BiometricsGait biometrics shows promise
A new biometric technology may soon join retinal scans, voice recognition, and fingerprints as a means to identify individuals: gait pattern biometrics; a method of identifying individuals by the way they walk, saunter, swagger, or sashay has achieved accuracy of about 90 percent in early trials
Gait biometrics are showing effectiveness // Source: inate.org
A new study released yesterday (Wednesday) says that the way a bare foot strikes the ground as one walks reveals one’s identity almost as accurately as a fingerprint.
The new biometric technology – called foot pressure biometrics — may soon join other technologies — retinal scans, voice recognition, fingerprints as a means to identify individuals.
“Gait patterns” biometrics examines the way an individual walks, saunters, swaggers, or sashays — with up to 90-percent accuracy.
AFP reports that scientists led by Todd Pataky at Shinshu University in Tokida, Japan, built on this findings by measuring how the foot hits and leaves the ground during walking. They used 3-D image processing and a technique called image extraction to follow 104 volunteers(who walked barefoot) and analyze the heel strike, roll-to-forefoot, and push-off by the toes. Footstep patterns were matched to the individual with 99.6 percent accuracy, according to their paper, published on Wednesday in the Journal of the Royal Society Interface.
In an e-mail exchange with AFP, Pataky said the technology would be useful in security checks — but it would only work in situations where an individual wants to be recognized, “since anyone can modify their gait,” he explained. “Automated airport security checks, ATM security, controlled building access — in all these cases, an individual could walk normally to be positively identified.”
Pataky said that the next, well, step, is to conduct trial of the technology with the subject walking with shoes on.
— Read more in Todd C. Pataky et al., “Gait recognition: highly unique dynamic plantar pressure patterns among 104 individuals,” Journal of Royal Society Interface (7 September 2011) (doi: 10.1098/rsif.2011.0430)
