Image processing method and apparatus

• US 7,995,855 B2
• Filed: 05/06/2008
• Issued: 08/09/2011
• Est. Priority Date: 01/18/2008
• Status: Active Grant

• Current Assignee: FotoNation Limited (Adeia Inc.)
• Original Assignee: Tessera Technologies Ireland Limited (Adeia Inc.)
• Inventors: Zamfir, Adrian, Florea, Corneliu, Drimbarean, Alexandru, Albu, Felix
• Primary Examiner(s): Bhatnagar; Anand
• Assistant Examiner(s): TSAI, TSUNG YIN
• Application Number:
  US12/116,140
• Publication Number:
  US 20090185753A1
• Time in Patent Office: 1,190 Days
• Field of Search: None
• US Class Current: 382/260
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/94 based on local image properties, e.g. for local contrast enhancement
• CPC Class Codes: H04N 1/56 Processing of colour picture signals H04N1/52 takes precedence
• CPC Class Codes: H04N 1/60 Colour correction or control H04N1/54 takes precedence

Abstract

A method of processing an image includes traversing pixels of an image in a single pass over the image. An inverting function is applied to the pixels. A recursive filter is applied to the inverted pixel values. The filter has parameters which are derived from previously traversed pixel values of the image. A pixel value is combined with a filter parameter for the pixel to provide a processed pixel value for a processed image.

Claims

1. A method of processing an image comprising:

acquiring the image;

using a processor;

traversing multiple pixels of the image in a single pass over the image;

applying an inverting function to said pixels;

applying a recursive filter to one or more inverted pixel values, said filter having one or more parameters derived from previously traversed pixel values of said image; and

combining said one or more pixel values with said one or more parameters to provide processed pixel values for a processed image wherein said inverting function comprises;

2. A method according to claim 1 wherein said image comprises one or a combination of YCC or RGB format.

3. A method according to claim 2 wherein said image comprises RGB format and wherein said inverting function inverts a combination of one or more color plane values for a pixel.

4. A method according to claim 3 further comprising:

providing an estimate of the average of the red, green and blue planes from previously traversed pixel values of said image;

providing correction terms for one or more of said planes, said correction terms being dependent on said color channel average estimates; and

wherein said combining includes multiplying a pixel value with a correction term.

5. A method according to claim 4 wherein said correction terms are limited by respective upper and lower thresholds.

6. A method according to claim 4 wherein said providing an estimate of the average of said red, green and blue channels comprises:

R=β

·

{right arrow over (R)}+(1−

β

)·

G(i,j,1)

G=β

·

G+(1−

β

)·

G(i,j,2)

B=β

·

B+(1−

β

)·

G(i,j,3),

where;

G(i,j,k) is the pixel value for the respective red (R), green (G) or blue (B) color plane; and

β

is a coefficient between 0 and 1.

7. A method according to claim 4 wherein said correction terms

γ

_R,γ

_B for each of said red and blue color planes comprise;

8. A method as claimed in claim 7 wherein γ

_R and γ

_B are limited to between 0.95 and 1.05.

9. A method according to claim 4 wherein combining comprises:

G₁(i,j,1)=G(i,j,1)·

H(i,j)·

γ

_R

G₁(i,j,2)=G(i,j,2)·

H(i,j)

G₁(i,j,3)=G(i,j,3)·

H(i,j)·

γ

_B

where;

γ

_R,γ

_B are said correction terms;

H(i,j) is said filter parameter; and

G(i,j,k) is the pixel value for the respective red (R), green (G) or blue (B) color plane, oro combinations thereof.

10. A method according to claim 4 wherein combining comprises:

11. A method according to claim 2 wherein said image comprises YCC format and wherein said inverting function inverts an intensity value for a pixel.

12. A method according to claim 11 wherein said inverting function comprises:

13. A method according to claim 1 wherein said traversing comprises one or a combination of row wise;

column wise;

or traversing an arbitrary path across the image.

14. A method according to claim 1 wherein said combining comprises applying a linear model.

15. A method according to claim 14 wherein said combining comprises multiplying said pixel value, a correction term and said filter parameter for said pixel to provide a pixel value for a processed image.

16. A method according to claim 1 wherein said recursive filter parameters H(i,j) comprise:

H(i,j)=α

H(i,j−

1)+(1−

α

)(f(G(i,j,k)))

where;

α

is the pole of the filter;

G(i,j,k) is the pixel value for the respective red (R), green (G) or blue (B) color plane, or combinations thereof; and

f(G(i,j,k)) is said inverting function.

17. A method according to claim 16 wherein α

is between 0.05 and 0.8.

18. A method according to claim 16 wherein the combining includes applying a linear or logarithmic model, or both, to process said one or more pixel values to obtain said processed pixel values.

19. A method according to claim 1 wherein said image is in YCC format and said recursive filter parameters H(i,j) comprise:

H(i,j)=α

H(i,j−

1)+(1−

α

)(f(Y(i,j)))

where;

α

is the pole of the IIR filter;

Y(i,j) is said pixel intensity value; and

f(Y(i,j)) is said inverting function.

20. A method according to claim 19 wherein said inverting function comprises:

21. A method according to claim 19 wherein said combining comprises:

Y₁(i,j)=Y(i,j)[1+ε

(i,j)·

(1−

H(i,j))]

where;

H(i,j) is said filter parameter; and

ε

(i,j) is a gain factor.

22. A method according to claim 21 wherein ε

(i, j) is either constant or varies for said image.

23. The apparatus of claim 1, wherein said combining comprises applying the linear model for the luminance channel and the logarithmic model for the chrominance channels.

24. The apparatus of claim 23, wherein the applying the logarithmic model for the chrominance channels comprises using an H(i,j) coefficient computed on the luminance channel.

25. The method of claim 1, further comprising varying a degree of color or brightness boost, or both, by varying one or more parameters of the model.

26. The method of claim 1, further comprising adjusting color saturation by adjusting a parameter of the model when computing new chrominance values.

27. The method of claim 1, further comprising varying brightness of an enhanced image by adjusting a multiplier of a luminance channel by a parameter of the model.

28. The method of claim 27, wherein the parameter comprises a different parameter than that for adjusting one or more chrominance values.

29. A method according to claim 1, further comprising iteratively applying said traversing and combining to one or more successively processed images, each of said applying said inverting function and said applying said recursive filter and said traversing being respectively adapted to improve image luminance or image sharpness or both.

30. An image processing apparatus comprising:

one or more optics and a sensor for acquiring a digital image;

a processor; and

one or more storage media having digital code embodied therein for programming the processor to perform an image processing method comprising;

acquiring the image;

using the processor;

traversing multiple pixels of the image in a single pass over the image;

applying an inverting function to said pixels;

applying a recursive filter to one or more inverted pixel values, said filter having one or more parameters derived from previously traversed pixel values of said image; and

combining said one or more pixel values with said one or more parameters to provide processed pixel values for a processed image wherein said inverting function comprises;

31. The apparatus of claim 30, wherein said image is in one of YCC or RGB format.

32. The apparatus of claim 31, wherein said images comprise RGB format and wherein said inverting function inverts a combination of one or more color plane values for a pixel.

33. The apparatus of claim 32, further comprising:

providing an estimate of an average of red, green and blue planes from previously traversed pixel values of said image;

providing correction terms for one or more of said planes, said correction terms being dependent on said average; and

wherein said combining includes multiplying a pixel value with a correction term.

34. The apparatus of claim 33, wherein said correction terms are limited by respective upper and lower thresholds.

35. The apparatus of claim 31, wherein said image comprises YCC format and wherein said inverting function inverts an intensity value for a pixel.

36. The apparatus of claim 30, wherein said combining comprises applying a linear model.

37. The apparatus of claim 36, wherein said combining comprises multiplying said pixel value, a correction term and said filter parameter for said pixel to provide a pixel value for a processed image.

38. The apparatus of claim 30, wherein said traversing comprises one or a combination of row wise;

column wise;

or traversing an arbitrary path across the image.

39. The apparatus of claim 30, wherein said combining comprises applying the linear model for the luminance channel and the logarithmic model for the chrominance channels.

40. The apparatus of claim 39, wherein the applying the logarithmic model for the chrominance channels comprises using an H(i,j) coefficient computed on the luminance channel.

41. The apparatus of claim 30, wherein the method further comprises varying a degree of color or brightness boost, or both, by varying one or more parameters of the model.

42. The apparatus of claim 30, wherein the method further comprises adjusting color saturation by adjusting a parameter of the model when computing new chrominance values.

43. The apparatus of claim 30, wherein the method further comprises varying brightness of an enhanced image by adjusting a multiplier of a luminance channel by a parameter of the model.

44. The apparatus of claim 43, wherein the parameter comprises a different parameter than that for adjusting one or more chrominance values.

45. One or more non-transitory computer-readable media having code embedded therein for programming a processor to perform a method of processing an acquired image, wherein the method comprises:

traversing multiple pixels of the image in a single pass over the image;

applying an inverting function to said pixels;

applying a recursive filter to one or more inverted pixel values, said filter having one or more parameters derived from previously traversed pixel values of said image; and

combining said one or more pixel values with said one or more parameters to provide processed pixel values for a processed image wherein said inverting function comprises;

46. One or more computer-readable media according to claim 45 wherein said image comprises one or a combination of YCC or RGB format.

47. One or more computer-readable media according to claim 45 wherein said traversing comprises one or a combination of row wise;

column wise;

or traversing an arbitrary path across the image.

48. One or more computer-readable media according to claim 45 wherein said combining comprises one of a linear or a logarithmic combination, or a combination thereof.

49. One or more computer-readable media according to claim 45 wherein said recursive filter parameters H(i,j) comprise:

H(i,j)=α

H(i,j−

1)+(1−

α

)(f(G(i,j,k)))

where;

α

is the pole of the filter;

G(i,j,k) is the pixel value for the respective red (R), green (G) or blue (B) color plane, or combinations thereof; and

f(G(i,j,k)) is said inverting function.

50. One or more computer-readable media according to claim 45 wherein said image is in YCC format and said recursive filter parameters H(i,j) comprise:

H(i,j)=α

H(i,j−

1)+(1−

α

)(f(Y(i,j)))

where;

α

is the pole of the IIR filter;

Y(i,j) is said pixel intensity value; and

f(Y(i,j)) is said inverting function.

51. One or more computer-readable media according to claim 45, comprising iteratively applying the method to one or more successively processed images and the steps being respectively adapted to improve image luminance or image sharpness or both.

52. One or more computer-readable media according to claim 45, wherein said combining comprises applying the linear model for the luminance channel and the logarithmic model for the chrominance channels.

53. One or more computer-readable media according to claim 45, wherein the applying the logarithmic model for the chrominance channels comprises using an H(i,j) coefficient computed on the luminance channel.

54. One or more computer-readable media according to claim 45, wherein the method further comprises varying a degree of color or brightness boost, or both, by varying one or more parameters of the model.

55. One or more computer-readable media according to claim 45, wherein the method further comprises adjusting color saturation by adjusting a parameter of the model when computing new chrominance values.

56. One or more computer-readable media according to claim 45, wherein the method further comprises varying brightness of an enhanced image by adjusting a multiplier of a luminance channel by a parameter of the model.

57. One or more computer-readable media according to claim 56, wherein the parameter comprises a different parameter than that for adjusting one or more chrominance values.