A new team of researchers has developed software that uses Artificial Intelligence to identify geese with facial recognition, according to a recent report.
While it may have taken the researchers a few years to complete their work, scientists have created goose recognition software that is now 97% accurate.
The success of the goose recognition software has been shared in the Journal of Ornithology, as the team that created it shared the unique story of how it came to be with NPR.
At the Konrad Lorenz Research Center for Behavior and Cognition in Vienna, Austria, Sonia Kleindorfer was faced with a problem. As the new director of the center, Kleindorfer was faced with a difficult challenge, identifying 30 different Greylag Geese simply by their faces.
While Kleindorfer’s predecessor, the famous Austrian biologist Konrad Lorenz, was able to identify all 30 of the birds, Kleindorfer was unable to.
“I can do five, but when the next five come, I start to have a mental meltdown,” she says. “So I’m actually not as good as I would like to be.”
Looking for a better way to identify the birds, Kleindorfer went to a colleague to see if they could write a program to identify the specific features of the geese, like their beak and facial colorings.
Years later, the program works and uses AI to tell the difference between each unique goose at the center. However, the achievement left researchers wondering what the point was.
“So we have nailed the AI, but then you have to ask yourself ... does it matter in the life of a goose?” she asks.
When it comes to why geese, Kleindorfer says that the animal’s lives are filled with “drama,” as they have “arch rivals, and jealousy and retribution,” NPR reports.
Wanting to know more about how the geese interacted with the drama, Kleindorfer and her team used full-sized cutouts of the animals and paired them with their partner or another member of the flock.
Through the tests, they were able to see that the geese recognized the other birds but not themselves since they can’t see their own reflection.
Knowing that the birds can tell themselves apart confirms the drama that Kleindorfer and others have observed over time.
“This facial recognition, we think, might be a key component in higher-level social organization among unrelated individuals,” Kleindorfer said.