Security Cameras Can Now Identify People By The Way They Walk

Business Insider
September 27, 2012

That Guy Fawkes mask won't be the great identity equalizer much longer.

The U.K.'s National Physical Laboratory (NPL) has developed a recognition system that can identify people by the way they walk, Phys.Org reports. 

The technology can be used to track individuals through areas monitored by CCTV (i.e. security) cameras by analyzing their "gait signature," or specific way of walking.

The system combines computer models of the area with CCTV feeds to record a person's gait signature at certain points, check where else that person has been in the area and display the results on a central computer.

As of now it cannot pick someone out of a crowd with 100 percent certainty.

Perhaps that's why the NPL and its partners—the Centre for Advanced Software Technology, the BBC and BAE Systems—are "particularly focused" on standardizing the gait recognition measurement, which involves variables including equipment, timing and position.

YouTube/NPLdigital This is the "gait cycle," which is used to generate a gait signature.

The new technology raises the bar of biometrics (i.e. the identification and analysis of humans by their traits) and privacy concerns along with it because a database of collected gait signature information could potentially be merged with TrapWire's facial recognition database and the National Security Agency's domestic surveillance database of electronic communications.

And with the FBI's biometric database and domestic drones on the horizon, privacy advocates have plenty to be worried about.

"Ultimately this provides a security system that combines real-time video and automated biometrics recognition," the NPL states. "It presents results in an interactive and intuitive 3D model, which provides the security control room with a better spatial understanding of events and locations than could be achieved with a wall of … unrelated cameras."

Check out the video: