HEAD DETECTION INSIDE VEHICLES WITH A MODIFIED SVM FOR SAFER AIRBAGS

This paper proposes a vision system for detecting and localizing a passenger's head within the cockpit of a vehicle. A monocular camera is used to detect the passenger's head and to give its 2D position to the detection system. A had detection algorithm is based on a modified version of one of the most recent neural network methods known as Support Vector Machines (SVM). 3D information is then used to give the exact position of the passenger's head. The information regarding the presence of a person in the passenger seat and the position of their head is then used to control the deployment of the passenger airbag.

  • Supplemental Notes:
    • Publication Date: 2001. IEEE Service Center, Piscataway NJ
  • Corporate Authors:

    Siemens Aktiengesellschaft

    Machtfinger Strasse 10, Postfach 100079
    8000 Munich,   Germany 

    Texas A&M University, College Station

    Department of Mechanical Engineering
    College Station, TX  United States  77843-3123

    National Science Foundation

    1800 G Street, NW
    Washington, DC  United States  20550

    University of California, Berkeley

    California PATH Program, Institute of Transportation Studies
    Richmond Field Station, 1357 South 46th Street
    Richmond, CA  United States  94804-4648

    Universidad Nacional Autonoma de Mexico. Instituto de Ingenieria

    ,    

    University of California, Berkeley

    Department of Mechanical Engineering
    Berkeley, CA  United States  94720-1740

    DaimlerChrysler Forschungszentrum in Ulm

    ,    

    Technische Universitat Munchen. Fachgebiet Verkehrsplanung und Verkehrsplanung

    ,    

    University of California, Irvine

    Institute of Transportation Studies
    4000 Anteater Instruction and Research Building
    Irvine, CA  United States  92697

    Nanyang Technological University

    Centre for High Performance Embedded Systems
    Singapore,   Singapore 
  • Authors:
    • Reyna, Roberto
    • Giralt, Alain
    • Esteve, Daniel
  • Conference:
  • Publication Date: 2001

Language

  • English

Media Info

  • Pagination: p. 268-272

Subject/Index Terms

Filing Info

  • Accession Number: 00963574
  • Record Type: Publication
  • Source Agency: UC Berkeley Transportation Library
  • Files: PATH
  • Created Date: Oct 2 2003 12:00AM