• DocumentCode
    2556931
  • Title

    A real-time eye detection system based on the active IR illumination

  • Author

    Guang-yuan, Zhang ; Bo, Cheng ; Zhe, Jin ; Jia-wen, Li

  • Author_Institution
    State Key Lab. of Automotive Safety & Energy, Tsinghua Univ., Beijing
  • fYear
    2008
  • fDate
    2-4 July 2008
  • Firstpage
    1255
  • Lastpage
    1260
  • Abstract
    Non-intrusive real-time eye detection is widely used in vision-based man-machine interaction. These methods based on IR illumination have little sensitivity to lighting condition change and complicated background environments. In this paper, we present the realization of eye detection system based on active IR illumination. The inner and outer rings of IR LEDs illuminated the eyes and generated two kinds of pupil images: bright and dark pupil images. The IR LEDs were controlled by the even/odd video signal information. It was easy to get the eye position by difference the image of bright and dark pupil images. Experimental results using real images demonstrated the accuracy and robustness of the proposed solution.
  • Keywords
    computer vision; object detection; user interfaces; active IR illumination; complicated background environments; lighting condition; nonintrusive real-time eye detection; pupil images; video signal information; vision-based man-machine interaction; Automotive engineering; Charge coupled devices; Computer vision; Eyes; Face detection; Infrared detectors; Light emitting diodes; Lighting; Pattern recognition; Real time systems; IR illumination; Image difference; Non-intrusive; Real-time; eye detection;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Control and Decision Conference, 2008. CCDC 2008. Chinese
  • Conference_Location
    Yantai, Shandong
  • Print_ISBN
    978-1-4244-1733-9
  • Electronic_ISBN
    978-1-4244-1734-6
  • Type

    conf

  • DOI
    10.1109/CCDC.2008.4597516
  • Filename
    4597516