Image binarization
View Patent ↗Systems and methods convert to binary an input image having pixels defining text and background. Thresholds are determined by which pixels in the input image and a corresponding blurred image will be defined as either binary black or binary white. Thresholds derive from grouped together neighboring pixels having pixels separated out that correspond to the background. For pixels of the input image defined as binary black and having corresponding pixels in the blurred image defined as binary black relative to their thresholds, those are set to black in the binary image, else they are set white. Techniques for devising thresholds, blurring images, grouping together pixels, statistical analysis, etc., typify the embodiments.
1. A method for converting an input image to a binary image, the input image having pluralities of pixels defining text and background, comprising:
blurring the input image with one or more controllers to obtain a blurred image;
determining thresholds for the pixels of the input image and corresponding pixels in the blurred image by which the pixels and corresponding pixels will be defined as either a black pixel or a white pixel; and
with the one or more controllers, setting to black pixels in the binary image those pixels of the input image being defined as said black pixels and having corresponding pixels in the blurred image being defined as said black pixels relative to the thresholds.
2. The method of claim 1 , further including grouping together neighboring pixels and devising a distribution of data therefore.
3. The method of claim 2 , further including identifying pixels that represent the background in the grouped together neighboring pixels.
4. The method of the claim 3 , further including defining an offset of pixels away from the identified pixels that have pixel value intensities similar to pixel value intensities of the identified pixels that represent the background.
5. The method of claim 3 , further including identifying a most common color of pixel value intensities in the grouped together neighboring pixels.
6. The method of claim 3 , further including separating out remaining pixels from the identified pixels that represent the background.
7. The method of claim 6 , further including determining the thresholds relative to the remaining pixels.
8. The method of claim 3 , wherein the determining the thresholds further includes applying a predetermined threshold value by which the pixels and corresponding pixels will be defined as either the black pixel or the white pixel if a significant number of pixels do not exist in the distribution of data after the identifying pixels that represent the background in the grouped together neighboring pixels.
9. The method of claim 2 , wherein the devising a distribution of data further includes determining a histogram of pixel value intensities relative to a frequency of occurrence in the grouped together neighboring pixels.
10. The method of claim 1 , wherein the blurring the input image further includes grouping together small groups of pixels and changing to lighter pixel value intensities of the plurality of pixels of the input image that are relatively dark and changing to darker pixel value intensities of the plurality of pixels of the input image that are relatively light.
11. The method of claim 10 , wherein the grouping together small groups of pixels further includes grouping together a 3×3 square of nine pixels.
12. The method of claim 2 , wherein the grouping together said neighboring pixels further includes grouping together a 21×21 square of 441 pixels, a threshold being determined for a center pixel of the square.
13. The method of claim 2 , wherein the grouping together said neighboring pixels further includes grouping together a 11×11 square of 121 pixels.
14. The method of claim 2 , further including grouping together a second group of neighboring pixels and devising a second distribution of data therefore.
15. The method of claim 14 , further including identifying pixels that represent the background in the grouped together said second group of neighboring pixels.
16. The method of claim 15 , further including separating out remaining pixels from the identified pixels that represent the background in the grouped together said second group of neighboring pixels.
17. The method of claim 16 , further including determining another threshold from the remaining pixels of the identified pixels that represent the background in the grouped together said second group of neighboring pixels.
18. The method of claim 14 , wherein the grouping together said second group of neighboring pixels further includes advancing a bounding box surrounding a centering pixel in the grouped together neighboring pixels.
19. The method of claim 18 , wherein the advancing the bounding box further includes advancing the bounding box across pixels defining a width of the input image and across pixels defining a length of the input image.
20. The method of claim 1 , further including comparing pixel value intensities of the pixels of the input image to a pixel value intensity corresponding to a threshold and determining if the pixel value intensities are less than and equal to the pixel value intensity corresponding to the threshold or are greater than.
21. The method of claim 1 , further including comparing pixel value intensities of the corresponding pixels of the blurred image to a pixel value intensity corresponding to a threshold and determining if the pixel value intensities of the corresponding pixels are less than and equal to the pixel value intensity corresponding to the threshold or are greater than.
22. The method of claim 1 , further including determining as many thresholds as there exist pixels under consideration in the input image and corresponding pixels in the blurred image.