A research project in Scotland aims to see if facial recognition software could be used to identify the specific emotional state of pigs.
Based at Scotland’s Rural College (SRUC) Pig Research Centre in Midlothian, it is hoped that the technology could be used to improve pig welfare and production outcomes.
The project aims to see if it is possible to detect a pig’s emotional state through facial recognition. If successful, the technology could be pared with software to monitor pigs and trigger an alert when health or welfare issues arise.
Work previously carried out by SRUC demonstrated that pigs are capable of communicating with each other, and have a range of different facial expressions.
The current project is using 2D and 3D imaging to capture facial images of breeding sows. The resulting images are then processed by the University of the West of England (UWE Bristol), which has automated the process of identifying emotions.
One area which could benefit from the use of such technology is the identification and treatment of pain.
“Machine vision technology offers the potential to realise a low-cost, non-intrusive and practical means to biometrically identify individual animals on the farm,” said Professor Melvyn Smith from the University of West England’s Centre for Machine Vision.
“Our work has already demonstrated a 97% accuracy at facial recognition in pigs. Our next step will be, for the first time, to explore the potential for using machine vision to automatically recognise facial expressions that are linked with core emotion states, such as happiness or distress, in the identified pigs.”
According to Dr Emma Baxter from Scotland’s Rural Centre, the technology could bring a financial benefit to farmers.
“Early identification of pig health issues gives farmers the potential to improve animal wellbeing by tackling any problems quickly and implementing tailored treatment for individuals.
“This will reduce production costs by preventing impact of health issues on performance.”
Ewe Look Nice
A 2017 study in the UK revealed that sheep are capable of identifying specific human faces.
Cambridge University researchers successfully trained sheep to distinguish between the profiles of actors Jake Gyllenhaal and Emma Watson, former US President Barack Obama and The Antiques Roadshow host Fiona Bruce.
A flock of eight Welsh Mountain ewes were trained to distinguish the celebrities from unfamiliar faces by using food as a reward. Researchers displayed photos on computer screens and the sheep indicated their choice by nudging the release of a treat.
After establishing that the ewes were capable of recognising specific human faces, researchers tested the sheep to see if they could recognise the celebrities from different angles, and whether they could identify their own handlers when included with a line up of random faces. On both tests, the ewes performed significantly above chance.
The research showed that the facial recognition skills of sheep are similar to those of monkeys, apes and humans. As well as showing potential for improvements in sheep welfare, the research could provide insight into neurodegenerative diseases such as Huntington’s and Parkinson’s.