IP Library › Granted Patent US 10,679,089
Granted Patent B2
US 10,679,089 · App. 15/858,355 · Granted Jun 9, 2020

Systems and methods for optical character recognition

Inventor: David Annis (Edmond, OK)
Assignee: Business Imaging Systems, Inc.
G06K9/344G06K9/00463G06K9/46G06K2209/01
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,679,089
App. No.
15/858,355
Granted
Jun 9, 2020
Kind
B2
Abstract

A method for iterative optical character recognition includes: performing an optical character recognition pass on an image; identifying a first character located in a first portion of the image, wherein an identification of the first character is associated with a first character confidence level that is at least a threshold character confidence level; identifying a second character located in a second portion of the image, wherein a first identification of the second character is associated with a second character confidence level that is below the threshold character confidence level; and performing an additional optical character recognition pass on the second portion of the image to identify the second character a second time when the second character confidence level that is below the threshold character confidence level.

Claims (92)

1. A method for iterative optical character recognition, comprising:

performing an optical character recognition pass on an image to identify a plurality of characters of the image;

identifying a first character located in a first portion of the image, wherein an identification of the first character is associated with a first character confidence level that is at least a threshold character confidence level;

identifying a second character located in a second portion of the image, wherein a first identification of the second character is associated with a second character confidence level that is below the threshold character confidence level;

performing an additional optical character recognition pass on the second portion of the image to identify the second character a second time, wherein a second identification of the second character is associated with a third character confidence level; and

selecting either the first identification of the second character or the second identification of the second character based upon a comparison of the second character confidence level and the third character confidence level.

2. The method of claim 1 , wherein performing the additional optical character recognition pass on the second portion of the image to identify the second character the second time includes:

storing a copy of the image in a memory; and

modifying the copy of the image to remove the first portion of the image from the copy of the image and leave the second portion of the image in the copy of the image; and

performing the additional optical character recognition pass on the copy of the image after modifying the copy of the image to remove the first portion of the image from the copy of the image and leave the second portion of the image in the copy of the image.

3. The method of claim 1 , further comprising:

defining at least a first section and a second section of an image;

performing a first optical character recognition pass on the first section of the image to identify a first plurality of characters of the image; and

performing a second optical character recognition pass on the second section of the image to identify a second plurality of characters of the image.

4. The method of claim 3 , wherein performing the first optical character recognition pass on the first section of the image to identify the first plurality of characters of the image includes:

storing a copy of the image in a memory; and

cropping the copy of the image to remove the second section of the image from the copy of the image and leave the first section of the image in the copy of the image; and

performing the first optical character recognition pass on the copy of the image after cropping the copy of the image to remove the second section of the image from the copy of the image and leave the first section of the image in the copy of the image.

5. The method of claim 3 , wherein performing the second optical character recognition pass on the second section of the image to identify the second plurality of characters of the image includes:

storing a copy of the image in a memory; and

cropping the copy of the image to remove the first section of the image from the copy of the image and leave the second section of the image in the copy of the image; and

performing the second optical character recognition pass on the copy of the image after cropping the copy of the image to remove the first section of the image from the copy of the image and leave the second section of the image in the copy of the image.

6. The method of claim 3 , wherein the first and second sections of the image are defined based on at least one of: a user defined page layout, a programmed page layout, or a detected page layout.

7. The method of claim 1 , further comprising:

detecting a plurality of segments defined by contiguous sequences of characters in the image;

performing a first optical character recognition pass on the image to identify characters in the plurality of segments of the image;

identifying a first set of one or more characters located in a first segment of the image, wherein an identification of the first set of one or more characters is associated with a first segment confidence level that is at least a threshold segment confidence level;

identifying a second set of one or more characters located in a second segment of the image, wherein a first identification of the second set of one or more characters is associated with a second segment confidence level that is below the threshold segment confidence level;

performing a second optical character recognition pass on the second segment of the image to identify the second set of one or more characters a second time, wherein a second identification of the second set of one or more characters is associated with a third segment confidence level; and

selecting the first identification of the second set of one or more characters or the second identification of the second set of one or more characters based upon a comparison of the second segment confidence level and the third segment confidence level.

8. The method of claim 7 , wherein performing the second optical character recognition pass on the second segment of the image to identify the second set of one or more characters the second time includes:

storing a copy of the image in a memory; and

cropping the copy of the image to remove portions of the image other than the second segment of the image in the copy of the image; and

performing the second optical character recognition pass on the copy of the image after cropping the copy of the image to remove portions of the image other than the second segment of the image in the copy of the image.

9. A system for iterative optical character recognition, comprising:

a controller including at least one processor configured to execute one or more modules stored by a memory that is communicatively coupled to the at least one processor, the one or more modules, when executed, causing the processor to:

perform an optical character recognition pass on an image to identify a plurality of characters of the image;

identify a first character located in a first portion of the image, wherein an identification of the first character is associated with a first character confidence level that is at least a threshold character confidence level;

identify a second character located in a second portion of the image, wherein a first identification of the second character is associated with a second character confidence level that is below the threshold character confidence level;

perform an additional optical character recognition pass on the second portion of the image to identify the second character a second time, wherein a second identification of the second character is associated with a third character confidence level; and

select the first identification of the second character or the second identification of the second character based upon a comparison of the second character confidence level and the third character confidence level.

10. The system of claim 9 , wherein the one or more modules, when executed, cause the processor to perform the additional optical character recognition pass on the second portion of the image to identify the second character the second time by:

storing a copy of the image in the memory; and

modifying the copy of the image to remove the first portion of the image from the copy of the image and leave the second portion of the image in the copy of the image; and

performing the additional optical character recognition pass on the copy of the image after modifying the copy of the image to remove the first portion of the image from the copy of the image and leave the second portion of the image in the copy of the image.

11. The system of claim 9 , wherein the one or more modules, when executed, further cause the processor to:

define at least a first section and a second section of an image;

perform a first optical character recognition pass on the first section of the image to identify a first plurality of characters of the image; and

perform a second optical character recognition pass on the second section of the image to identify a second plurality of characters of the image.

12. The system of claim 11 , wherein the one or more modules, when executed, cause the processor to perform the first optical character recognition pass on the first section of the image to identify the first plurality of characters of the image by:

storing a copy of the image in a memory; and

cropping the copy of the image to remove the second section of the image from the copy of the image and leave the first section of the image in the copy of the image; and

performing the first optical character recognition pass on the copy of the image after cropping the copy of the image to remove the second section of the image from the copy of the image and leave the first section of the image in the copy of the image.

13. The system of claim 11 , wherein the one or more modules, when executed, cause the processor to perform the second optical character recognition pass on the second section of the image to identify the second plurality of characters of the image by:

storing a copy of the image in a memory; and

cropping the copy of the image to remove the first section of the image from the copy of the image and leave the second section of the image in the copy of the image; and

performing the second optical character recognition pass on the copy of the image after cropping the copy of the image to remove the first section of the image from the copy of the image and leave the second section of the image in the copy of the image.

14. The system of claim 11 , wherein the first and second sections of the image are defined based on at least one of: a user defined page layout, a programmed page layout, or a detected page layout.

15. The system of claim 9 , wherein the one or more modules, when executed, further cause the processor to:

detect a plurality of segments defined by contiguous sequences of characters in the image;

perform a first optical character recognition pass on the image to identify characters in the plurality of segments of the image;

identify a first set of one or more characters located in a first segment of the image, wherein an identification of the first set of one or more characters is associated with a first segment confidence level that is at least a threshold segment confidence level;

identify a second set of one or more characters located in a second segment of the image, wherein a first identification of the second set of one or more characters is associated with a second segment confidence level that is below the threshold segment confidence level;

perform a second optical character recognition pass on the second segment of the image to identify the second set of one or more characters a second time, wherein a second identification of the second set of one or more characters is associated with a third segment confidence level; and

select the first identification of the second set of one or more characters or the second identification of the second set of one or more characters based upon a comparison of the second segment confidence level and the third segment confidence level.

16. The system of claim 15 , wherein the one or more modules, when executed, cause the processor to perform the second optical character recognition pass on the second segment of the image to identify the second set of one or more characters the second time by:

storing a copy of the image in a memory; and

cropping the copy of the image to remove portions of the image other than the second segment of the image in the copy of the image; and

performing the second optical character recognition pass on the copy of the image after cropping the copy of the image to remove portions of the image other than the second segment of the image in the copy of the image.

17. A system for iterative optical character recognition, comprising:

an imaging device;

a controller in communication with the imaging device, the controller including at least one processor configured to execute one or more modules stored by a memory that is communicatively coupled to the at least one processor, the one or more modules, when executed, causing the processor to:

perform an optical character recognition pass on an image received from the imaging device to identify a plurality of characters of the image;

identify a first character located in a first portion of the image, wherein an identification of the first character is associated with a first character confidence level that is at least a threshold character confidence level;

identify a second character located in a second portion of the image, wherein a first identification of the second character is associated with a second character confidence level that is below the threshold character confidence level;

perform an additional optical character recognition pass on the second portion of the image to identify the second character a second time, wherein a second identification of the second character is associated with a third character confidence level; and

select the first identification of the second character or the second identification of the second character based upon a comparison of the second character confidence level and the third character confidence level.

18. The system of claim 17 , wherein the one or more modules, when executed, cause the processor to perform the additional optical character recognition pass on the second portion of the image to identify the second character the second time by:

storing a copy of the image in the memory; and

modifying the copy of the image to remove the first portion of the image from the copy of the image and leave the second portion of the image in the copy of the image; and

performing the additional optical character recognition pass on the copy of the image after modifying the copy of the image to remove the first portion of the image from the copy of the image and leave the second portion of the image in the copy of the image.

19. The system of claim 17 , wherein the one or more modules, when executed, further cause the processor to:

define at least a first section and a second section of an image;

perform a first optical character recognition pass on the first section of the image to identify a first plurality of characters of the image; and

perform a second optical character recognition pass on the second section of the image to identify a second plurality of characters of the image.

20. The system of claim 17 , wherein the one or more modules, when executed, further cause the processor to:

detect a plurality of segments defined by contiguous sequences of characters in the image;

perform a first optical character recognition pass on the image to identify characters in the plurality of segments of the image;

identify a first set of one or more characters located in a first segment of the image, wherein an identification of the first set of one or more characters is associated with a first segment confidence level that is at least a threshold segment confidence level;

identify a second set of one or more characters located in a second segment of the image, wherein a first identification of the second set of one or more characters is associated with a second segment confidence level that is below the threshold segment confidence level;

perform a second optical character recognition pass on the second segment of the image to identify the second set of one or more characters a second time, wherein a second identification of the second set of one or more characters is associated with a third segment confidence level; and

select the first identification of the second set of one or more characters or the second identification of the second set of one or more characters based upon a comparison of the second segment confidence level and the third segment confidence level.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 24, 2018
From: ANNIS, DAVID
To: BUSINESS IMAGING SYSTEMS, INC.
Reel/Frame 046696/0683 →
Continuity (2)
Provisional Application 62440757 · Dec 30, 2016
Related Publication 20180189592A1 · Jul 5, 2018
Cited By (8)
US 12,216,694 US 12,288,406 US 12,346,649 US 12,417,352 US 12,450,217 US 12,488,136 US 12,493,754 US 12,675,510