• DocumentCode
    3529276
  • Title

    Pedestrian detection in near-infrared night vision system

  • Author

    Luo, Yun ; Remillard, Jeffrey ; Hoetzer, Dieter

  • Author_Institution
    Chassis Syst. Control, Robert Bosch LLC, Plymouth, MI, USA
  • fYear
    2010
  • fDate
    21-24 June 2010
  • Firstpage
    51
  • Lastpage
    58
  • Abstract
    Several premium automotive brands offer night vision systems to enhance the driver´s ability to see at night. Most recent generation night vision systems have added pedestrian detection as a feature to assist drivers to avoid potential collisions. This paper reviews pedestrian detection based on two different sensing technologies: active night vision operating in the near-infrared (NIR) region of the electromagnetic spectrum, and passive night vision operating in the far-infrared (FIR) spectrum. It also discusses the pros and cons of each type of night-vision system with-respect-to the pedestrian detection capability, the effectiveness for collision avoidance, and the commercial attractiveness. The paper introduces an enhancement to the NIR active lighting scheme that significantly improves the pedestrian detection performance. With improved pedestrian detection performance, we argue that the NIR night vision system is more effective at improving night-time driving safety and may achieve broader market acceptance.
  • Keywords
    accident prevention; collision avoidance; driver information systems; feature extraction; night vision; object detection; road safety; NIR active lighting scheme; active night vision; collision avoidance; driver assistance; electromagnetic spectrum; far-infrared spectrum; near-infrared night vision system; night time driving safety; passive night vision; pedestrian detection; sensing technology; Automotive engineering; Cameras; Collision avoidance; Control systems; Costs; Lighting; Night vision; Protection; Road safety; Vehicle driving;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Intelligent Vehicles Symposium (IV), 2010 IEEE
  • Conference_Location
    San Diego, CA
  • ISSN
    1931-0587
  • Print_ISBN
    978-1-4244-7866-8
  • Type

    conf

  • DOI
    10.1109/IVS.2010.5548089
  • Filename
    5548089