Image Understanding for Prevention of Vandalism in Metro Stations


N. Chleq and F. Brémond and M. Thonnat


We address here the issues of developing an interpretation system describing automatically human activities from image sequences. The class of applications weare interested in, is the automatic surveillance and monitoring of metro stations scenes observed by a monocular camera. We present related works in image sequence interpretation, then we analyze the problem of interpretation of image sequences both from a general point of view and in metro stations, and we proposea software architecture. Finally, we present the points that are really specific to the surveillance of metro stations, and some results obtained in the framework of the AVS-PV european Esprit project.

visual surveillance, image understanding, image sequence

BibTeX reference:

  Dernière mise à jour: 15/03/01