Confirming automatically recognized handwritten answers

• US 9,665,786 B2
• Filed: 02/20/2015
• Issued: 05/30/2017
• Est. Priority Date: 02/20/2015
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Inventors: Wshah, Safwan R., Skinner, Gary W., Gross, Eric M., Simmons, Jr., Isaiah L.
• Primary Examiner(s): Kholdebarin, Iman K
• Application Number:
  US14/627,457
• Publication Number:
  US 20160247035A1
• Time in Patent Office: 830 Days
• Field of Search: None
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/224 of printed characters having additional code marks or containing code marks
• CPC Class Codes: G06V 30/262 using context analysis, e.g. lexical, syntactic or semantic context
• CPC Class Codes: G06V 30/36 Matching; Classification

Abstract

An image is received that contains handwritten characters as answers to an inquiry. An automatic character recognition process is performed on the image to generate initially recognized characters from the handwritten characters. The initially recognized characters of an incorrect answer to a question of the inquiry are analyzed to automatically identify alternative recognized characters of the incorrect answer. Then, it can be determined whether one or more of the alternative recognized characters, when substituted in place of the initially recognized characters of the incorrect answer, change the incorrect answer to a correct answer to the question. If the alternative recognized characters substituted in place of the initially recognized characters in the correct answer exceed a minimum character recognition confidence value, the incorrect answer is changed to the correct answer to modify the initially scored answers to modified scored answers, and such are output.

Claims

1. A method comprising:

receiving an image into an image processor, said image containing handwritten characters as answers to an inquiry;

performing an automatic character recognition process on said image to generate initially recognized characters from said handwritten characters based on a first division of said handwritten characters by determining character recognition confidence values of said initially recognized characters based on said first division of said handwritten characters;

evaluating said initially recognized characters of an incorrect answer to a question of said inquiry to automatically identify alternative recognized characters of said incorrect answer based on a second division of said handwritten characters that is different from said first division of said handwritten characters by determining character recognition confidence values of said alternative recognized characters based on said second division of said handwritten characters that is different from said first division of said handwritten characters;

determining whether said one or more of said alternative recognized characters substituted in place of said initially recognized characters of said incorrect answer change said incorrect answer to a correct answer to said question;

determining whether said alternative recognized characters substituted in place of said initially recognized characters in said correct answer exceed a minimum character recognition confidence value;

changing said incorrect answer to said correct answer based on said alternative recognized characters in said correct answer exceeding said minimum character recognition confidence value, to produce modified scored answers; and

outputting said modified scored answers.

2. The method according to claim 1, said first division and said second division distinguishing individual handwritten characters of said handwritten characters within said image.

3. The method according to claim 1, said alternative recognized characters being different from corresponding ones of said initially recognized characters.

4. The method according to claim 1, further comprising obtaining said image from at least one of:

a scanner operatively connected to said image processor; and

a graphic user interface comprising a free-form input area receiving user handwriting marks, operatively connected to said image processor.

5. A method comprising:

receiving an image into an image processor, said image containing handwritten characters as answers to an inquiry;

automatically performing an automatic character recognition process on said image to generate initially recognized characters from said handwritten characters based on a first division of said handwritten characters by determining character recognition confidence values of said initially recognized characters based on said first division of said handwritten characters, using said image processor;

automatically determining whether said initially recognized characters represent correct answers or incorrect answers to said inquiry based on previously established correct and incorrect answers, to produce initially scored answers, using said image processor;

automatically evaluating said initially recognized characters of an incorrect answer, of said incorrect answers, to a question of said inquiry to automatically identify alternative recognized characters of said incorrect answer based on a second division of said handwritten characters that is different from said first division of said handwritten characters by determining character recognition confidence values of said alternative recognized characters based on said second division of said handwritten characters that is different from said first division of said handwritten characters, said alternative recognized characters of said incorrect answer being generated by said automatic character recognition process and having lower character recognition confidence values relative to character recognition confidence values of said initially recognized characters of said incorrect answer;

automatically determining whether said one or more of said alternative recognized characters substituted in place of said initially recognized characters of said incorrect answer change said incorrect answer to a correct answer to said question based on said previously established correct and incorrect answers, using said image processor;

automatically determining whether said lower character recognition confidence values of said alternative recognized characters substituted in place of said initially recognized characters in said correct answer exceed a minimum confidence value, using said image processor;

automatically changing said incorrect answer to said correct answer based on said lower character recognition confidence values of said alternative recognized characters in said correct answer exceeding said minimum confidence value, to modify said initially scored answers to modified scored answers, using said image processor; and

automatically outputting said modified scored answers from said image processor.

6. The method according to claim 5, said first division and said second division distinguishing individual handwritten characters of said handwritten characters within said image.

7. The method according to claim 5, said alternative recognized characters being different from corresponding ones of said initially recognized characters.

8. The method according to claim 5, further comprising obtaining said image from at least one of:

a scanner operatively connected to said image processor; and

a graphic user interface comprising a free-form input area receiving user handwriting marks, operatively connected to said image processor.

9. A system comprising:

a computerized device;

an image processor operatively connected to said computerized device;

a scanner operatively connected to said computerized device; and

a graphic user interface operatively connected to said computerized device,

said computerized device receiving an image into said image processor from at least one of said scanner and said graphic user interface,

said image containing handwritten characters as answers to an inquiry,

said computerized device automatically performing an automatic character recognition process on said image to generate initially recognized characters from said handwritten characters based on a first division of said handwritten characters by determining character recognition confidence values of said initially recognized characters based on said first division of said handwritten characters,

said computerized device automatically determining whether said initially recognized characters represent correct answers or incorrect answers to said inquiry based on previously established correct and incorrect answers, to produce initially scored answers,

said computerized device automatically evaluating said initially recognized characters of an incorrect answer, of said incorrect answers, to a question of said inquiry to automatically identify alternative recognized characters of said incorrect answer based on a second division of said handwritten characters that is different from said first division of said handwritten characters by determining character recognition confidence values of said alternative recognized characters based on said second division of said handwritten characters that is different from said first division of said handwritten characters,

said alternative recognized characters of said incorrect answer being generated by said automatic character recognition process and having lower character recognition confidence values relative to character recognition confidence values of said initially recognized characters of said incorrect answer,

said computerized device automatically determining whether said one or more of said alternative recognized characters substituted in place of said initially recognized characters of said incorrect answer change said incorrect answer to a correct answer to said question based on said previously established correct and incorrect answers,

said computerized device automatically determining whether said lower character recognition confidence values of said alternative recognized characters substituted in place of said initially recognized characters in said correct answer exceed a minimum confidence value,

said computerized device automatically changing said incorrect answer to said correct answer based on said lower character recognition confidence values of said alternative recognized characters in said correct answer exceeding said minimum confidence value, to modify said initially scored answers to modified scored answers, and

said computerized device automatically outputting said modified scored answers.

10. The system according to claim 9, said first division and said second division distinguishing individual handwritten characters of said handwritten characters within said image.

11. The system according to claim 9, said alternative recognized characters being different from corresponding ones of said initially recognized characters.

12. The system according to claim 9, said graphic user interface comprising a free-form input area receiving user handwriting marks.

13. A computerized device comprising:

a first connection to an image processor;

a second connection to a scanner; and

a third connection to a graphic user interface,

said computerized device receiving an image into said image processor from at least one of said scanner and said graphic user interface,

said image containing handwritten characters as answers to an inquiry,

said computerized device automatically performing an automatic character recognition process on said image to generate initially recognized characters from said handwritten characters based on a first division of said handwritten characters by determining character recognition confidence values of said initially recognized characters based on said first division of said handwritten characters,

said computerized device automatically determining whether said initially recognized characters represent correct answers or incorrect answers to said inquiry based on previously established correct and incorrect answers, to produce initially scored answers,

said computerized device automatically evaluating said initially recognized characters of an incorrect answer, of said incorrect answers, to a question of said inquiry to automatically identify alternative recognized characters of said incorrect answer based on a second division of said handwritten characters that is different from said first division of said handwritten characters by determining character recognition confidence values of said alternative recognized characters based on said second division of said handwritten characters that is different from said first division of said handwritten characters,

said alternative recognized characters of said incorrect answer being generated by said automatic character recognition process and having lower character recognition confidence values relative to character recognition confidence values of said initially recognized characters of said incorrect answer,

said computerized device automatically determining whether said one or more of said alternative recognized characters substituted in place of said initially recognized characters of said incorrect answer change said incorrect answer to a correct answer to said question based on said previously established correct and incorrect answers,

said computerized device automatically determining whether said lower character recognition confidence values of said alternative recognized characters substituted in place of said initially recognized characters in said correct answer exceed a minimum confidence value,

said computerized device automatically changing said incorrect answer to said correct answer based on said lower character recognition confidence values of said alternative recognized characters in said correct answer exceeding said minimum confidence value, to modify said initially scored answers to modified scored answers, and

said computerized device automatically outputting said modified scored answers.

14. The computerized device according to claim 13, said first division and said second division distinguishing individual handwritten characters of said handwritten characters within said image.

15. The computerized device according to claim 13, said alternative recognized characters being different from corresponding ones of said initially recognized characters.

16. The computerized device according to claim 13, said graphic user interface comprising a free-form input area receiving user handwriting marks.