (i.e., storing knowledge gained while adding extra-classes), or by redefining the dataset and following the methods described in this paper.…”

“…All procedures were conducted in accordance with the Animals (Scientific Procedures) Act 1986, European Directive EU 2010/63, with the approval of Animal Welfare and Ethical Review Body (AWERB) of Newcastle University 10 , 35 .…”

“…Fifteen pigs (Landrace/Large White dams synthetic sire line, Hermitage Seaborough Ltd., North Tawton, UK) were housed, under commercial conditions, in a single, fully-slatted pen ( ) from 9 to 14 weeks of age 35 ; this pig density is equivalent to that required by UK commercial conditions (MAFF 1994). Ear tags allowed pigs to be individually identified.…”

“…Ear tags allowed pigs to be individually identified. Food and water were provided to the pens using 4 drinking sources and 4 feeding troughs respectively 10 , 35 . To match commercial enrichment standards, a hanging chain with plastic pipes was also provided 35 .…”