Image evaluation for reading mode in a reading machine

• US 8,249,309 B2
• Filed: 04/01/2005
• Issued: 08/21/2012
• 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, Gashel, James, Gibson, Lucy, Lvovsky, Lev
• Primary Examiner(s): Ge, Yuzhen
• Application Number:
  US11/096,703
• Publication Number:
  US 20060008122A1
• Time in Patent Office: 2,699 Days
• Field of Search: None
• US Class Current: 382/114
• CPC Class Codes: G06V 10/993 Evaluation of the quality of the acquired pattern
• CPC Class Codes: G06V 30/142 using hand-held instruments; Constructional details of the instruments

Abstract

A portable reading machine detects poor image conditions for performing optical character recognition processing. The portable reading machine receives an image of sufficient resolution to distinguish lines of text but not necessarily of sufficient resolution to distinguish individual characters and processes the image to determine imaging conditions from the image. The reading machine reports imaging conditions to the user.

Claims

1. A computer program product embodied on a computer readable medium for causing a computing device of a portable reading machine for visually disabled persons to detect poor imaging conditions comprising instructions for causing the computing device to:

receive an image of sufficient resolution to distinguish lines of text but not necessarily of sufficient resolution to distinguish individual characters;

process the image to determine imaging light conditions under which the image was taken, by instructions to;

examine contrast characteristics of different parts of the image by examining a distribution of light intensities across the image, with regions of the image that have poor contrast characterized by a relatively narrow distribution of light intensity values compared to regions of good contrast;

determine whether poor contrast is present due to lighting that is too dim or too bright; and

determine whether glare conditions saturate the image and blur or obscure text that may be present in the image;

generate directions to convey to an operator of the portable reading machine based on processing the image; and

apply the generated directions to a text to speech synthesis module to provide an audio report of determined imaging lighting conditions as one of the imaging conditions and generated directions to the operator.

2. The computer program product of claim 1 wherein instructions to process the image to determine imaging conditions comprise instructions to cause the computing device to:

determine quality of at least one condition that is related to positioning of the reading machine that the computing device is operating in.

3. The computer program product of claim 1 wherein instructions to determine the quality of the lighting conditions, comprise instructions to:

compute the distribution of light intensities across the image.

4. The computer program product of claim 1 wherein the instructions to generate directions to the operator comprise instructions to cause the computing device to:

determine lighting conditions that the computing device is operating in and

report that an image input device may be out of focus if the determination of lighting quality has determined that there is low contrast in a region of the image.

5. The computer program product of claim 1 wherein the computer program product causes the device to determine if a plurality of characters are on the right side or the left side of the image, and if on the left side, to report that the text may be upside-down.

6. The computer program product of claim 1 wherein instructions to process the image to determine imaging conditions comprise instructions to:

determine that the text is too small; and

the instructions to report,

report text as too small as one of the imaging conditions.

7. The computer program product of claim 1 wherein instructions to process the image to determine imaging conditions comprise instructions to:

determine that the text is too large; and

the instructions to report,

report the text as too large as one of the imaging conditions.

8. The computer program product of claim 1 wherein instructions to process the image to determine imaging conditions comprise instructions to cause the computing device to:

detect that an image input device or an object whose image was captured by the image input device was tilted and that as a result text in the image is rotated or distorted; and

the instructions to report, report tilt as one of the imaging conditions.

9. The computer program product of claim 1 wherein instructions to process the image to determine imaging conditions comprise instructions to cause the computing device to:

detect that text is incomplete; and

the instructions to report, report that text is incomplete as one of the imaging conditions.

18. The computer program product of claim 1 wherein the computing program further comprises instructions to:

generate directions to convey to an operator of the machine a change in placement of the machine relative to a scene of which the image was taken, based on processing the image to determine imaging conditions; and

apply text to speech synthesis to the generated directions to provide an audio file that when rendered to the operator reports the generated directions to change placement.

10. A machine comprises:

a processing device;

a computer program product residing on a computer readable medium for causing the processing device to detect poor reading conditions for the reading machine, comprising instructions for causing the machine to;

receive an image of a scene;

process the image to determine imaging lighting conditions under which the image was taken, by instructions to;

examine contrast characteristics of different parts of the image by examining a distribution of light intensities across the image, with regions of the image that have poor contrast characterized by a relatively narrow distribution of light intensity values compared to regions of good contrast;

determine whether poor contrast is present due to lighting that is too dim or too bright; and

determine whether glare conditions saturate the image and blur or obscure text that may be present in the image; and

report the determined imaging lighting conditions.

11. The machine of claim 10 wherein the computing program product further comprises instructions to:

examine contrast characteristics of different parts of the image;

determine the brightness in different regions of the image; and

compute a distribution of light intensities across the image.

12. The machine of claim 10 wherein the computing program product further comprises instructions to:

detect and report incomplete or unreadable text.

13. The machine of claim 10 wherein the computing program product further comprises instructions to:

detect that an image input device or an object whose image was captured by the image input device was tilted and that as a result text in the captured image is rotated or distorted.

14. A method of operating a machine, the method comprises:

receiving an image of sufficient resolution to distinguish lines of text but not necessarily of sufficient resolution to distinguish individual characters;

processing the image by a computing device provided in the machine to determine imaging lighting conditions under which the image was taken, by;

examining by the computing device contrast characteristics of different parts of the image by determining a distribution of light intensities across the image, with regions of the image that have poor contrast characterized by a relatively narrow distribution of light intensity values compared to regions of good contrast;

determining by the computing device whether poor contrast is present due to lighting that is too dim or too bright; and

determining by the computing device whether glare conditions saturate the image and blur or obscure text that may be present in the image; and

producing a speech file by the computing device, to convey the determined imaging conditions audibly to a user.

15. The method of claim 14 further comprising:

examining contrast characteristics of different parts of the image;

determining the brightness in different regions of the image; and

computing a distribution of light intensities across the image.

16. The method of claim 14 further comprising:

detecting and reporting incomplete or unreadable text.

17. The method of claim 14 further comprising:

detecting that an image input device or an object whose image was captured by the image input device was tilted and that as a result text in the captured image is rotated or distorted.

19. The machine of claim 14 wherein the computing program further comprises instructions to:

generate directions to convey to an operator of the machine a change in placement of the machine relative to a scene of which the image was taken, based on processing the image to determine imaging conditions; and

apply text to speech synthesis to the generated directions to provide an audio file that when rendered to the operator reports the generated directions to change placement.

20. The method of claim 14 further comprising:

generating directions to convey to an operator of the machine changes in placement of the machine relative to a scene of which the image was taken, based on processing the image to determine light conditions; and

generating speech by applying by the computing device the generated directions to text to speech processing to produce an audio file that when rendered to the operator reports the generated directions to change placement.