Method of and apparatus for optical character recognition based on geometric and color attribute hypothesis testing

• US 5,875,261 A
• Filed: 12/20/1991
• Issued: 02/23/1999
• Est. Priority Date: 12/20/1991
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Keller, Robert S., Fitzpatrick, Greg, Johnson, William J., Williams, Marvin L.
• Primary Examiner(s): Johns, Andrew W.
• Application Number:
  US07/812,249
• Time in Patent Office: 2,622 Days
• Field of Search: 382/17, 382/39, 382/162, 382/165, 382/227, 382/228, 101/211, 395/109, 395/110, 395/117, 395/131, 395/150, 283/114, 283/901, 235/469, 358/79, 358/539, 380/54, 345/431
• US Class Current: 382/165
• CPC Class Codes: G06F 18/254 of classification results, e.g. of results related to same input data
• CPC Class Codes: G06K 7/12 using a selected wavelength, e.g. to sense red marks and ignore blue marks
• CPC Class Codes: G06V 30/1918 Fusion techniques, i.e. combining data from various sources, e.g. sensor fusion
• CPC Class Codes: G06V 30/224 of printed characters having additional code marks or containing code marks

Abstract

Method of, and apparatus for, optical character recognition of predominate and non-predominate color coded characters based on hypothesis testing utilizing both geometric optical character recognition and color coded optical character recognition techniques. In this hypothesis testing technique, geometric OCR is used to generate a null hypothesis that a target character has been recognized, and an alternative hypothesis that the target character has not been recognized. Color coded OCR is used to select either the null hypothesis or the alternative hypothesis. Alternatively, the roles of geometric OCR and color coded OCR may be reversed such that color coded OCR is used for hypothesis generation, and geometric OCR is used for hypothesis selection.

Claims

1. A method of recognizing a color coded character, wherein said character comprises plural pixels which are of a first color, said first color pixels being predominate so as to allow humans to distinguish said character and a small amount of a second color associated with said character, said second color being non-predominate so as to be non-distracting to humans, but distinguishable by a color scanner, comprising the steps of:

processing said character in a first processing step by an optical character recognition means selected from an optical character recognition group consisting of geometric optical character recognition and non-predominate color coded optical character recognition; and

processing said character in a second precessing step by an optical character recognition means from said optical character recognition group not selected in said first processing step.

2. The method of claim 1, wherein said first processing step further comprises:

generating a null hypothesis that a character is recognized and an alternative hypothesis that said character is not recognized,

and wherein said second processing step further comprises;

selecting one hypothesis from a hypothesis group consisting of said null hypothesis and said alternative hypothesis.

3. The method of claim 2, wherein said step of selecting one hypothesis from a hypothesis group consisting of said null hypothesis and said alternative hypothesis, further comprises the steps of:

calculating, by said optical character recognition means from said optical character recognition group not selected in said first processing step, a probability of a hypothesis from said hypothesis group; and

selecting one hypothesis from said hypothesis group consisting of said null hypothesis and said alternative hypothesis based on said probability.

4. The method of claim 3, wherein said step of selecting a hypothesis from said hypothesis group consisting of said null hypothesis and said alternative hypothesis based on said probability, further comprises the steps of:

selecting said alternative hypothesis if said probability is greater than a threshold; and

selecting said null hypothesis if said probability is not greater than said threshold.

5. The method of claim 3, wherein said step of selecting a hypothesis from said hypothesis group consisting of said null hypothesis and said alternative hypothesis based on said probability, further comprises the steps of:

selecting said null hypothesis if said probability is greater than a threshold; and

selecting said alternative hypothesis if said probability is not greater than said threshold.

6. An apparatus for recognizing a color coded character, wherein said character comprises plural pixels which are of a first color, said first color pixels being predominate so as to allow humans to distinguish said character and a small amount of a second color associated with said character, said second color being non-predominate so as to be non-distracting to humans, but distinguishable by a color scanner, comprising:

a first means for processing said character by an optical character recognition means selected from an optical character recognition group consisting of geometric optical character recognition and non-predominate color coded optical character recognition; and

a second means for processing said character by an optical character recognition means from said optical character recognition group not selected in said first processing step.

7. The apparatus of claim 6, wherein said first means further comprises:

means for generating a null hypothesis that a character is recognized and an alternative hypothesis that said character is not recognized,

and wherein said second means further comprises;

means for selecting one hypothesis from a hypothesis group consisting of said null hypothesis and said alternative hypothesis.

8. The apparatus of claim 7, wherein said means for selecting one hypothesis from a hypothesis group consisting of said null hypothesis and said alternative hypothesis, further comprises:

means for calculating, by said optical character recognition means from said optical character recognition group not selected in said first means, a probability of a hypothesis from said hypothesis group; and

means for selecting one hypothesis from said hypothesis group consisting of said null hypothesis and said alternative hypothesis based on said probability.

9. The apparatus of claim 8, wherein said means for selecting a hypothesis from said hypothesis group consisting of said null hypothesis and said alternative hypothesis based on said probability, further comprises:

means for selecting said alternative hypothesis if said probability is greater than a threshold; and

means for selecting said null hypothesis if said probability is not greater than said threshold.

10. The apparatus of claim 8, wherein said means for selecting a hypothesis from said hypothesis group consisting of said null hypothesis and said alternative hypothesis based on said probability, further comprises:

means for selecting said null hypothesis if said probability is greater than a threshold; and

means for selecting said alternative hypothesis if said probability is not greater than said threshold.

11. A method of permitting the recognition of a color coded character, wherein said character comprises plural pixels which are of a first color, said first color pixels being predominate so as to allow humans to distinguish said character and a small amount of a second color associated with said character, said second color being non-predominate so as to be non-distracting to humans, but distinguishable by a color scanner, comprising the steps of:

providing a first means for processing said character by an optical character recognition means selected from an optical character recognition group consisting of geometric optical character recognition and non-predominate color coded optical character recognition; and

providing a second means for processing said character by an optical character recognition means from said optical character recognition group not selected in said first processing step.

12. The method of claim 11, wherein said step of providing a first means further comprises:

providing means for generating a null hypothesis that a character is recognized and an alternative hypothesis that said character is not recognized,

and wherein said step of providing a second means further comprises;

providing means for selecting one hypothesis from a hypothesis group consisting of said null hypothesis and said alternative hypothesis.

13. The method of claim 12, wherein said step of providing means for selecting one hypothesis from a hypothesis group consisting of said null hypothesis and said alternative hypothesis, further comprises the steps of:

providing means for calculating, by said optical character recognition means from said optical character recognition group not selected in said first processing step, a probability of a hypothesis from said hypothesis group; and

providing means for selecting one hypothesis from said hypothesis group consisting of said null hypothesis and said alternative hypothesis based on said probability.

14. The method of claim 13, wherein said step of providing means for selecting one hypothesis from said hypothesis group consisting of said null hypothesis and said alternative hypothesis based on said probability, further comprises the steps of:

providing means for selecting said alternative hypothesis if said probability is greater than a threshold; and

providing means for selecting said null hypothesis if said probability is not greater than said threshold.

15. The method of claim 13, wherein said step of providing means for selecting one hypothesis from said hypothesis group consisting of said null hypothesis and said alternative hypothesis based on said probability, further comprises the steps of:

providing means for selecting said null hypothesis if said probability is greater than a threshold; and

providing means for selecting said alternative hypothesis if said probability is not greater than said threshold.