• DocumentCode
    943046
  • Title

    Image perception and enhancement for the visually impaired

  • Author

    Myers, Lemuel R., Jr. ; Rogers, Steven K. ; Kabrisky, Matthew ; Burns, Thomas J.

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Air Force Inst. of Technol., Wright-Patterson AFB, OH, USA
  • Volume
    14
  • Issue
    5
  • fYear
    1995
  • Firstpage
    594
  • Lastpage
    602
  • Abstract
    Wavelets have been used extensively as a method of extracting information contained in images and even in modeling human physiological data processing. Wavelet implementation for image processing parallels the analysis done by Peli concerning image enhancement and human perception. In this article, wavelets were used to provide a new image enhancement procedure by manipulating the decomposition detail coefficients. The true usefulness of the wavelet coefficients as enhancement tools can only be determined by testing. The only sure statement that can be made about image enhancement for the visually impaired, without first having a better understanding of the human visual system, is that individual subject preferences will be the deciding factor when choosing the optimal enhancement technique. Wavelet enhancements will provide different information to the low-vision observer and may prove to be superior to other enhancement techniques for some patients
  • Keywords
    image enhancement; sensory aids; spatial filters; vision defects; visual perception; wavelet transforms; decomposition detail coefficients; human physiological data processing; human visual system; image enhancement; image perception; image processing; information extraction; low-vision observer; optimal enhancement technique; patients; visually impaired; Data mining; Data processing; Humans; Image analysis; Image enhancement; Image processing; Testing; Visual system; Wavelet analysis; Wavelet coefficients;
  • fLanguage
    English
  • Journal_Title
    Engineering in Medicine and Biology Magazine, IEEE
  • Publisher
    ieee
  • ISSN
    0739-5175
  • Type

    jour

  • DOI
    10.1109/51.464776
  • Filename
    464776