Browse Prior Art Database

A Compact Error-Correction Method for OCR Mark Fields

IP.com Disclosure Number: IPCOM000123083D
Original Publication Date: 1998-May-01
Included in the Prior Art Database: 2005-Apr-04
Document File: 2 page(s) / 46K

Publishing Venue

IBM

Related People

Sumikawa, M: AUTHOR [+2]

Abstract

Disclosed is a method for compact error-correction in OCR mark fields. This method reduces area for visual check and enhances the user interface of mark fields correction in OCR by overlapping display of mark fields elements.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 81% of the total text.

A Compact Error-Correction Method for OCR Mark Fields

   Disclosed is a method for compact error-correction in OCR
mark fields.  This method reduces area for visual check and enhances
the user interface of mark fields correction in OCR by overlapping
display of mark fields elements.

   The elements for correction of mark fields are following
three.
  1.  Identification of item
  2.  Actual image
  3.  Recognition result

   Identification of item is explanation of mark field.  For
example "Male" or "Female".  Actual image is bit map image filled by
people.  Recognition result is code data obtained by OCR.

   Conventional methods of mark fields correction use three
lines for visual check.  The first line is used for identification
of item, the second line is used for actual image and the third line
is used for recognition result.  See Fig. 1.

   Our method reduces the area for the visual check by
overlapping the three elements for correction of mark fields.
Following are three cases that reduce the area.
  case 1.  Display an identification of item, actual image
            and recognition result to one line by overlapping
            them.  For example it shows identification of item,
            "Male", in blue, actual image, "x", in black and
            recognition result as pink or yellow background.  See
            Fig.2.  This is the most compact case.
  case 2.  Display an identificat...