DETECTION OF LOITERING INDIVIDUALS IN PUBLIC TRANSPORTATION AREAS

This paper describes a vision-based method for automatically detecting individuals loitering around inner-city bus stops is presented. Images from a single stationary color camera are classified using an appearance-based method involving short-term biometrics. A pedestrian segmentation module and a correlation module are also used in the extraction and classification processes. Results from experiments suggest that the system may be applicable in monitoring applications at bus stops where the prolonged presence of loitering individuals may be indicative of unsafe activities such as drug dealing.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01005293
  • Record Type: Publication
  • Source Agency: UC Berkeley Transportation Library
  • Files: BTRIS, TRIS
  • Created Date: Oct 14 2005 8:25AM