ELEDIA@UniTN

New ELEDIA@UniTN Website
The new ELEDIA@UniTN website is available at www.eledia.org/eledia-unitn   This website has been discontinued, please refer to the new website.

Detection, Density Estimation and Tracking of Crowds

 
 
Crowds monitoring can be of critical relevance, expecially when dealing with large scale locations, such as airports, supermarkets and stadiums, where crowds tend to appear for many different reasons. The use of automated techniques for monitoring crowds, such as estimating a crowd’s density, tracking its movements (and even observing its behaviour) is nowadays necessary in a large number of applications. In fact, areas where people are likely to cluster (e.g. an airport) need careful observation to ensure crowd safety (i.e. ensuring a fast and safe evacuation in case of danger, distributing people over all the available security doors). Crowd density estimation and detection can be of fundamental relevance also when dealing with marketing strategies and investigations, where the ability to observe people concentrations (e.g. inside a supermarket, in correspondence to the exposition of a brand new product) directly translates into an analysis of people main interests. Moreover, in some applications people monitoring through classical video surveillance systems (e.g. CCTV) may not be the proper choice, or in some cases it may even be not feasible (e.g for privacy reasons). In those scenarios, it would be of great interesest the use of the avaliable wireless infrastructure to infer informations about crowd density and movements, by simply analyzing the electromagnetic field distribution inside a wide area.
   
Large scale locations, airports, supermarkets, stadiums


Airport
Supermarket
 
Members of the ELEDIA Research Center have developed several techniques for the simulation of the electromagnetic field distribution in very-large scale crowded locations. The basic idea is to apply the concepts of Device-Free Localization (DFL) and Tracking when dealing with a big number of targets, which are spread over a wide area (indoor or outdoor). As a matter of fact, people inside a scenario under test interact with the electromagnetic signals transmitted by the infrastructured wireless devices (e.g. the Wi-Fi access points), thus modifying the received signal strength by each device. The crowd detection and tracking problem can then be reformulated in terms of a simplified electromagnetic inverse scattering problem, which is carried out by means of a learning-by-example (LBE) strategy.
   
E-Field distribution over a wide area
Probability map of crowd density inside a wide area

Keywords: Crowd Density Estimation, Device-Free Passive Localization, Crowd Detection, Crowd Tracking


See Also
  • F. Viani, L. Lizzi, P. Rocca, M. Benedetti, M. Donelli, and A. Massa, "Object tracking through RSSI measurements in wireless sensor networks," Electronics Letters, vol. 44, no. 10, pp. 653-654, May 8 2008
    doi: 10.1049/el:20080509
  • F. Viani, G. Oliveri, and A. Massa, "Real-time tracking of transceiver-free objects for homeland security," European Radar Conference (EuRAD 2009), pp.621-624, Sept. 30 2009-Oct. 2 2009
  • F. Viani, M. Donelli, M. Salucci, P. Rocca, and A. Massa, "Opportunistic exploitation of wireless infrastructures for homeland security," 2011 IEEE International Symposium on Antennas and Propagation (APSURSI), pp. 3062-3065, 3-8 July 2011
    doi: 10.1109/APS.2011.5997177
  • F. Viani, M. Salucci, P. Rocca, G. Oliveri, A. Massa, "A multi-sensor WSN backbone for museum monitoring and surveillance," 6th European Conference on Antennas and Propagation (EUCAP 2012), Prague, CZ, pp. 51-52, 26-30 Mar. 2012.
    doi:10.1109/EuCAP.2012.6206050
  • G. Menduni, F. Viani, F. Robol, E. Giarola, A. Polo, G. Oliveri, P. Rocca, and A. Massa, "A WSN-based architecture for the E-Museum - The experience at 'Sala dei 500' in Palazzo Vecchio (Florence)," IEEE Antennas and Propagation Society International Symposium (APSURSI 2013), Orlando, FL, United States, pp. 1114-1115, 7-13 Jul. 2013.
    doi:10.1109/APS.2013.6711217