The Fraunhofer Institute's Barbara Krausz has developed a system for determining when crowds have become too large by observing the way people sway slowly from side to side to keep their balance when they become trapped in a highly congested area.
The software detects this motion by analyzing the movement of each pixel between different frames of crowd video footage, as highly symmetrical pixel motions indicate swaying. "It does not recognize that there is a person, we just check the pixel," Krausz says.
The system highlights areas in red where there is a sudden increase in symmetrical movements, enabling event organizers to quickly investigate crowd congestion. Krausz applied the system to footage of the overcrowding at the Love Parade music festival in Duisburg last year that killed 21 people and injured more than 500. The system highlighted areas where people squeezed together as emergency vehicles arrived, producing its highest alert about half an hour before the disaster.
Krausz wants to develop the system further to run crowd simulations that would advice security personnel on the best solutions, such as opening a gate at a given location.
From New Scientist
View Full Article - May Require Free Registration
Abstracts Copyright © 2011 Information Inc., Bethesda, Maryland, USA
No entries found