Removing character from text in non-image form where location of character in image of text falls outside of valid content boundary

• US 8,682,075 B2
• Filed: 12/28/2010
• Issued: 03/25/2014
• Est. Priority Date: 12/28/2010
• Status: Active Grant

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Reddy, Prakash
• Primary Examiner(s): Mehta, Bhavesh
• Assistant Examiner(s): Niu, Feng
• Application Number:
  US12/980,036
• Publication Number:
  US 20120163718A1
• Time in Patent Office: 1,183 Days
• Field of Search: None
• US Class Current: 382/176
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1444 Selective acquisition, locating or processing of specific regions, e.g. highlighted text, fiducial marks or predetermined fields
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables

Abstract

Data representing an image of text is received, as is data representing the text in non-image form. A valid content boundary within the image of the text is determined. For each character within the text in the non-image form, a location of the character within the image of the text is determined. Where the location of the character within the image of the text falls outside the valid content boundary, the character is removed from the data representing the text in the non-image form.

Claims

1. A method comprising:

receiving, by a processor, data representing an image of text, and data representing the text in non-image form;

determining, by the processor, a valid content boundary within the image of the text, the valid content boundary dividing a portion of the image corresponding to valid text of the image from and excluding another portion of the image corresponding to one or more of stray marks, dirt, debris, and handwritten notes;

for each character of a plurality of characters within the text in the non-image form,

determining, by the processor, a location of the character within the image of the text; and

where the location of the character within the image of the text falls outside the valid content boundary, removing the character from the data representing the text in the non-image form, by the processor.

2. The method of claim 1, further comprising, where the location of the character within the image of the text falls inside the valid content boundary, keeping the character within the data representing the text in the non-image form, by the processor.

3. The method of claim 1, further comprising generating the data representing the text in the non-image form by:

inputting the data representing the image of the text to an optical character recognition (OCR) engine;

receiving output from the OCR engine, the output being the data representing the text in the non-image form.

4. The method of claim 3, wherein determining the location of the character within the image of the text comprises receiving a specification of the location of the character within the image of the text from the OCR engine.

5. The method of claim 3, wherein the valid content boundary is determined independent of the OCR engine that generated the data representing the text in the non-image form.

6. The method of claim 1, wherein the image of the text is a given image of a plurality of images representing text and that are related to one another,

and wherein determining the valid content boundary within the image of the text comprises:

determining the valid content boundary based on one or more of the images that are related to one another.

7. The method of claim 6, wherein the images correspond to pages within a book.

8. The method of claim 6, wherein the image of the text corresponds to one of an odd-numbered page within the book and an even-numbered page within the book,

wherein where the image corresponds to an odd-numbered page within the book, determining the valid content boundary is based on the images that correspond to odd-numbered pages within the book,

and wherein where the image corresponds to an even-numbered page within the book, determining the valid content boundary is based on the images that correspond to even-numbered pages within the book.

9. The method of claim 1, wherein determining the valid content boundary within the image of the text comprises:

for each line of a plurality of lines of the image of the text,

determining whether the line includes content that is separated by white space from other content within the line by more than a threshold;

where the line includes content that is separated by white space from other content within the line by more than the threshold, setting a portion of the valid content boundary corresponding to the line so that the content that is separated by white space from the other content within the line is outside the valid content boundary.

10. The method of claim 1, wherein determining the valid content boundary within the image of the text comprises:

receiving a specification of the valid content boundary from a user.

11. The method of claim 1, wherein the character is not removed from the data representing the text in the non-image form where the location of the character within the image of the text falls outside the valid content boundary, the character has a font matching a font of a second character within the text that falls inside the valid content boundary, and the character is part of a word that is found within a dictionary.

12. A non-transitory computer-readable data storage medium storing a computer program executable by a processor, execution of the computer program by the processor causing a method to be performed, the method comprising:

receiving first data representing an image of text and a marking other than the text, and second data representing the text and the marking in non-image form, the marking represented within the second data as if the marking were part of the text;

determining a valid content boundary within the image, the valid content boundary dividing a portion of the image corresponding to the text from and excluding another portion of the image corresponding to the marking;

for each character of a plurality of characters within the second data,

determining a location of the character within the first data; and

where the location of an image portion within the first data corresponding to the character falls outside the valid content boundary, removing the character from the second data.

13. The non-transitory computer-readable data storage medium of claim 12, wherein the method further comprises generating the second data by:

inputting the first data to an optical character recognition (OCR) engine;

receiving output from the OCR engine, the output being the second data,

wherein the valid content boundary is determined independent of the OCR engine that generated the second data.

14. The non-transitory computer-readable data storage medium of claim 12, wherein the image is a given image of a plurality of images that are related to one another,

and wherein determining the valid content boundary within the image comprises:

determining the valid content boundary based on one or more of the images that are related to one another.

15. The non-transitory computer-readable data storage medium of claim 12, wherein determining the valid content boundary within the image of the text comprises:

for each line of a plurality of lines of the image of the text,

determining whether the line includes content that is separated by white space from other content within the line by more than a threshold;

where the line includes content that is separated by white space from other content within the line by more than the threshold, setting a portion of the valid content boundary corresponding to the line so that the content that is separated by white space from the other content within the line is outside the valid content boundary.

16. The non-transitory computer-readable data storage medium of claim 12, wherein the marking comprises one or more of stray marks, dirt, debris, and handwritten notes.

17. A computing system comprising:

a processor;

a computer-readable data storage medium to store first data representing an image of text and a marking other than the text, and second data representing the text and the marking in non-image form, the marking represented within the second data as if the marking were part of the text; and

logic executable by the processor to determine a valid content boundary within the image, the valid content boundary dividing a portion of the image corresponding to the text from and excluding another portion of the image corresponding to the marking, and for each character of a plurality of characters within the second data, to remove the character from the second data where the logic has determined that a location of an image portion within the first data corresponding to the character falls outside the valid content boundary.

18. The computing system of claim 17, further comprising an optical character recognition (OCR) engine to generate the second data,

wherein the logic is to determine the valid content boundary independent of the OCR engine generating the second data.

19. The computing system of claim 17, wherein the image is a given image of a plurality of images that are related to one another,

and wherein the logic is to determine the valid content boundary within the image based on one or more of the images that are related to one another.

20. The computing system of claim 17, wherein the images correspond to pages within a book, and the image of the text corresponds to one of an odd-numbered page within the book and an even-numbered page within the book,

wherein where the image corresponds to an odd-numbered page within the book, the logic is to determine the valid content boundary based on the images that correspond to odd-numbered pages within the book,

and wherein where the image corresponds to an even-numbered page within the book, the logic is to determine the valid content boundary based on the images that correspond to even-numbered pages within the book.

21. The computing system of claim 17, wherein the logic is to determine the valid content boundary within the image by:

for each line of a plurality of lines of the image of the text,

determining whether the line includes content that is separated by white space from other content within the line by more than a threshold;

where the line includes content that is separated by white space from other content within the line by more than the threshold, setting a portion of the valid content boundary corresponding to the line so that the content that is separated by white space from the other content within the line is outside the valid content boundary.

22. The computing system of claim 17, wherein the marking comprises one or more of stray marks, dirt, debris, and handwritten notes.