A multi-sensor framework for event detection in wide environments

G.L. Foresti

Department of Mathematics and Computer Science, University of Udine, Italy

In this talk we present a hierarchical framework that uses networks of static and active cameras. Static cameras are organized into subnets for the local monitoring of sub-areas of the whole environment. Each subnet is organized in such a way that data acquired are first locally fused. As data are collected, information about position, velocity and shapes of the moving objects are extracted and used to identify and classify occurred events. A communication protocol allows the sharing of information between different networks and especially between the hierarchy levels. This establishes a cooperation that, driven by an operative control unit, allows the identification of complex events occurred inside the environment and not only in restricted local areas. Finally, as result of the event analysis, active cameras can be tasked to track and acquire suspicious targets at higher resolution for surveillance purposes as identification, recognition and behavior analysis.

Paper presented at Measuring Behavior 2005 , 5^th International Conference on Methods and Techniques in Behavioral Research, 30 August - 2 September 2005, Wageningen, The Netherlands.

© 2005 Noldus Information Technology bv