Scientists Develop Facial Recognition Technology to Identify Birds


If you’ve ever struggled to identify a species of birds, facial recognition software is coming to the rescue.

According to the New York Post, a new tool created by an international team of scientists has been developed in hopes to boost conservation efforts and eliminate monitoring devices attached to animals.

The artificial intelligence technology is 92% accurate and similar to software being used to recognize facial features on social media.

“We show computers can consistently recognize dozens of individual birds – even though we cannot ourselves tell them apart,” said Dr. Andre Ferreira, lead author of the study which led to the tool's development. “In doing so, our study provides the means of overcoming one of the greatest limitations in the study of wild birds – reliably recognizing individuals.”

The study, which was published in Methods in Ecology and Evolution, collected thousands of labeled images of birds to help save bird populations from a number of causes including climate change, intensive farming, and deforestation.

“To our knowledge, this is the first successful attempt at performing such an individual recognition in small birds,” Ferreira said, noting it was previously only used on larger mammals.

The team used building feeders with camera traps and sensors to study bird populations in South Africa and Germany. The publication notes they worked similarly to microchips implanted in pet cats and dogs.

“The development of methods for automatic, non-invasive identification of animals completely unmarked and unmanipulated by researchers represents a major breakthrough in this research field,” said Ferreira. “Ultimately, there is plenty of room to find new applications for this system and answer questions that seemed unreachable in the past.”

The publication notes the current method used to identify animals is expensive and time-consuming but necessary to protect species. The advancement allows the elimination of current anxiety-inducing methods such as tagging birds with a colored bang.

Currently, the AI model works by being able to identify those it has been shown before.

“It is able to identify birds from new pictures as long as the birds in those pictures are previously known,” said Ferreira. “This means that if new birds join the study population the computer will not be able to identify them.”

More research will be required, according to Ferriera.

“Overall, our work demonstrates the feasibility of applying state-of-the-art deep learning tools for individual identification of birds, both in the laboratory and in the wild," the researcher said.

Facebook | Twitter | Instagram