Browse Prior Art Database

Searching Method for Black Marks in Optical Character Recognition Forms

IP.com Disclosure Number: IPCOM000117808D
Original Publication Date: 1996-Jun-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 4 page(s) / 109K

Publishing Venue

IBM

Related People

Arima, T: AUTHOR [+2]

Abstract

Disclosed is a program that detects black marks in Optical Character Recognition (OCR) forms fast and accurately.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 56% of the total text.

Searching Method for Black Marks in Optical Character Recognition
Forms

Disclosed is a program that detects black marks in Optical Character
Recognition (OCR) forms fast and accurately.

      The purpose of black marks in OCR forms is to detect (1) the
form position in a scanned image, (2) form's skew ratio in the image,
and (3) image's shrink/expansion ratio.  They are used for character
segmentation.  The search of black marks generally spends much time
because it scans large image data.  Images by facsimile are generally
skewed much more than those by scanners and shifting more widely.
This disclosure is effective to detect black marks fast and
accurately.
  o  This search method scans only small image area by using narrow
      strips.  So it can find black marks fast.
  o  This method searches edges in the neighborhood of the corner of
      black marks.  So the detected corners are highly reliable even
if
      the form image is skewed.

      The following shows the process flow of this disclosure to
detect upper-right corner position of the top left black mark.
  1.  Mark position estimation
       Estimate the mark position by calculating it based on the
       information of a user definition table.
  2.  Rough search of mark's upper edge (Fig. 1)
       Search white-to-black transition point  downward in a vertical
       strip.  If no edge found, shift the zonary area horizontally
to
       righ...