13th November 2008 - Crowd Vision has today been named as the winner of the ideas competition. Swiss-based Crowd Vision is the inaugural winner of the Crowded Places Award, a new ideas competition conceived by Secure Futures and announced at the London Grand Final of the Global Security Challenge, an international business plan competition to find the most promising security technology start-ups in the world.

Crowd Vision has developed real-time surveillance software that can detect and track individuals within a crowd. The software has been refined using video material provided by the Saudi Government relating to the crowds visiting Mecca for the Hajj pilgrimage. In the last two years, fatalities at Mecca have been reduced, in part due to the use of Crowd Vision to pre-empt danger and intervene.

The Crowded Places Award is the first ideas competition dedicated to identifying and showcasing technologies that can help keep us all safer in the crowded places we frequent each day. From city streets to town centres and from train stations to nightclubs, the protection of crowded places is at the heart of securing our society and securing our citizens.

The Crowded Places ideas competition was conceived by Secure Futures, a new strategic consultancy focused exclusively on the emerging Public Security sector, where national security, law enforcement and public safety converge. Secure Futures aims to energise the Public Security sector by providing strategic insights and connectivity to innovators and investors, policymakers and security practitioners. Secure Futures can be found at: www.secure-futures.com

The award was handed to Anders Johansson, the founder of Crowd Vision, by Stephen Dennis, the Technical Director at HSARPA, the advanced research agency of the US Department of Homeland Security, and by Fiona Strens, founder of Secure Futures.

What is Crowd Vision about?

Crowd Vision is the brainchild of Anders Johansson, working under the guidance of renowned crowd dynamics researcher Dirk Helbing, at the Swiss Federal Institute of Technology in Zurich (ETH Zurich). Anders has been researching crowd flows for several years, first as the focus of his PhD and more recently as a research assistant at ETH Zurich.  He has developed software that detects and tracks people in very crowded conditions as they move past CCTV cameras.

The Saudi Government has been involved in providing Anders with considerable video material from the Hajj pilgrimages to Mecca, which has been used to test and refine the software under development. In the last two years, fatalities at Mecca have been reduced, in part due to the use of Crowd Vision to pre-empt danger and intervene.

Crowd dynamics is a highly complex subject. Crowds do not behave as fluids. Analysis of the Mecca data has highlighted unique crowd phenomenon, such as stop-and-go waves and crowd turbulence – sudden changes in crowd patterns that indicate imminent problems. The research has established a new set of metrics, such as ‘crowd pressure’, which can be used to measure and forecast crowd criticality.

Crowd Vision now aims to commercialise its software and optimise it for other real-world applications, taking feeds from dozens of CCTV cameras distributed over a large area, and then aggregating results to indicate overcrowding and to forecast where overcrowding is likely to occur.

For previous media coverage of Crowd Vision see:
Our article on nature.com
Our article on sciencemag.org
Our Article on newscientist.com