IMAGE BINARIZATION BASED ON GREY MEMBERSHIP PARAMETER OF PIXELS

• US 20130188865A1
• Filed: 03/07/2013
• Published: 07/25/2013
• Est. Priority Date: 12/02/2009
• Status: Active Grant

• Current Assignee: Jadavpur University
• Original Assignee: Jadavpur University
• Inventors: SAHA, Satadal, BASU, Subhadip, NASIPURI, Mita, BASU, Dipak Kumar
• Granted Patent: US 8,744,208 B2
• Time in Patent Office: Days
• US Class Current: 382/164
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/136 involving thresholding
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: H04N 1/40012 Conversion of colour to monochrome
• CPC Class Codes: H04N 1/40062 Discrimination between different image types, e.g. two-tone, continuous tone
• CPC Class Codes: H04N 1/4078 using gradational references, e.g. grey-scale test pattern analysis

Abstract

Briefly, in accordance with one aspect, a method of binarizing an image is provided. The method includes partitioning the image into a plurality of image segments, each image segment having a plurality of image pixels and partitioning each of the image segments into subsegments, each image subsegment having a plurality of image pixels. The method also includes estimating a grey membership parameter for each image pixel for each of the plurality of image segments and subsegments, combining the grey membership parameter for each of the plurality of image pixels from each of the plurality of image segments and subsegments to estimate a net grey membership parameter for each image pixel and assigning black or white color to each of the plurality of image pixels based on the estimated net grey membership parameter of the respective pixel.

Claims

1. A method comprising:

partitioning an image into a first plurality of image segments, of which a first image segment includes a first plurality of image pixels;

partitioning the image into a second plurality of image segments, of which a second image segment includes a second plurality of image pixels;

estimating a grey value for an image pixel of the first and second plurality of image segments;

estimating mean and standard deviations of the grey value of the image pixel; and

assigning a first grey membership parameter to the image pixel having the standard deviation below the pre-determined threshold and assigning a second grey membership parameter to the image pixel having the standard deviation above the pre-determined threshold.

2. The method of claim 1, further comprising estimating the first grey membership parameter by dividing the mean of the grey value with a pre-determined maximum grey value of the image pixel.

3. The method of claim 1, further comprising estimating the second grey membership parameter by dividing a contrast-stretched grey value with a pre-determined maximum grey value of the image pixel.

4. The method of claim 1, further comprising combining the first and second grey membership parameter to estimate a net grey membership parameter for the image pixel of the first and second plurality of image segments.

5. The method of claim 4, further comprising assigning black or white to the image pixel of the first and second plurality of image segments based on the net estimated grey membership parameter.

6. The method of claim 4, wherein the combining comprises determining a weighted average of the first and second grey membership parameter to estimate the net grey membership parameter.

7. The method of claim 6, wherein the weighted average comprises a function of a number of image segments in the first and second plurality of image segments.

8. The method of claim 6, wherein the weighted average is a function of a square of a number of image segments in the first and second plurality of image segments.

9. The method of claim 1, wherein the estimating the grey value comprising enhancing a contrast level of the first and second plurality of image.

10. The method of claim 9, wherein the enhancing the contrast level comprises stretching the contrast level of the image pixel using histogram equalization.

11. A method comprising:

partitioning an image into a first plurality of image segments, of which a first image segment includes a first plurality of image pixels;

partitioning the image into a second plurality of image segments, of which a second image segment includes a second plurality of image pixels;

estimating a grey value for an image pixel of the first and second plurality of image segments by enhancing a contrast level of the first and second plurality of image segments;

estimating a grey membership parameter for the image pixel based upon the grey value of the image pixel; and

assigning black or white to the image pixel based on at least one of the grey membership parameter and a number of partition levels.

12. The method of claim 11, further comprising assigning a black color to the image pixel when a net grey membership parameter is below a first predetermined threshold and assigning a white color to the image pixel when the net grey membership parameter is above the predetermined threshold.

13. The method of claim 11, wherein the enhancing the contrast level comprises stretching the contrast level of the image pixel using histogram equalization.

14. An image binarization system, comprising:

a memory circuit configured to store an image; and

an image processing circuit configured to;

partition the image into a first plurality of image segments, of which a first image segment includes a first plurality of image pixels;

partition the image into a second plurality of image segments, of which a second image segment includes a second plurality of image pixels;

estimate a grey value for an image pixel of the first and second plurality of image segments;

estimate mean and standard deviation of the grey value of the image pixel; and

assign a first grey membership parameter to the image pixel having the standard deviation below the pre-determined threshold and assign a second grey membership parameter to the image pixel having the standard deviation above the pre-determined threshold.

15. The image binarization system of claim 14, wherein the image processing circuit is further configured to determine a weighted average of grey membership parameters for the image pixel from the first and second plurality of image pixels to estimate the net grey membership parameter.

16. The image binarization system of claim 14, wherein the image processing circuit is further configured to generate a binary output image by assigning black or white color to the image pixel based upon an estimated net grey membership parameter of the image pixel.

17. The image binarization system of claim 14, wherein the image processing circuit is further configured to reduce noise in the binary output image using morphological open-close technique.

18. The image binarization system of claim 14, further comprising enhance a contrast level of the plurality of first and second image segments using histogram equalization.

19. The method of claim 15, wherein the enhancing the contrast level comprises stretching the contrast level of the image pixel using histogram equalization.