IP Library Granted Patent US 7,231,098
Granted Patent B2
US 7,231,098 · App. 10/320,681 · Granted Jun 12, 2007

Shading noise filter

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 7,231,098
App. No.
10/320,681
Granted
Jun 12, 2007
Kind
B2
Abstract

A shading correction is employed for a scanner to correct shading distortion. However, an image corrected with the shading corrective curve has shading noise lines due to the effects of various factors in the producing process of the shading corrective curve. The characteristic of the shading noise is that the each value of any primary color channel, of each pixel in a line is higher or lower than of the adjacent two pixels in other lines, wherein a color channel is one of red, green, or blue channel. Hence, the quality of the image is improved by removing the shading noise detected from the characteristic described above.

Claims (47)

1. A method for eliminating shading noise, comprising:

detecting one or more shading noise lines of a shading noise in an image based on two adjacent lines;

eliminating said shading noise from said one or more shading noise lines in said image;

wherein said detecting one or more shading noise lines comprises detecting one or more shading noise lines by a characteristic of said shading noise; and

wherein a detecting function of said shading noise is: [X(i−1,j)−X(i,j)][X(i,j)−X(i+1,j)]<0, wherein X(i,j) comprises one color value in a coordinate (i,j) of an image, and wherein X(i−1,j) and X(i+1,j) comprise color values of same said color as X(i,j).

2. The method for eliminating shading noise according to claim 1 , wherein said characteristic comprises a line with one pixel width and at least one color value of one color in said shading noise line higher than adjacent two pixels in said two adjacent lines.

3. The method for eliminating shading noise according to claim 1 , wherein said characteristic comprises a line with one pixel width and at least one color value of one color in said shading noise line higher a certain percent than adjacent two pixels in said two adjacent lines.

4. The method for eliminating shading noise according to claim 1 , wherein said characteristic comprises a line with one pixel width and at least one color value of one color in said shading noise line lower than adjacent two pixels in said two adjacent lines.

5. The method for eliminating shading noise according to claim 1 , wherein said characteristic comprises a line with one pixel width and at least one color value of one color in said shading noise line lower a certain percent than adjacent two pixels in said two adjacent lines.

6. The method for eliminating shading noise according to claim 1 , wherein a line in said image has shading noise if a detecting result of said detecting function is true over a certain percent.

7. The method for eliminating shading noise according to claim 6 , wherein said certain percent comprises 80 percent.

8. The method for eliminating shading noise according to claim 2 , wherein said color value is red, green, or blue.

9. The method for eliminating shading noise according to claim 3 , wherein said color value is red, green, or blue.

10. The method for eliminating shading noise according to claim 1 , wherein said shading noises are detected by multiplying a first difference with a second difference, wherein the first difference is taken between a color value of a first adjacent pixel and a color value of a given pixel, and wherein the second difference is taken between a color value of a second adjacent pixel and said color value of said given pixel.

11. The method for eliminating shading noise according to claim 10 , wherein a new color value of said given pixel comprises an average of said color value of said first adjacent pixel and said color value of said second adjacent pixel for eliminating said shading noise.

12. The method for eliminating shading noise according to claim 10 , wherein said color value of said given pixel is not transformed if detecting shading noise is false.

13. A method for eliminating shading noises, comprising:

detecting a shading noise in an image by multiplying a first difference with a second difference, wherein the first difference is taken between color values of a first adjacent pixel and a color value of a given pixel, and wherein the second difference is taken between a color value of a second adjacent pixel and said color value of said given pixel;

determining one or more shading noise lines; and eliminating said shading noise from said one or more shading noise lines in said image; and

wherein a detecting function of said shading noise is: [X(i−1,j)−X(i,j)][X(i,j)−X(i+1,j)]<0, wherein X(i,j) comprises one color value in a coordinate (i,j) of an image, and wherein X(i−1,j) and X(i+1,j) comprise color values of same said color as X(i,j).

14. The method for eliminating shading noise according to claim 13 , wherein a line in said image has shading noise if a detecting result of said detecting function is true over a certain percent.

15. The method for eliminating shading noise according to claim 14 , wherein said certain percent comprises 80 percent.

16. The method for eliminating shading noise according to claim 13 , wherein a new color value of said given pixel comprises an average of said color value of said first adjacent pixel and said color value of said second adjacent pixel for eliminating said shading noise.

17. The method for eliminating shading noise according to claim 13 , wherein said color value of said given pixel is not transformed if detecting shading noise is false.

18. An apparatus, comprising:

means for detecting one or more shading noise lines of a shading noise in an image based on two adjacent lines;

means for eliminating said shading noise from said one or more shading noise lines in said image;

wherein said means for detecting one or more shading noise lines comprises means for detecting one or more shading noise lines by a characteristic of said shading noise; and

wherein said means for detecting one or more shading noise lines comprises a detecting function of said shading noise comprising: [X(i−1,j)−X(i,j)][X(i,j)−X(i+1,j)]<0 wherein X(i,j) comprises one color value in a coordinate (i,j) of an image; X(i−1,j) and X(i+1,j) comprise color values of same said color.

19. The apparatus of claim 18 , wherein said characteristic comprises a line with one pixel width and at least one color value of one color in said shading noise line higher than adjacent two pixels in said two adjacent lines.

20. The apparatus of claim 18 , wherein said characteristic comprises a line with one pixel width and at least one color value of one color in said shading noise line higher a certain percent than adjacent two pixels in said two adjacent lines.

21. The apparatus of claim 18 , wherein said characteristic comprises a line with one pixel width and at least one color value of one color in said shading noise line lower than adjacent two pixels in said two adjacent lines.

22. The apparatus of claim 18 , wherein said characteristic comprises a line with one pixel width and at least one color value of one color in said shading noise line lower a certain percent than adjacent two pixels in said two adjacent lines.

23. The apparatus of claim 18 , wherein a line in said image has shading noise if a detecting result of said detecting function is true over a certain percent.

24. The apparatus of claim 23 , wherein said certain percent comprises 80 percent.

25. The apparatus of claim 18 , wherein said means for detecting one or more shading noise lines comprises means for detecting one or more shading noise lines by multiplying a first difference with a second difference, wherein the first difference is taken between a color value of a first adjacent pixel and a color value of a given pixel, and wherein the second difference is taken between a color value of a second adjacent pixel and said color value of said given pixel.

26. The apparatus of claim 18 , wherein a new color value of said given pixel comprises an average of said color value of said first adjacent pixel and said color value of said second adjacent pixel for eliminating said shading noise.

27. The apparatus of claim 25 , wherein said color value of said given pixel is not transformed if detecting shading noise is false.

28. An apparatus, comprising:

a scanner, said scanner capable of:

detecting a shading noise in an image by multiplying a first difference with a second difference, wherein the first difference is taken between color values of a first adjacent pixel and a color value of a given pixel, and wherein the second difference is taken between a color value of a second adjacent pixel and said color value of said given pixel;

determining one or more shading noise lines; and eliminating said shading noise from said one or more shading noise lines in said image; and

wherein said scanner is further capable of detecting a shading noise by a detecting function of said shading noise comprising: [X(i−1,j)−X(i,j)][X(i,j)−X(i+1,j)]<0 wherein X(i,j) comprises one color value in a coordinate (i,j) of an image; X(i−1,j) and X(i+1,j) comprise color values of same said color.

29. The apparatus of claim 28 , wherein a line in said image has shading noise if a detecting result of said detecting function is true over, a certain percent.

30. The apparatus of claim 29 , wherein said certain percent comprises 80 percent.

31. The apparatus of claim 28 , wherein a new color value of said given pixel comprises an average of said color value of said first adjacent pixel and said color value of said second adjacent pixel for eliminating said shading noise.

32. The apparatus of claim 28 , wherein said color value of said given pixel is not transformed if detecting shading noise is false.

Assignments (3)
MERGER Recorded Jun 18, 2013
From: TITUSVILLE CANAVERAL LLC
To: INTELLECTUAL VENTURES I LLC
Reel/Frame 030639/0330 →
MERGER Recorded Jun 17, 2013
From: TRANSPACIFIC SYSTEMS, LLC
To: TITUSVILLE CANAVERAL LLC
Reel/Frame 030628/0681 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 14, 2009
From: TRANSPACIFIC IP LTD.
To: TRANSPACIFIC SYSTEMS, LLC
Reel/Frame 023107/0267 →