Character string identification

• US 7,444,021 B2
• Filed: 10/15/2002
• Issued: 10/28/2008
• Est. Priority Date: 10/15/2001
• Status: Expired due to Fees

• Current Assignee: Silverbrook Research Pty Limited
• Original Assignee: Silverbrook Research Pty Limited
• Inventors: Napper, Jonathon Leigh
• Primary Examiner(s): Ahmed; Samir A.
• Assistant Examiner(s): RASHIDIAN, MOHAMMAD M
• Application Number:
  US10/492,161
• Publication Number:
  US 20050226512A1
• Time in Patent Office: 2,205 Days
• Field of Search: 382/186, 382/188, 382/220, 382/229, 382/309, 382/310, 382/311, 348/143, 345/159, 345/160, 345/161, 345/173
• US Class Current: 382/186
• CPC Class Codes: G06F 18/254 of classification results, e.g. of results related to same input data
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/2272 with lexical matching
• CPC Class Codes: G06V 30/268 Lexical context

Abstract

The present invention provides a method of identifying a string formed from a number of hand-written characters, such as hand-written words. In order to achieve this, the method operates to determine character probabilities for each character in the string, as well as to determine the probability of the string corresponding to a predetermined form of template. In this regard, each template represents a respective combination of character types. The template and character probabilities are then combined to determine string probabilities, with the character string being identified in accordance with the determined string probabilities.

Claims

1. A method of identifying a string formed from a number of hand-written characters, the method including steps of:

a) Determining character probabilities for each character in the string, each character probability representing the likelihood of the respective character being a respective one of a number of predetermined characters;

b) Determining template probabilities for the string, each template probability representing the likelihood of the string corresponding to a respective one of a number of templates, each template representing a respective combination of character types, wherein the step of determining the template probabilities for the string includes;

(b)(i) Determining the number of characters in the string;

(b)(ii) Selecting templates having an identical number of characters; and

(b)(iii) Obtaining a template probability for each selected template;

c) Determining string probabilities in accordance with the determined character and template probabilities; and

,

d) Identifying the character string in accordance with the determined string probabilities.

2. A method according to claim 1, each predetermined character having a respective character type.

3. A method according to claim 1, the character types including at least one of:

a) Digits;

b) Letters; and

,

c) Punctuation marks.

4. A method according to claim 1, the method of determining the character probabilities including using a character classifier.

5. A method according to claim 1, the template probability being predetermined by statistical analysis of a text corpus.

6. A method according to claim 1, the method including determining a potential character string corresponding to each template by:

a) Determining the character type of each character in the string from the template, and,

b) Selecting one of the predetermined characters for each character in the template, the predetermined character being selected in accordance with the determined character type and the character probability.

7. A method according to claim 6, the selected predetermined character being the predetermined character having the highest character probability.

8. A method according to claim 6, the method of identifying the character string including:

a) Determining a string probability for each potential string, the string probability being determined by concatenating the character probabilities of each selected character and the respective template probability; and

,

b) Determining the character string to be the potential string having the highest string probability.

9. A method according to claim 1, the method being performed using a processing system having:

a) A store for storing at least one of;

i) The predetermined characters;

ii) Template data representing at least one of;

(1) The templates; and

,

(2) The template probabilities; and

,

b) A processor, the processor being adapted to;

i) Receive the character string;

ii) Determine the character probabilities for each character in the string;

iii) Determine the template probabilities;

iv) Determine string probabilities in accordance with the determined character and template probabilities; and

,

v) Identify the character string in accordance with the determined string probabilities.

10. Apparatus for identifying a string formed from a number of hand-written characters, the apparatus including:

a) A store for storing at least one of;

i) A number of predetermined characters; and

,

ii) Template data representing a number of templates; and

,

b) A processor, the processor being adapted to;

i) Determine character probabilities for each character in the string, each character probability representing the likelihood of the respective character being a respective one of a number of predetermined characters;

ii) Determine template probabilities for the string, each template probability representing the likelihood of the string corresponding to a respective one of a number of templates, each template representing a respective combination of character types, wherein the processor is configured to determine the template probabilities for the string by;

Determining the number of characters in the string;

Selecting templates having an identical number of characters; and

Obtaining a template probability for each selected template;

iii) Determine string probabilities in accordance with the determined character and template probabilities; and

,

iv) Identify the character string in accordance with the determined string probabilities.

11. Apparatus according to claim 10, the processor being coupled to an input, the processor being further adapted to receive the string of hand-written characters via the input.