• DocumentCode
    1469727
  • Title

    Recognition of transformed English letters with modified higher-order neural networks

  • Author

    He, Zhengquan ; Siyal, M.Y.

  • Author_Institution
    Sch. of Electr. & Electron. Eng., Nanyang Technol. Univ., Singapore
  • Volume
    34
  • Issue
    25
  • fYear
    1998
  • fDate
    12/10/1998 12:00:00 AM
  • Firstpage
    2415
  • Lastpage
    2416
  • Abstract
    Higher-order neural networks (HONNs) have been proven to be an effective neural model for invariant pattern recognition. However, one of the main problems in HONNs is that they are sensitive to distortion of the input patterns. The authors propose a modification scheme, and a modified HONN is employed to recognise transformed printed English letters
  • Keywords
    character recognition; neural nets; distortion; higher order neural network; invariant pattern recognition; modified HONN; transformed English letter recognition;
  • fLanguage
    English
  • Journal_Title
    Electronics Letters
  • Publisher
    iet
  • ISSN
    0013-5194
  • Type

    jour

  • DOI
    10.1049/el:19981654
  • Filename
    744020