By Matt Liebowitz
Like palm-activated ATMs and retina-scanning smartphones, tablet computers like the iPad may soon authenticate their rightful users by reading the unique movement of their hands, not their passwords.
Napa Sae-Bae, a doctoral student at the Polytechnic Institute of New York University, is working to build an app that, using multitouch sensors, will biometrically authenticate tablet users' hand gestures. Her goal, she told NextGov, is for the tablet device to recognize its owner's specific biological traits — their hand shape and finger length, for example — and use those unique characteristics, which do not change, to replace passwords, which run the risk of being cracked.
Although she says it will be at least a year before her app is ready, Sae-Bae has already developed an iPad app that asks users to make gestures on a touch screen, such as rotating an open palm and opening a closed fist, in order to verify their identity. The hand sensor technologies she's using are currently available and already used, in different capacities, on iPads and Android tablets.
She also built a biometric-analyzing algorithm that will be the technological basis of the app; in experiments with 34 people, she achieved a 90 percent accuracy rate in authenticating the hand movements made by each participant, NextGov said.
Sae-Bae's work falls in line with other advancements in biometric authentication that have made headlines in recent months, both for the consumer market and for the government. As part of its human measurements and signatures intelligence (human MASINT) program, the U.S. Air Force has expressed interest in developing security cameras that can detect a suspect's age and ethnicity, and even whether that person is a terrorist smuggling a bomb.
Last month, a Japanese bank announced that beginning in September, it would equip 10 ATMs with biometric sensors that read customers' palms for identity verification.
