Method of selecting one of a plurality of binarization programs

• US 6,411,737 B2
• Filed: 12/19/1997
• Issued: 06/25/2002
• Est. Priority Date: 12/19/1997
• Status: Expired due to Term

• Current Assignee: NCR Atleos Corp.
• Original Assignee: NCR Corporation (NCR Voyix Corporation)
• Inventors: Wesolkowski, Slawomir B., Hassanein, Khaled S.
• Primary Examiner(s): ROGERS, SCOTT A
• Application Number:
  US08/994,700
• Publication Number:
  US 20010043748A1
• Time in Patent Office: 1,649 Days
• Field of Search: 382/237, 382/270, 382/272, 382/172, 382/171, 382/169, 382/137, 382/156, 358/462, 358/465, 358/466, 358/455, 358/456
• US Class Current: 382/237
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/162 Quantising the image signal
• CPC Class Codes: H04N 1/40062 Discrimination between different image types, e.g. two-tone, continuous tone

Abstract

A document processing apparatus comprises a scanner for scanning a bank check to obtain gray scale image data associated with the bank check. A processor is provided for (i) selecting one of a plurality of binarization programs based upon the gray scale image data, and (ii) applying the binarization program selected to at least a portion of the gray scale image data to provide a binary image of at least a portion of the bank check.

Claims

1. A document processing apparatus comprising:

a scanner for scanning a bank check to obtain gray scale image data associated with the bank check;

a memory for storing a plurality of different selectable binarization programs, each of the plurality of different selectable binarization programs having a non-selectable threshold associated therewith; and

a processor for (i) selecting one of the plurality of different selectable binarization programs based upon the gray scale image data associated with the bank check, and (ii) then applying the selected binarization program to at least a portion of the gray scale image data to provide a binary image of at least a portion of the bank check.

2. A document processing apparatus according to claim 1, wherein the processor includes a neural network which is applied to at least one value which has been calculated based upon the gray scale image data.

3. A document processing apparatus according to claim 2, wherein the neural network accepts as input a value indicative of average intensity of pixels associated with the check.

4. A document processing apparatus according to claim 2, wherein the neural network accepts as input a value indicative of standard deviation of pixels associated with the check.

5. A document processing apparatus according to claim 2, wherein the neural network accepts as input a value indicative of standard deviation of pixels associated with the check.

6. A method of processing a bank check, the method comprising the steps of:

(a) scanning the bank check to obtain gray scale image data associated with the bank check;

(b) selecting one of a plurality of different selectable binarization programs having non-selectable thresholds based upon the gray scale image data obtained in step (a); and

(c) applying the binarization program selected in step (b) to at least a portion of the gray scale image data obtained in step (a) to provide a binary image of at least a portion of the bank check.

7. A method according to claim 6, wherein step (b) includes the step of:

(b-1) applying a neural network to at least one value which has been calculated based upon the gray scale image data obtained in step (a).

8. A method according to claim 7, wherein step (b-1) includes the step of:

(b-1) accepting as input a value indicative of an average intensity of pixels associated with the check.

9. A method according to claim 7, wherein step (b-1) includes the step of:

(b-1-1) accepting as input a value indicative of standard deviation of intensity of pixels associated with the check.

10. A method according to claim 7, wherein step (b-1) includes the step of:

(b-1-1) accepting as input a value indicative of histogram type of pixels associated with the check.

11. A document processing apparatus comprising:

a scanner for scanning a bank check to obtain gray scale image data associated with the bank check;

a memory for storing a plurality of different selectable binarization programs, each of the plurality of different selectable binarization programs having a single predetermined threshold associated therewith and

a processor for (i) selecting one of the plurality of different selectable binarization programs based upon the gray scale image data associated with the bank check, and (ii) then applying the selected binarization program to at least a portion of the gray scale image data to provide a binary image of at least a portion of the bank check.

12. A document processing apparatus according to claim 11, wherein the processor includes a neural network which is applied to at least one value which has been calculated based upon the gray scale image data.

13. A document processing apparatus according to claim 12, wherein the neural network accepts as input a value indicative of average intensity of pixels associated with the check.

14. A document processing apparatus according to claim 12, wherein the neural network accepts as input a value indicative of standard deviation of pixels associated with the check.

15. A document processing apparatus according to claim 12, wherein the neural network accepts as input a value indicative of histogram type of pixels associated with the check.

16. A method of processing a bank check, the method comprising the steps of:

(a) scanning the bank check to obtain gray scale image data associated with the bank check;

(b) selecting one of a plurality of different selectable binarization programs having fixed predetermined thresholds based upon the gray scale image data obtained in step (a); and

(c) applying the binarization program selected in step (b) to at least a portion of the gray scale image data obtained in step (a) to provide a binary image of at least a portion of the bank check.

17. A method according to claim 16, wherein step (b) includes the step of:

(b-1) applying a neural network to at least one value which has been calculated based upon the gray scale image data obtained in step (a).

18. A method according to claim 17, wherein step (b-1) includes the step of:

(b-1-1) accepting as input a value indicative of an average intensit y of pixels associated with the check.

19. A method according to claim 17, wherein step (b-1) includes the step of:

(b-1-1) accepting as input a value indicative of standard deviation of intensity of pixels associated with the check.

20. A method according to claim 17, w herein step (b-1-1) includes the step of:

(b-1-1) accepting as input a value indicative of histogram type of pixels