Using AI to follow one goose in a flock could help with conservation efforts
ARI SHAPIRO, HOST:
Artificial intelligence is also coming to the animal kingdom. As NPR's Geoff Brumfiel reports, some researchers are starting to use advanced facial recognition techniques to track species.
GEOFF BRUMFIEL, BYLINE: This story is about facial recognition for geese. Yes, that's right. Someone's actually developed an AI tool for goose faces. Her name is Sonia Kleindorfer. She's a biologist at the University of Vienna.
SONIA KLEINDORFER: And I'm the director of the Konrad Lorenz Research Center for Behavior and Cognition.
BRUMFIEL: It's that job - directing the Konrad Lorenz Research Center - that set everything into motion. You see, Lorenz was a famous Austrian biologist who spent much of his career studying the behavior of local graylag geese.
(SOUNDBITE OF ARCHIVED RECORDING)
UNIDENTIFIED NARRATOR: For more than 30 years, Dr. Lorenz has literally lived with flocks of birds, trying to understand their life and their world.
BRUMFIEL: Famously, he could identify each individual in the flock by looking at their goosey face. And when Kleindorfer got the director job a few years back, she felt pressure she should learn to do it too, but she kept getting confused.
KLEINDORFER: I can do five. But then when the next five come in, I'm just like, oh - you know, I start to have, like, a mental meltdown. So I'm actually not as good as I would like to be.
BRUMFIEL: It was embarrassing, frankly. So she contacted a more technically minded colleague and asked him...
KLEINDORFER: Would you be able to, you know, write a program to distinguish these faces?
BRUMFIEL: He said yes, but he'd need a database of geese photos to work with. Kleindorfer got her team out there, snapping pics of the geese from every angle. After building the database, they wrote a piece of facial recognition AI that could ID a goose by looking at specific features of its beak. It took a couple of years, but writing in the Journal of Ornithology, the team reports that their goose recognition software is now about 97% accurate.
KLEINDORFER: So we had nailed the AI. But, you know, then you have to ask yourself, well, does it matter in the life of a goose?
BRUMFIEL: And here, let's just take a moment to talk about the lives of geese. Because if you, like me, have lived around them, you know they are not the most pleasant animals. Kleindorfer says that's in part because they have a lot going on.
KLEINDORFER: Geese have such drama - there's archrivals and jealousy and retribution.
BRUMFIEL: I mean, does this help explain why geese are such jerks to everyone else?
KLEINDORFER: (Laughter) No. No, I would not use those words.
BRUMFIEL: In the same study, Kleindorfer showed evidence that geese seem to recognize photos of their partners and friends. She thinks that makes sense because all that goose drama can only happen if they can tell each other apart.
KLEINDORFER: This facial recognition, we think, might be a key component in higher-level social organization among unrelated individuals.
BRUMFIEL: That's probably why humans are good at recognizing each other, too. This is not the first facial recognition for animals. In recent years, researchers have used it on everything from lemurs to bears.
KRISTA INGRAM: One thing that's really exciting about this is that it's AI really being used for good.
BRUMFIEL: Krista Ingram is a biologist at Colgate University in New York who has developed SealNet, an AI tool that can tell harbor seals apart. Before SealNet, she said the only way to identify individual seals was by tagging them, but that wasn't going great. They're pretty slippery, and the best way to do it is to try and shoot them with tracking darts.
INGRAM: It's very time-consuming, costly and, to be honest, it stresses them out.
BRUMFIEL: SealNet can ID seals with high accuracy using just a photo - way easier and faster and just better for everyone. Both Ingram and Kleindorfer think that facial recognition is going to play a really important role in conservation and ecology. Researchers will be able to tell how many individuals are in a population. They'll see who's hanging with whom.
INGRAM: Yeah, I do think it's the wave of the future. And we need more computer scientists trained in behavioral ecology. And we need more conservation biologists trained in computer science. But working together, I think we can do this.
BRUMFIEL: And citizen scientists could be part of it, too. Sonia Kleindorfer hopes birdwatchers will someday be able to snap a picture of a goose, ID it, and share its location with scientists. But, she adds, just remember, her new research suggests that birdwatching goes both ways. Geese can remember faces, too.
KLEINDORFER: If you were ever not kind to a goose, then woe to you the rest of your life. It shall not be a happy one if that goose finds you again.
BRUMFIEL: So be nice to the geese, and maybe they'll be nice back.
Geoff Brumfiel, NPR News.
(SOUNDBITE OF ADELE SONG, "RUMOUR HAS IT") Transcript provided by NPR, Copyright NPR.
NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.