Facial recognition technology used to discover emotional state of pigs

Artificial Intelligence, News

Scientists are using facial recognition technology to assess pigs’ emotional states in a project to help improve animal wellbeing.

Researchers hope that being able to tell from a pig’s expression if it is content or distressed will lead to the development of a tool that can monitor individual faces and alert farmers to any health and welfare problems.

Pigs are highly expressive and previous research from Scotland’s Rural College (SRUC) has shown they can signal their intentions to other pigs using different facial expressions.

There is also evidence they make different expressions when in pain or under stress.

A 3D camera is used to analyse images of pigs’ faces (Scotland’s Rural College/PA)

In the new project, scientists at SRUC’s Pig Research Centre in Midlothian are capturing 3D and 2D facial images of the breeding sow population under typical situations that are likely to result in different emotional states.

Sows can experience lameness and could show different facial expressions relating to pain before and after being given pain relief, while pigs appear calm and content when satiated, which could be reflected in their facial expressions.

The images are then processed at the University of the West of England Bristol’s Centre for Machine Vision, where various state-of-the-art machine learning techniques are being developed to automatically identify different emotions conveyed by particular facial expressions.

Dr Emma Baxter from SRUC said: “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.

“By focusing on the pig’s face, we hope to deliver a truly animal-centric welfare assessment technique, where the animal can ‘tell’ us how it feels about its own individual experiences and environment.

“This allows insight into both short-term emotional reactions and long-term individual ‘moods’ of animals under our care.”

After validating the techniques, the team will develop the technology for use on farms with commercial partners where individual sows in large herds will be monitored continuously.

Professor Melvyn Smith from UWE Bristol’s Centre for Machine Vision, part of the Bristol Robotics Laboratory, said: “Machine vision technology offers the potential to realise a low-cost, non-intrusive and practical means to biometrically identify individual animals on the farm.

“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.”

The study, funded by the Biotechnology and Biological Sciences Research Council (BBSRC), is also being supported by industry stakeholders JSR Genetics Ltd and Garth Pig Practice as well as precision livestock specialists Agsenze.

Chris Price
For latest tech stories go to TechDigest.tv