Facial recognition technology might help monitor animal moods

It’s 2019 and science hasn’t even figured out how to talk to animals yet. What gives?

Fortunately, dedicated and resourceful researchers around the world have greatly improved our understanding of animal behavior, to the point where we can often figure out what our pets and livestock are expressing, even if they can’t talk to us.

One of the ways they do this: facial recognition software.

A new study at Scotland’s Rural College is facial recognition technology, for example, to examine subtle changes in pigs’ faces. These expressions are often the first signs of potential health issues.

“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,” said the college’s Emma Baxter. “This will reduce production costs by preventing the impact of health issues on performance.”

Baxter and her team are working with machine vision experts from the University of West England, and they chose pigs as test subjects because of the animals’ highly expressive nature — pigs have been observed before to use facial expressions to communicate to each other.

Using 3D and 2D images of breeding sow faces in typical commercial situations, the researchers analyze for signs of health and wellbeing issues. How do the faces look before and after the sow is given pain relief? Before and after eating?

The images are processed at a state-of-the-art machine vision center, where different facial expressions are linked to emotions. As the center builds up a library of facial expressions and what they mean, it hopes to develop technology that can be applied on the farm, granting farmers greater insight into the welfare of their herds.

“Our work has already demonstrated a 97% accuracy at facial recognition in pigs,” said Melvyn Smith of the University of West England.

As with most fields, more knowledge is more power when it comes to farming.

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