Systems and methods for adaptive binarization of an image

• US 8,391,599 B1
• Filed: 10/17/2008
• Issued: 03/05/2013
• Est. Priority Date: 10/17/2008
• Status: Active Grant

• Current Assignee: United Services Automobile Association
• Original Assignee: United Services Automobile Association
• Inventors: Medina, Reynaldo III
• Primary Examiner(s): Mehta, Bhavesh
• Assistant Examiner(s): SHAH, UTPAL D
• Application Number:
  US12/253,278
• Time in Patent Office: 1,600 Days
• Field of Search: 382/168, 382/172, 382/171
• US Class Current: 382/168
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/162 Quantising the image signal

Abstract

A system for processing an image for binarization comprises at least one subsystem that breaks the image into multiple sub-images, at least one subsystem that generates a histogram for each sub-image, and at least one subsystem that determines optimal thresholding values for image binarization by statistical analysis of the histogram for each sub-image.

Claims

1. A system for processing an image for binarization comprising:

a memory; and

a processor configured to execute instructions stored in the memory to;

break the image into multiple sub-images;

generate a histogram for a sub-image among the multiple sub-images; and

determine a thresholding value for image binarization by statistical analysis of the histogram for the sub-image, where determining the thresholding value comprises performing an image processing clamp method on the sub-image to find the thresholding value, with input to the clamp method being boundary gray level values determined from statistical analysis of the histogram of the sub-image.

2. The system of claim 1 where the processor is further configured to execute instructions stored in the memory to:

find a highest Y axis value on the histogram, denoted as second_high herein, wherein a Y axis on the histogram is counts of gray and an X axis on the histogram is gray value;

find a second_highest Y value on the histogram, denoted herein as first_high;

obtain a gray index value of the first_high;

find a lowest Y value traversing from left to right on the histogram, denoted as first_min herein, using the first_high and a calculated value as boundaries;

find a lowest Y value traversing from right to left on the histogram, denoted as second_min herein, using the first_min and the second_high as boundaries; and

where the input to the clamp method is boundary gray level values of the first_min and the second_min.

3. The system of claim 2 wherein the processor is further configured to execute instructions stored in the memory to binarize the sub-image using the thresholding value of the clamp method as input.

4. The system of claim 1 where the processor is further configured to execute instructions stored in the memory to binarize the sub-image using the determined thresholding value.

5. The system of claim 1 where the processor is further configured to execute instructions stored in the memory to:

binarize the multiple sub-images using determined thresholding values; and

put the binarized multiple sub-images back together to create a single binarized image.

6. The system of claim 5 where the processor is further configured to execute instructions stored in the memory to use the single binarized image to read characters on the single binarized image.

7. A method for processing an image for binarization comprising:

breaking the image into multiple sub-images;

generating a histogram for a sub-image among the multiple sub-images; and

determining a thresholding value for image binarization by statistical analysis of the histogram for the sub-image, where determining the thresholding value comprises performing an image processing clamp method on the sub-image to find the thresholding value, with input to the clamp method being boundary gray level values determined from statistical analysis of the histogram of the sub-image.

8. The method of claim 7 wherein the determining the thresholding value comprises:

finding a highest Y axis value on the histogram, denoted as second_high herein, wherein a Y axis on the histogram is counts of gray and an X axis on the histogram is gray value;

finding a second_highest Y value on the histogram, denoted herein as first_high;

obtaining a gray index value of the first_high;

finding a lowest Y value traversing from left to right on the histogram, denoted as first min herein, using the first_high and a calculated value as boundaries;

finding a lowest Y value traversing from right to left on the histogram, denoted as second_min herein, using the first_min and the second_high as boundaries; and

where the input to the clamp method is boundary gray level values of the first_min and the second_min.

9. The method of claim 8 further comprising binarizing the sub-image using the thresholding value of the clamp method as input.

10. The method of claim 7 further comprising binarizing the sub-image using the determined thresholding value.

11. The method of claim 7 further comprising:

binarizing the multiple sub-images using determined thresholding values; and

putting the binarized multiple sub-images back together to create a single binarized image.

12. The method of claim 11 further comprising using the single binarized image to read characters on the single binarized image.

13. A non-transitory computer readable medium for processing an image for binarization comprising computer readable instructions for:

breaking the image into multiple sub-images;

generating a histogram for a sub-image among the multiple sub-images; and

determining a thresholding value for image binarization by statistical analysis of the histogram for the sub-image, where determining the thresholding value comprises performing an image processing clamp method on the sub-image to find the thresholding value, with input to the clamp method being boundary gray level values determined from statistical analysis of the histogram of the sub-image.

14. The non-transitory computer readable medium of claim 13 wherein the computer readable instructions for determining the thresholding value comprise computer readable instructions for:

finding a highest Y axis value on the histogram, denoted as second_high herein, wherein a Y axis on the histogram is counts of gray and an X axis on the histogram is gray value;

finding a second_highest Y value on the histogram, denoted herein as first_high;

obtaining a gray index value of the first_high;

finding a lowest Y value traversing from left to right on the histogram, denoted as first_min herein, using the first_high and a calculated value as boundaries;

finding a lowest Y value traversing from right to left on the histogram, denoted as second_min herein, using the first_min and the second_high as boundaries; and

where the input to the clamp method is boundary gray level values of the first_min and the second_min.

15. The non-transitory computer readable medium of claim 14 further comprising computer readable instructions for binarizing the sub-image using the thresholding value of the clamp method as input.

16. The non-transitory computer readable medium of claim 13 further comprising computer readable instructions for binarizing the sub-image using the determined thresholding value.

17. The non-transitory computer readable medium of claim 13 further comprising computer readable instructions for:

binarizing the multiple sub-images using determined thresholding values; and

putting the binarized multiple sub-images back together to create a single binarized image.

18. The non-transitory computer readable medium of claim 13 further comprising computer readable instructions for using the single binarized image to read characters on the single binarized image.