Skip to main content

Facial recognition software for lemurs could help save endangered species

Facial recognition software for lemurs could help save endangered species

Share this story

Newborn Ring Tailed Lemur At San Francisco Zoo
Photo by Justin Sullivan/Getty Images

A team of biologists and computer scientists have created the first ever facial recognition system for lemurs, able to identify more than 100 different individuals with 98.7 percent accuracy. It’s hoped that the software, dubbed LemurFaceID, will help with conservation efforts for the primate; giving researchers an easier and less invasive way to track individuals and whole families across generations.

“Like humans, lemurs have unique facial characteristics that can be recognized by this system," biometrics expert Anil Jain, who worked on the software, told “Once optimized, LemurFaceID can assist with long-term research of endangered species by providing a rapid, cost-effective and accurate method for identification."

The system is adapted from facial recognition software for humans. It was fed images of some 462 different lemurs covering a variety of different species, with the main focus being the red-bellied lemur — a species found in the rainforests of eastern Madagascar that’s vulnerable, but not endangered. A full description of the software is published in a report this week in the open access journal BMC Zoology.

A selection of lemur faces used to train the software.
A selection of lemur faces used to train the software.

Usually researchers rely on characteristics like injuries, scars, body size and shape to identify individual lemurs. However, these markers can be difficult to track over time as lemurs mature and grow. By comparison, using more stable facial markers for identification should make it easier to track lemurs over time. It’s also less invasive than “capture and collar” methods usually to identify animals for research.

“Studying lemur individuals and populations over long periods of time provides crucial data on how long individuals live in the wild, how frequently they reproduce, as well as rates of infant and juvenile mortality and ultimately population growth and decline," Stacey Tecot, a co-author of the report, told "Using LemurFaceID can inform conservation strategies for lemurs, a highly endangered group of mammals." 

LemurFaceID will have to be tested thoroughly in the wild to better evaluate its potential as a research tool, but its creators are confident it will be useful. In the future, they think it could also be adapted to work with other primates, like monkeys and gibbons. And at a time when scientists says more than 60 percent of all known primate species are at risk of extinction, conversationists need all the help they can get.