September 8, 2011
Fingerprinting With Footsteps
Every person´s gait is so unique that their way of walking could be used as a way of identifying them, according to a new study published on Wednesday.
The study, led by Todd Pataky at Shinshu University in Japan, looked at the barefoot walking characteristics of 104 volunteers, such as the way the heel strikes the ground, the way the forefoot rolls and how the foot pushes off. The researchers found that footstep patterns identified each person with 99.6 percent accuracy, nearly as accurate as a person´s fingerprint.
The finding means that footsteps could one day be as useful in confirming identification as voice recognition, fingerprinting, retinal scans and mug shots.
Previous studies have already found that everyone does have a unique stride, but Pataky´s study is the first to measure how the foot hits and leaves the ground.
The team used a computer to determine gait patterns in individuals. The 3D imaging technique -- called image extraction -- analyzed the foot motions in the volunteers when they stepped and when they lifted.
The study, published in Britain´s Journal of the Royal Society Interface, was “proof of concept,” meaning it was carried out in experimental conditions among volunteers who were barefoot to see whether the theory was sound.
Pataky told AFP in an email that the technology would prove 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,” said Pataky.
The researchers noted that the study is preliminary and volunteers were only examined walking barefoot. It is not known whether these unique walking patterns occur when people have on shoes. Further studies are needed with shoed people to determine if their gait remains the same or not.
“We have some pilot data for walking with shoes, but have not yet conducted systematic testing,” Pataky concluded.
On the Net: