• DocumentCode
    1925547
  • Title

    Comprehensive Check Image Reader

  • Author

    Shridhar, M. ; Houle, G.F. ; Kimura, F.

  • fYear
    2007
  • fDate
    5-7 March 2007
  • Firstpage
    407
  • Lastpage
    416
  • Abstract
    A check contains more than just an amount. The fields of interests are courtesy and legal amounts, date, check number, MICR line, payer, payee, memo, signature, signature instruction, fractional validation numbers, endorsement, security features, etc. In fact a bank can literally construct a very meaningful profile of the account holder from an analysis of the checks written and received by an account holder. Since October 2005 the "Check 21" law defines an image of a check as being as valid as the original document. The agreement between banks is that a binary image 200-240 dpi is the most practical format to use. Since most banks are now creating images and must be able to handle incoming images it creates a need for new image tools to validate the quality and readability of each field. This paper describes the challenges in finding and recognizing the fields of interest on the broad document types. These fields contain machine printed and handwritten words. The remaining of the paper would focus on handwritten legal line recognition which remains the most challenging field
  • Keywords
    cheque processing; document image processing; handwriting recognition; Check 21 law; account holder profile; comprehensive check image reader; document processing; handwritten legal line recognition; Background noise; Cameras; Data mining; Handwriting recognition; Image texture analysis; Law; Layout; Legal factors; Printing; Security;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computing: Theory and Applications, 2007. ICCTA '07. International Conference on
  • Conference_Location
    Kolkata
  • Print_ISBN
    0-7695-2770-1
  • Type

    conf

  • DOI
    10.1109/ICCTA.2007.42
  • Filename
    4127404