• DocumentCode
    2093977
  • Title

    Using image processing methods to reduce dazzle in the eyes from a digital projector

  • Author

    Bai, Ying-Wen ; Liu, Yu-Cheng ; Tsai, Cheng-Hung

  • Author_Institution
    Dept. of Electr. Eng., Fu Jen Catholic Univ., Taipei, Taiwan
  • fYear
    2011
  • fDate
    10-12 May 2011
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    When someone is standing in front of a presentation screen, the contrast between the projector light and the indoor illumination is, as a rule, too great. This causes frequent contraction of the pupils of a person´s eyes and thereby great discomfort. In this paper we use the image processing method to reduce the light intensity towards a person´s face to the approximate level of the indoor illumination, thus reducing the pupil contraction frequency. Our design proposes three methods by using both the current computer vision technology-motion detection and face detection. Motion detection tracks the person´s location and infers the head position. Face detection, by using a combination of the two methods, tracks the location and speeds up the face examination process. These three methods all use a black oval mask to cover the face so that the eyes can avoid any dazzling light.
  • Keywords
    computer displays; computer vision; eye; optical projectors; screens (display); computer vision technology motion detection; dazzle reduction; digital projector; eye discomfort; face detection; head position; image processing method; indoor illumination; light intensity reducion; presentation screen; pupil contraction frequency; Face; Face detection; Gray-scale; Lighting; Motion detection; Software; Computer Vision; Face Detection; Motion Detection; Projector;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Instrumentation and Measurement Technology Conference (I2MTC), 2011 IEEE
  • Conference_Location
    Binjiang
  • ISSN
    1091-5281
  • Print_ISBN
    978-1-4244-7933-7
  • Type

    conf

  • DOI
    10.1109/IMTC.2011.5944023
  • Filename
    5944023