If you’re looking to identify someone, forget taking his fingerprints—try pointing a computer’s camera at his ear. A group of researchers have created an algorithm that can identify a person based on the shape of his ear with 99.6% accuracy, Wired reports. They think ear recognition could be the most accurate, least intrusive biometric method available for identifying someone.
Your ear shape is unique, researchers explain, and unlike fingerprints, it can’t rub off or callous over. “When you’re born, your ear is fully formed,” says the team’s leader. “The lobe descends a little, but overall, it stays the same. It’s a great way to identify people.” The technology—which uses rays to "draw" the curves of the ear, which are then translated into a unique set of numbers—is already “roughly equivalent if not better” than facial recognition, according to another man performing similar research. He admits it probably won't replace fingerprints, but "when you're doing surveillance ... you want anything you can get." Click here for the specifics of how the algorithm works.