New facial recognition tech to help track, protect endangered primates

Researchers have made good use of facial recognition technology in a new way by using it to help track and protect endangered primates.

Researchers from Michigan State University (MSU) have created a new facial recognition software program, PrimNet, which can help in saving lives of endangered primates.

Tracking methods that are currently used requires capturing and tagging animals with tracking devices to keep tabs on them. In addition to being costly, devices ranging from $400 to $4,000, the method can also harm the animals causing them physical injuries, stress or even death, reported Futurism.

The MSU researchers claim that their created software can avoid all these drawbacks resulting from the usual tracking methods. Their study has been published via arXiv.

Chinese school uses facial recognition to observe inattentive students

In order to develop PrimNet, researchers first generated an image dataset for three species of primates, golden monkeys, chimpanzees and lemurs, taking numerous pictures of these animals in the wild. Then, they made use of this dataset to train a neural network for recognizing the individual animals.

Then, in order to make the software easy to use, a corresponding Android app was made that they called PrimID. Wildlife researchers or conservationists simply captures images of an animal and drop it in the app. The system then either produces a match, that is 90% accurate as put by MSU researchers, or it the match is not found, the system narrows down the field to five probable candidates, as per MSU.

“We compared PrimID to our own benchmark primate recognition system and two, open-source human face recognition systems, and the performance of PrimNet was superior in verification one-to-one comparison and identification, or one-to-many comparisons, scenarios,” said senior authors Anil Jain.

For the future, PrimNet could prove to be very beneficial as many endangered primates are on the edge of extinction. With this software, future losses can be prevented along with being non-invasive.

Jain said, “Intervention is necessary to halt and reverse these population declines. Automated facial recognition is one way we can help combat these loses.”

Moreover, this software can also slow illegal wildlife trafficking. As MSU explains, if a captured great ape is photographed and identified, knowing its origin can provide insights to its capture and help improve efforts to prevent future crimes.

“Moving forward, we plan to enlarge our primate datasets, develop a primate face detector, and share our efforts through open-source websites,” he added.