Noise reduction method utilizing probabilistic weighting, apparatus, and program for digital image processing

• US 6,681,054 B1
• Filed: 05/01/2000
• Issued: 01/20/2004
• Est. Priority Date: 05/01/2000
• Status: Expired due to Fees

• Current Assignee: Eastman Kodak Company
• Original Assignee: Eastman Kodak Company
• Inventors: Gindele, Edward B.
• Primary Examiner(s): Johns, Andrew W.
• Assistant Examiner(s): Alavi, Amir
• Application Number:
  US09/564,409
• Time in Patent Office: 1,359 Days
• Field of Search: 382/173, 382/224, 382/254, 382/260, 382/272, 382/275, 348/606, 348/624, 358/1.9, 358/3.26, 358/3.27, 358/447, 358/461, 358/463
• US Class Current: 382/272
• CPC Class Codes: G06T 2207/20076 Probabilistic image processing
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06V 10/30 Noise filtering

Abstract

A method of processing a digital image channel to remove noise, includes the steps of: identifying a pixel of interest; identifying at least two sampled local regions of pixels about the pixel of interest; calculating a noise free pixel estimate for each sampled local region of pixels; using the noise free pixel estimates, a noise model and the pixel of interest to calculate a probabilistic weighting factor for each noise free pixel estimate; and using the noise free pixel estimates and the probabilistic weighting factors for calculating a noise reduced pixel value.

Claims

1. A method of processing a digital image channel to remove noise, comprising the steps of:

a) identifying a pixel of interest;

b) identifying at least two sampled local regions of pixels about the pixel of interest;

c) calculating a noise free pixel estimate for each sampled local region of pixels;

d) using the noise free pixel estimates, a noise model of the expected noise level in the digital image channel and the pixel of interest to calculate a probabilistic weighting factor for each noise free pixel estimate; and

e) using the noise free pixel estimates and the probabilistic weighting factors for calculating a noise reduced pixel value.

2. The method claimed in claim 1, wherein the noise model is a Gaussian probability function.

3. A computer program product for performing the method of claim 2.

4. The method claimed in claim 1, wherein the noise model is a Laplacian probability function.

5. A computer program product for performing the method of claim 4.

6. The method claimed in claim 1, wherein the noise free pixel estimate is calculated as a rank order statistical value derived from the values of the pixels from the sampled local regions of pixels.

7. The method claimed in claim 6, wherein the rank order statistical value is the statistical median of the pixel values in the sampled local region.

8. A computer program product for performing the method of claim 7.

9. A computer program product for performing the method of claim 6.

10. The method claimed in claim 1, wherein the noise free pixel estimate is calculated as a function of the difference between the value of the pixel of interest and the pixel values in the sampled local region.

11. A computer program product for performing the method of claim 10.

12. The method claimed in claim 1, wherein one of the sampled local regions is the pixel of interest.

13. A computer program product for performing the method of claim 12.

14. The method claimed in claim 1, wherein the sampled local regions include the pixel of interest.

15. A computer program product for performing the method of claim 14.

16. The method claimed in claim 1, wherein the sampled local regions do not include the pixel of interest.

17. A computer program product for performing the method of claims 16.

18. The method claimed in claim 1, wherein the probabilistic weighting factor is calculated using a noise level parameter to calculate the probabilistic weighting factors.

19. A computer program product for performing the method of claim 18.

20. The method claimed in claim 18, wherein the noise level parameter is dependent on the value of the pixel of interest.

21. A computer program product for performing the method of claim 20.

22. The method claimed in claim 1, wherein steps a) through e) are repeated for all pixels in the image.

23. A computer program product for performing the method of claim 22.

24. The method claimed in claim 1, wherein the image includes a plurality of channels and wherein steps a) through e) are repeated for all pixels in all of the channels of the digital image.

25. A computer program product for performing the method of claim 24.

26. The method claimed in claim 1, wherein the sampled local regions define radial patterns of pixels emanating from the pixel of interest.

27. A computer program product for performing the method of claim 26.

28. A computer program product for performing the method of claim 1.

29. Apparatus for processing a digital image channel to remove noise, comprising:

a) means for identifying a pixel of interest;

b) means for identifying at least two sampled local regions of pixels about the pixel of interest;

c) means for calculating a noise free pixel estimate for each sampled local region of pixels;

d) means for using the noise free pixel estimates, a noise model of the expected noise level in the digital image channel and the pixel of interest to calculate a probabilistic weighting factor for each noise free pixel estimate; and

e) means for using the noise free pixel estimates and the probabilistic weighting factors for calculating a noise reduced pixel value.

30. The apparatus claimed in claim 29, wherein the noise model is a Gaussian probability function.

31. The apparatus claimed in claim 29, wherein the noise model is a Laplacian probability function.

32. The apparatus claimed in claim 29, wherein the noise free pixel estimate is calculated as a rank order statistical value derived from the values of the pixels from the sampled local regions of pixels.

33. The apparatus claimed in claim 32, wherein the rank order statistical value is the statistical median of the pixel values in the sampled local region.

34. The apparatus claimed in claim 29, wherein the noise free pixel estimate is calculated as a function of the difference between the value of the pixel of interest and the pixel values in the sampled local region.

35. The apparatus claimed in claim 29, wherein one of the sampled local regions is the pixel of interest.

36. The apparatus claimed in claim 29, wherein the sampled local regions include the pixel of interest.

37. The apparatus claimed in claim 29, wherein the sampled local regions do not include the pixel of interest.

38. The apparatus claimed in claim 29, wherein the probabilistic weighting factor is calculated using a noise level parameter to calculate the probabilistic weighting factors.

39. The apparatus claimed in claim 38, wherein the noise level parameter is dependent on the value of the pixel of interest.

40. The apparatus claimed in claim 29, wherein the image includes a plurality of channels and wherein means a) through e) are provided for each of the channels of the digital image.

41. The apparatus claimed in claim 29, wherein the sampled local regions define radial patterns of pixels emanating from the pixel of interest.