Computer scientists with Kinalisoft in the Czech Republic have developed software designed to work with closed-circuit TV footage that can identify people based on the way they walk.
The software tracks a person by separating their shadow from the background, and then extracts additional information from the person's outline and the angle of their joints to find their unique gait.
Kinalisoft says the software can identify someone in as few as four steps, and effecting a limp is unlikely to throw it off. The software is correct 99 times out of 100, with just a 0.8-percent chance of confusing a person's normal walk for someone else, according to a report in Horizon magazine.
The researchers say the software could help police identify suspects who have their face covered or are out of frame. "Our system can help recognize such people and trace them as they move in front of cameras," says Kinalsoft computer scientist Radek Stourac.
From Daily Mail (United Kingdom)
View Full Article
Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA
No entries found