Reducing processing latency in optical character recognition for portable reading machine

• US 8,531,494 B2
• Filed: 12/08/2009
• Issued: 09/10/2013
• Est. Priority Date: 04/02/2004
• Status: Active Grant

• Current Assignee: National Federation of the Blind
• Original Assignee: K-NFB Reading Technology Inc.
• Inventors: Kurzweil, Raymond C., Albrecht, Paul, Gibson, Lucy
• Primary Examiner(s): Dharia, Prabodh M
• Application Number:
  US12/632,949
• Publication Number:
  US 20100088099A1
• Time in Patent Office: 1,372 Days
• Field of Search: 345/8, 345/156, 345/184, 345/418, 345/419, 345/473, 345/581, 345/690, 345/696, 345/698, 704/251, 704258-260, 704267-276, 382/100, 382/114, 382/153, 382/155, 382/164, 382/166, 382/173, 382/176, 382/181, 348/62, 348/157, 348/207.1, 348/222.1, 348/229.1, 235/454, 235/379
• US Class Current: 345/696
• CPC Class Codes: G06V 10/96 Management of image or video recognition tasks
• CPC Class Codes: G09B 21/001 Teaching or communicating with blind persons G09B21/02 - G09B21/06 take precedence
• CPC Class Codes: G09B 21/006 using audible presentation of the information

Abstract

A portable reading device includes a computing device and a computer readable medium storing a computer program product to receive an image and select a section of the image to process. The product processes the section of the image with a first process and when the first process is finished processing the section of the image, process a result of the first process with a second process. While the second process is processing, repeats the first process on another section of the image.

Claims

1. A computer program product tangibly embodied on a computer readable storage device, the computer program product comprising instructions for causing a portable reading device to:

receive an image;

select a first section of the image to process;

apply optical character recognition to the first section of the image and when finished processing at least a portion of the first section of the image,

process a result of the optical character recognition with a different process, and while the different process is processing the result of optical character recognition,

select a subsequent section of the image to process by optical character recognition; and

apply optical character recognition on the subsequent section of the image after the first section of the image is processed by the optical character recognition.

2. The computing program product of claim 1 further comprising instructions to:

parse the image into sections to detect regions of monochromatic and mixed content.

3. The computing program product of claim 2 wherein the different process is text to speech synthesis that uses as input a text output from the optical character recognition.

4. The computing program product of claim 1 wherein instructions to select are based on a user input corresponding to the user having the product in a directed reading mode.

5. The computing program product of claim 1 wherein instructions to select are based on selecting a first available portion of the image.

6. The computing program product of claim 1 wherein instructions to select are based on user-initiated input.

7. The computing program product of claim 1 wherein instructions to select are based on a default.

8. The computing program product of claim 1 further comprise instructions to:

determine if there are more sections in the image to process; and

select another section of the image to apply optical character recognition to until there are no more sections to apply optical character recognition to.

9. The computing program product of claim 1 further comprising instructions to:

search for characteristic text or marks in standardized areas of the document to determine regions.

10. The computing program product of claim 1 further comprising instructions to:

preprocess the image prior to optical character recognition to produce a binary or grayscale image for the optical character recognition processing.

11. The computing program product of claim 1 further comprising instructions to:

process the image using a template that is mapped to the image; and

apply voice recognition to a user'"'"'s verbal commands to select a part of the template that maps to the section of the image, causing the instructions to process that section of the image.

12. The computing program product of claim 1 further comprising instructions to:

determine whether the image is upside down, by determining that a relatively low number of words were recognized by the first process; and

change the orientation of the section of the image by 180 degrees and apply optical character recognition to the section of the image in the changed orientation; and

if changing the orientation produces enough words to surpass the threshold, then the instructions process all remaining sections of the image by changing the orientation of the remaining sections of the image.

13. A computer implemented method of operating a machine that processes an image, the method comprising:

receiving an image;

selecting by a computing device a first section of the image to process;

applying optical character recognition by the computing device to the first section of the image; and

when finished processing the first section of the image,

processing by the computing device a result of the optical character recognition with a different process, and while the different process is processing the result of optical character recognition,

selecting by the computing device a subsequent section of the image to process by optical character recognition; and

applying by the computer optical character recognition on the subsequent section of the image after the first section of the image is processed by the optical character recognition.

14. The method of claim 13 further comprising:

parsing the image into sections to detect regions of monochromatic and mixed content.

15. The method of claim 14 wherein the different process is text to speech synthesis that uses as input a text output from the optical character recognition.

16. The method of claim 13 wherein selecting is based on a user input corresponding to the user having the product in a directed reading mode.

17. The method of claim 13 wherein selecting is based on selecting a first available portion of the image.

18. The method of claim 13 wherein selecting is based on user-initiated input.

19. The method of claim 13 wherein selecting is based on a default.

20. The method of claim 13 further comprising:

determining if there are more sections in the image to process; and

selecting another section of the image to apply optical character recognition to until there are no more sections to apply optical character recognition to.

21. The method of claim 13 further comprising:

searching for characteristic text or marks in standardized areas of the document to determine regions.

22. The method of claim 13 further comprising instructions to:

preprocess the image prior to optical character recognition to produce a binary or grayscale image for the optical character recognition processing.

23. The method of claim 13 further comprising:

processing the section image using a template that is mapped to the image; and

applying voice recognition to a user'"'"'s verbal commands to select a part of the template that maps to the section of the image, causing the instructions to process that section of the image.

24. The method of claim 13 further comprising:

determining whether the image is upside down, by determining that a relatively low number of words were recognized by the first process;

changing the orientation of the section of the image by 180 degrees while applying optical character recognition to the section of the image; and

determining if changing the orientation produced enough words to surpass the threshold, then processing all remaining sections of the image by changing the orientation.

25. A portable reading device comprising:

a processor device;

a computer readable medium storing a computer program product comprising instructions for causing the reading machine to;

receive an image;

select a first section of the image to process;

apply optical character recognition to the first section of the image and when finished processing the first section of the image,

process a result of the optical character recognition with a different process; and

while the different process is processing the result of optical character recognition,

select a subsequent section of the image to process by optical character recognition; and

apply optical character recognition on the subsequent section of the image after the first section of the image is processed by the optical character recognition.

26. The portable reading device of claim 25 further comprising instructions to:

parse the image into sections to detect regions of monochromatic and mixed content.

27. The portable reading device of claim 25 wherein the different process is text to speech synthesis that uses as input a text output from the optical character recognition.

28. The portable reading device of claim 25 wherein instructions to select are based on a user input corresponding to the user having the portable reading device in a directed reading mode.

29. The portable reading device of claim 25 wherein instructions to select are based on selecting the first available portion of the image.

30. The portable reading device of claim 25 wherein instructions to select are based on user-initiated input.

31. The portable reading device of claim 25 wherein instructions to select are based on a default.

32. The portable reading device of claim 25 wherein instructions to process the section with a first process, further comprise instructions to:

determine if there are more sections in the image to process and if so select a next section of the image to apply optical character recognition to, until there are no more sections to apply optical character recognition to.

33. The portable reading device of claim 25 further comprising instructions to:

search for characteristic text or marks in standardized areas of the document to determine regions.

34. The portable reading device of claim 25 further comprising instructions to:

preprocess the image prior to optical character recognition to produce a binary or grayscale image for the optical character recognition processing.

35. The portable reading device of claim 25 further comprising instructions to:

process the image using a template that is mapped to the image; and

apply voice recognition to a user'"'"'s verbal commands to select a part of the template that maps to the section of the image, causing the instructions to process that section of the image.

36. The portable reading device of claim 25 further comprising instructions to:

determine whether the image is upside down, by determining that a relatively low number of words were recognized by the first process; and

change the orientation of the section of the image by 180 degrees and apply optical character recognition to the section of the image in the changed orientation; and

determine if changing the orientation produces enough words to surpass the threshold, then the instructions process all remaining sections of the image by changing the orientation.