Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Method for Supporting Stroke Variations in Online Handwritten Character Recognition

IP.com Disclosure Number: IPCOM000110085D
Original Publication Date: 1992-Oct-01
Included in the Prior Art Database: 2005-Mar-25
Document File: 2 page(s) / 52K

Publishing Venue

IBM

Related People

Kitamura, K: AUTHOR [+2]

Abstract

Disclosed is a method for supporting stroke variations in online handwritten character recognition by storing them in a dictionary.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 53% of the total text.

Method for Supporting Stroke Variations in Online Handwritten Character Recognition

       Disclosed is a method for supporting stroke variations in
online handwritten character recognition by storing them in a
dictionary.

      A significant problem in online handwritten character
recognition is to provide an efficient method of supporting
variations in the shape, number, and order of strokes.  If all of the
possible variations were stored in a dictionary, this would be
impractically large.

      Existing methods solve the problem by storing only major
variations in a dictionary.  At the time of recognition recognition,
an input character is matched with only the stored  major variations
in order to select (as the recognition result) the one that has the
minimum difference with the input character.  Therefore, existing
methods may not correctly recognize input characters that are minor
variations.

      The disclosed method solves the problem by compressing the
minor variations and merging them, thus enabling them to be stored in
a reasonably sized dictionary.  Although the merging usually results
in the loss of some information, it enables all of the possible
variations to be stored in the dictionary.  At the time of
implementation, an appropriate merging method can be selected, taking
account of the trade-off between the size of the dictionary and the
loss of information that is directly related to the recognition
accuracy.

      The following is an example of the simple merging method:

      Suppose that the followin...