Researchers Develop Facial Recognition Software for Birds
ITHACA, N.Y., June 8 (UPI) — Turns out, Big Brother is a bird nerd. Thanks to the hard work of biologists and computer scientists at Cornell University, facial recognition software has turned the smartphone into a bird-identification expert.
The software comes in the form of a new app, called Merlin Bird ID, available online. It’s the result of collaboration among researchers from Cornell Tech, Caltech and the Cornell Lab of Ornithology.
Though now rather effective, the road to a workable electronic identification system was bumpy.
“Asking computers to identify bird species is a challenge not only because some species look so alike, but also because their shape varies from moment to moment,” scientists at the Cornell Lab explained in a recent press release. “On top of that, photographs of birds often include complex backgrounds, and the birds may be far away or blurry.”
To overcome these obstacles, researchers built an algorithm that improves as it acquires more experience and information. As more birders utilize the software, the identification system recognizing new patterns and becomes smarter.
“Computers can process images much more efficiently than humans — they can organize, index and match vast constellations of visual information such as the colors of the feathers and shapes of the bill,” explained Serge Belongie, a professor of computer science at Cornell Tech. “The state-of-the-art in computer vision is rapidly approaching that of human perception, and with a little help from the user, we can close the remaining gap and deliver a surprisingly accurate solution.”
It works by analyzing photos uploaded by birders. Users must trace the bird and click on its tail, beak and eye, but the computer does the rest — capable of identifying hundreds of species within in the United States and Canada. The algorithm hones in on key attributes (color, feather patterns, beak and tail shapes) to narrow down the possibilities and select the most likely match.
“It gets the bird right in the top three results about 90 percent of the time, and it’s designed to keep improving the more people use it,” said Jessie Barry, an ornithologist at Cornell who headed up the project. “That’s truly amazing, considering that the computer vision community started working on the challenge of bird identification only a few years ago.”