Method and apparatus for a pattern recognition

• US 4,977,603 A
• Filed: 01/09/1989
• Issued: 12/11/1990
• Est. Priority Date: 01/18/1988
• Status: Expired due to Term

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Irie, Bunpei, Asada, Haruo
• Primary Examiner(s): Razavi, Michael
• Application Number:
  US07/294,649
• Time in Patent Office: 701 Days
• Field of Search: 382/34, 382/18, 382/22, 382/27, 382/21, 382/26, 382/28, 358/462
• US Class Current: 382/218
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/18095 Summing image-intensity values; Projection and histogram analysis

Abstract

A method and an apparatus for pattern recognition utilizing the multiple similarity method, capable of taking structural features of a pattern to be recognized into account, so that sufficiently accurate pattern recognition can be achieved even when the pattern may involve complicated and diverse variations. The method includes the steps of: counting a number of occurences within each one of localized regions which subdivides a pattern to be recognized, of local patterns indicating possible arrangements of picture elements; deriving a vector quantity indicating distribution of black picture elements which constitute the pattern, from the numbers of occurrences of the local patterns; calculating multiple similarity defined in terms of square of inner product of the vector quantity and one of prescribed standard vectors representing standard patterns; and recognizing the pattern by identifying the pattern with one of the standard pattern whose corresponding standard vectors gives the maximum values for the multiple similarity.

Claims

1. An apparatus for pattern recognition of a pattern constituted by an arrangement of black picture elements on a background constituted by white picture elements comprising:

means for counting a number of occurrences within each one of localized regions of local patterns indicating possible arrangements of the white and black picture elements, the localized regions subdividing the pattern;

means for deriving a vector quantity indicating distribution of black picture elements on a basis of the numbers of occurrences of the local patterns for each one of the localized regions counted by the counting means;

means for calculating multiple similarity defined in terms of the square of the inner product of the vector quantity derived by the deriving means and each one of prescribed standard vectors representing one of prescribed standard patterns;

means for recognizing the pattern by identifying the pattern with one of the standard patterns for which the multiple similarity calculated by the calculating means is maximum.

2. The apparatus of claim 1, wherein the local patterns are in forms of 2×

2 matrices.

3. The apparatus of claim 1, wherein the vector quantity contains information concerning densities of black picture elements evaluated across the pattern along more than one direction.

4. A method of pattern recognition of a pattern constituted by an arrangement of black picture elements on a background constituted by white picture elements, comprising the steps of:

(a) counting the number of occurrences within each one of localized regions of local patterns indicating possible arrangements of the white and black picture elements the localized regions subdividing the pattern;

(b) deriving a vector quantity indicating distribution of the black picture elements on a basis of the number of occurrences of the local patterns for each one of the localized regions counted at the step (a);

(c) calculating multiple similarity defined in terms of the square the inner product of the vector quantity derived at step (b) and each one of prescribed standard vectors representing one of prescribed standard patterns; and

(d) recognizing the pattern by identifying the pattern with one of the standard patterns for which the multiple similarity calculated by the calculating means is maximum.

5. The method of claim 4, wherein the local patterns are in forms of 2×

2 matrices.

6. The method of claim 4, wherein the vector quantity contains information concerning densities of the black picture elements evaluated across the pattern along more than one direction.