Image processing apparatus and method for binarizing a multilevel image

• US 7,167,281 B1
• Filed: 09/26/2000
• Issued: 01/23/2007
• Est. Priority Date: 11/26/1999
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Naoi, Satoshi, Fujimoto, Katsuhito, Ohara, Atsuko
• Primary Examiner(s): Grant, II; Jerome
• Assistant Examiner(s): Safaipour; Houshang
• Application Number:
  US09/669,663
• Time in Patent Office: 2,310 Days
• Field of Search: 358/464, 358/465, 358/401, 358/406, 358/474
• US Class Current: 358/464
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/162 Quantising the image signal
• CPC Class Codes: H04N 1/403 Discrimination between the two tones in the picture signal of a two-tone original

Abstract

It is judged for each pixel in an inputted multilevel image whether the pixel is a background pixel, and the pixel is locally binarized if it is judged not to be a background pixel. Then, it is judged whether the pixel belongs to a background or a stroke, such as of a character, ruled line, etc., and a binary image is generated.

Claims

1. An image processing apparatus, comprising:

a background judgment device judging whether a target pixel is a background pixel using a gray level difference and a standard deviation of gray levels of pixels in a vicinity area of the target pixel on receipt of a multilevel image, wherein

the gray level difference is an amount which is calculated based on a difference between an average gray level of white pixels in the vicinity area of the target pixel and an average gray level of black pixels in the vicinity area of the target pixel; and

a local binarization device locally binarizing the target pixel and outputting a binary image if it is judged that the target pixel is not the background pixel.

2. The apparatus according to claim 1, wherein said local binarization device uses an amount which is calculated based on an average and a standard deviation of gray levels of pixels in the vicinity area of the target pixel as a binarization threshold for the target pixel.

3. The apparatus according to claim 2, wherein the amount which is calculated based on the average and the standard deviation of the gray levels of the pixels in the vicinity area of the target pixel is calculated based on a sum of the average and a constant-multiple of the standard deviation.

4. The apparatus according to claim 2, wherein the vicinity area of the target pixel is a rectangular area of N×

N with a prescribed number of pixels N and the target pixel located at a center.

5. The apparatus according to claim 1, wherein said background judgment device judges whether the target pixel is the background pixel, using a standard deviation of gray levels of pixels in the vicinity area of the target pixel.

6. The apparatus according to claim 5, wherein said background judgment device judges whether the target pixel is the background pixel under a background judgment condition of σ

<

σ

_min with σ

as the standard deviation in the vicinity area of the target pixel and a min as a prescribed constant.

7. The apparatus according to claim 1, wherein said background judgment device judges whether the target pixel is the background pixel using a standard deviation of gray levels and a gray level difference of pixels in the vicinity area of the target pixel.

9. The apparatus according to claim 7, wherein said background judgment device judges whether the target pixel is the background pixel under a background judgment condition of Δ

g<

g_min with Δ

g as the gray level difference in the vicinity of the target pixel and Δ

g_min as a prescribed constant.

10. The apparatus according to claim 7, wherein the gray level difference is an amount which is calculated based on a difference between an average gray level of white pixels in the vicinity area of the target pixel and an average gray level of black pixels in the vicinity area of the target pixel.

12. The apparatus according to claim 1, further comprising:

a line element restriction device executing a process of the obtained binary image based on a ratio of black pixels in a shape-fixed line element mask including the target pixel and outputting a binary image.

13. The apparatus according to claim 12, wherein

said line element restriction device leaves the black pixels in the line element mask as black pixels if the ratio of black pixels In the line element mask is a prescribed ratio or more.

14. The apparatus according to claim 12, wherein said line element restriction device converts all pixels in the line element mask Into black pixels if the ratio of black pixels in the line element mask is a prescribed ratio or more.

15. The apparatus according to claim 12, wherein

said line element restriction device converts all pixels in the line element mask into white pixels if the ratio of black pixels in the line element mask is less than a prescribed ratio.

16. The apparatus according to claim 12, wherein said line element restriction device uses a plurality of line element masks.

17. The apparatus according to claim 1, further comprising:

a stroke separation device applying a partial pattern in a gray scale image corresponding to a black pixel joint element in the obtained binary Image and separating strokes of different gray levels.

18. The apparatus according to claim 17, wherein said stroke separation device judges whether to perform a stroke separation using one of an inter-class dispersion and a dispersion ratio between different strokes.

19. The apparatus according to claim 1, wherein

said local binarization device judges which of the background and the stroke a pixel, which is judged to be the background pixel by said background judgment device, belongs to based on a gray level of the pixel.

8. An image processing apparatus, comprising:

a background judgment device judging for each target pixel whether the target pixel is a background pixel on receipt of a multilevel image, and

a local binarization device locally binarizing the target pixel, judging which of a background and a stroke the target pixel belongs to, and outputting a binary image if it is judged that the target pixel is not the background pixel,

wherein said background judgment device judges whether the target pixel is the background pixel using standard deviation of gray levels and a gray level difference of pixels in a vicinity area of the target pixel, and

wherein said background judgment device judges whether the target pixel is the background pixel under a background judgment condition of r=σ

/Δ

g<

r min with σ

as the standard deviation in the vicinity area of the target pixel, Δ

g as the gray level difference in the vicinity of the target pixel and r_min as a prescribed constant.

11. An image processing apparatus, comprising:

a background judgment device judging for each target pixel whether the target pixel is a background pixel on receipt of a multilevel image, and

a local binarization device locally binarizing the target pixel, judging which of a background and a stroke the target pixel belongs to, and outputting a binary image if it is judged that the target pixel is not the background pixel, wherein said background judgment device judges whether the target pixel is the background pixel using standard deviation of gray levels and a gray level difference of pixels in a vicinity area of the target pixel, and

wherein said background judgment device judges whether the target pixel is a background pixel using a combination of a background judgment conditions σ

<

σ

_min, r=σ

/Δ

g<

r_min and Δ

g<

Δ

g_min with σ

as the standard deviation in the vicinity area of the target pixel, Δ

g as the gray level difference in the vicinity of the target pixel and σ

_min, r_min and Δ

g_min as a prescribed constant.

20. An image processing apparatus, comprising:

judging for each target pixel whether a target pixel is a background pixel using a gray level difference and a standard deviation of gray levels of pixels in a vicinity area of the target pixel on receipt of a multilevel image, wherein

the gray level difference is an amount which is calculated based on a difference between an average gray level of white pixels in the vicinity area of the target pixel and an average gray level of black pixels in the vicinity area of the target pixel; and

locally binarizing the target pixel and outputting a binary image if it is judged that the target pixel is not the background pixel.

21. The method according to claim 20, further comprising:

processing the obtained binary image based on a ratio of black pixels in a shape-fixed line element mask including the target pixel; and

outputting a binary image.

22. The method according to claim 20, further comprising:

binarizing a partial pattern in a gray scale image corresponding to a black pixel joint element in the obtained binary image; and

separating strokes of different gray levels.

23. A computer-readable storage medium on which is recorded a program for enabling a computer extracting a stroke included in an inputted multilevel image to perform a process, said process comprising:

judging for each target pixel whether a target pixel is a background pixel using a gray level difference and a standard deviation of gray levels of pixels in a vicinity area of the target pixel on receipt of a multilevel image, wherein

the gray level difference is an amount which is calculated based on a difference between an average gray level of white pixels in the vicinity area of the target pixel and an average gray level of black pixels in the vicinity area of the target pixel; and

locally binarizing the target pixel and outputting a binary image if it is judged that the target pixel is not the background pixel.

24. The storage medium according to claim 23, said process further comprising:

processing the obtained binary image based on a ratio of black pixels in a shape-fixed line element mask including the target pixel; and

outputting a binary image.

25. The storage medium according to claim 23, said process further comprising:

binarizing a partial pattern in a gray scale image corresponding to a black pixel Joint element in the obtained binary image; and

separating strokes of different gray levels.

26. An image processing apparatus, comprising:

input means for receiving a multilevel image; and

background judgment means for judging whether a target pixel is a background pixel using a gray level difference and a standard deviation of gray levels of pixels in a vicinity area of the target pixel;

wherein

the gray level difference is an amount which is calculated based on a difference between an average gray level of white pixels in the vicinity area of the target pixel and an average gray level of black pixels in the vicinity area of the target pixel; and

local binarizatiod means for locally binarizing the target pixel and outputting a binary image if it is judqed that the target pixel is not the background pixel.

27. A transmission signal transmitting to a computer, which extracts a stroke included in an inputted multilevel image, a program for enabling the computer to perform a process, said process comprising:

judging for each target pixel whether a target pixel is a background pixel using a gray level difference and a standard deviation of gray levels of pixels in a vicinity area of the target pixel on receipt of a multilevel image, wherein

the gray level difference is an amount which is calculated based on a difference between an average gray level of white pixels in the vicinity area of the target pixel and an average gray level of black pixels in the vicinity area of the target pixel; and

locally binarizing the target pixel and outputting a binary image if it is judged that the target pixel is not the background pixel.

28. An image processing method, comprising:

receiving a multilevel image;

measuring an average gray level of white pixels in the vicinity area of the target pixel of the multilevel image;

measuring an average gray level of black pixels in the vicinity area of the target pixel;

calculating a gray level difference based on a difference between the average gray level of white pixels and the average gray level of black pixels in the vicinity area of the target pixel; and

judging whether the target pixel is a background pixel using the gray level difference and a standard deviation of gray levels of pixels in the vicinity area of the target pixel.