Optical word recognition by examination of word shape

• US 5,557,689 A
• Filed: 06/01/1995
• Issued: 09/17/1996
• Est. Priority Date: 11/19/1991
• Status: Expired due to Term

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Hopcroft, Michael J., Huttenlocher, Daniel P.
• Primary Examiner(s): COUSO, YON JUNG
• Application Number:
  US08/456,413
• Time in Patent Office: 474 Days
• Field of Search: 382/181, 382/173, 382/176, 382/174, 382/177, 382/229, 382/190, 382/289, 382/290, 382/203
• US Class Current: 382/177
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1478 of characters or characters lines
• CPC Class Codes: G06V 30/182 by coding the contour of the pattern
• CPC Class Codes: G06V 30/2253 Recognition of characters printed with magnetic ink G06V30/2247 takes precedence
• CPC Class Codes: G06V 30/2264 using word shape

Abstract

A method for reducing an image of a character or word string to one or more one dimensional signals, including the steps of determining page orientation, isolating character strings from adjacent character strings, establishing a set of references with respect to which measurements about the character string may be made, and deriving a plurality of measurements with respect to the references in terms of one dimensional signal, amenable to identification processing techniques.

Claims

1. A method of recognizing a string of symbols in data defining an image, comprising the steps of:

a) detecting in a set of image signals representing an image, a discrete symbol string in an image wherein symbols strings are made up of a plurality of discrete symbols, and image data representing a symbol string in the image data distinguishes symbol-to-symbol spacing from symbol string-to-symbol string spacing;

b) representing the image data of one such symbol string as at least one, one-dimensional signal, said step of representing image data of one such symbol string including the substeps of;

identifying a reference with respect to which a shape signal will be derived for the symbol string;

deriving a set of measurements from the reference with respect to the image data forming the symbol string representing the set of measurements as a curve; and

for a selected portion of the curve, representing that portion as a one dimensional signal f(t) derived with respect to the reference; and

c) comparing said representative signal of the symbol string with a representative signal of a second symbol string.

3. The method as defined in claim 1, where the symbol string is representative of a word.

4. The method as defined in claim 1, where the image data represents a bitmapped image.

5. The method as defined in claim 1, wherein the step of detecting image data encoding a symbol string in the data defining the image includes the substep of:

isolating the image data encoding the symbol string from adjacent portions of the image.

6. The method as defined in claim 1, wherein the set of measurements is a set of distance measurements, measured from the reference to the image data forming the symbol string.

7. The method as defined in claim 6, wherein the step representing the image data of one such symbol string includes the substeps of:

defining a second reference with respect to the image data of a second symbol string; and

deriving a second set of measurements from the second reference with respect to the image data forming the second symbol string;

representing the second set of measurements as a second curve; and

for a selected portion of the curve, representing that portion as a second one dimensional signal f(t) derived with respect to the reference;

combining said second one dimensional signal f(t) with f(t) for a subsequent comparison step.

8. The method as defined in claim 1, wherein the step of comparing said representative signal of a symbol string with a representation of another symbol string includes the substep of:

receiving from an external source a representative signal of another symbol string.

2. A method of recognizing a string of symbols forming a word object within data defining an image, comprising the steps of:

a) identifying a discrete symbol string in the data defining the image wherein symbols strings are made up of a plurality of discrete symbols, and image data representing a symbol string in the image data distinguishes symbol-to-symbol spacing from symbol string-to-symbol string spacing;

b) deriving a signal representative of at least a portion of the shape of the symbol string, wherein the step of representing image data of one such symbol string includes the substeps of;

identifying a first reference with respect to which a shape signal will be derived for the symbol string;

deriving a set of measurements from the reference with respect to the image data forming the symbol string;

representing the set of measurements as a curve; and

for a selected portion of the curve, representing that portion as a one dimensional signal f(t) derived with respect to the reference;

identifying a second reference with respect to which a shape signal will be derived for the symbol string;

deriving a second set of measurements from the second reference with respect to the image data forming the second symbol string;

representing the second set of measurements as a second curve; and

for a selected portion of the second curve, representing that portion as a one dimensional signal fit) derived with respect to the second reference;

c) comparing said derived signal to at least one other known signal for determination of equivalence therewith; and

d) indicating recognition based on said comparison.

9. The method as defined in claim 2, where the symbol string is representative of a word.

10. The method as defined in claim 2, where the image data represents a bitmapped image.

11. The method as defined in claim 2, wherein the step of detecting image data encoding a symbol string in the data defining the image includes the substep of:

isolating the image data encoding the symbol string from adjacent portions of the image.

12. The method as defined in claim 2, wherein the set of measurements is a set of distance measurements, measured from the reference to the image data forming the symbol string.

13. The method as defined in claim 2, wherein the step of comparing said representative signal of a symbol string with a representation of another symbol string includes the substep of:

receiving from an external source a representative signal of another symbol string.