Why the long face? Device can tell farmers when sheep are sad
KNOWING you have a happy flock is important for any sheep farmer.
But the animals’ lack of obvious facial expressions can make this a tricky task.
Help is at hand, however, from a computer program that can detect whether farm animals or pets are contented or in pain.
The artificially intelligent system has been designed by scientists at Cambridge University to recognise five different facial expressions in sheep and estimate whether an animal is suffering and how severe is its discomfort.
It is hoped cameras can be mounted on water troughs to monitor animals such as sheep to warn farmers if they appear distressed so that diseases within a flock can be picked up early.
It is based on a computer algorithm that scans faces for signs of pain such as eyes narrowing, cheeks tightening, ears folding forwards, lips pulling down and back, and nostrils changing from a U-shape to a V-shape.
It could also be used for other animals such as horses, rodents used in animal research and household pets.
“There’s been much more study over the years with people,” said Prof Peter Robinson, of Cambridge’s Computer Laboratory. “But a lot of the earlier work on the faces of animals was actually done by [Charles] Darwin, who argued that humans and many animals show emotion through remarkably similar behaviours. So, we thought there would likely be crossover between animals and our work in human faces.
“I do a lot of walking in the countryside, and after working on this project, I now often find myself stopping to talk to the sheep and make sure they’re happy,” he added.
“The interesting part is that you can see a clear analogy between actions in sheep faces and similar facial actions in humans when they are in pain – there is a similarity in terms of the muscles,” said co-author Dr Marwa Mahmoud.
To train the computer model, the Cambridge researchers used around 500 photographs of sheep gathered by veterinarians during treatment.
Severe pain in sheep is associated with conditions such as foot rot or mastitis, an inflammation of the udder caused by injury or bacterial infection.
The algorithm will be presented today at the 12th IEEE International Conference on Automatic Face and Gesture Recognition in Washington DC.