Method and system for altering defects in a digital image

• US 6,487,321 B1
• Filed: 09/15/2000
• Issued: 11/26/2002
• Est. Priority Date: 09/16/1999
• Status: Expired due to Term

• Current Assignee: Kodak Alaris Inc (KPP (no. 2) Trustees Ltd.)
• Original Assignee: Applied Science Fiction, Inc.
• Inventors: Edgar, Albert D., Lee, Raymond S.
• Primary Examiner(s): ROGERS, SCOTT A
• Application Number:
  US09/662,774
• Time in Patent Office: 802 Days
• Field of Search: 382/261, 382/262, 382/260, 382/264, 382/275, 358/463, 356/237.1, 356/239.1, 356/239.7, 356/239.8, 356/237.2, 356/237.3, 250/341.8, 250/339.11
• US Class Current: 382/260
• CPC Class Codes: H04N 1/4097 Removing errors due external factors, e.g. dust, scratches
• CPC Class Codes: H04N 1/58 Edge or detail enhancement; Noise or error suppression, e.g. colour misregistration correction H04N1/62 takes precedence

Abstract

One aspect of the invention is a method for altering defects in a digital image. At least a first pixel of a first channel of a digital image is filtered using digital circuitry to produce a filtered pixel by averaging the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel. At least one of the pixels is weighted in response to the intensity value of at least one pixel in a defect channel associated with the digital image. A corrected digital image is produced using the digital circuitry in response to the filtered pixel and the first channel of the digital image.

Claims

1. A method for altering defects in a digital image, comprising:

storing on a computer readable storage medium a digital image comprising at least one channel, the at least one channel comprising a plurality of pixels each having an intensity value;

storing on the computer readable storage medium a defect channel comprising a plurality of pixels each having an intensity value, at least some of the plurality of pixels in the defect channel having an intensity value proportional to defects in the digital image;

filtering a first pixel of the at least one channel of the digital image to produce a filtered pixel the filtering of the first pixel responsive to the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of at least one pixel in the defect channel; and

producing a corrected digital image in response to the filtered pixel and the at least one channel.

2. The method of claim 1:

wherein the digital image comprises three channels, each channel comprising a plurality of pixels;

wherein the digital image represents a color image; and

wherein the method further comprises

filtering at least a second pixel at the same location in each of the three channels to produce a second filtered pixel for each of the three channels, the filtering of the second pixel in each channel, for a particular channel, responsive to the intensity of the second pixel in the particular channel and a plurality of additional pixels in the neighborhood of the second pixel in the particular channel, wherein at least one of the pixels used in filtering the second pixel in the particular channel is weighted in response to the intensity value of at least one pixel in the defect channel;

producing a corrected digital image in response to the filtered pixel, the second filtered pixel for each channel, and the three channels.

3. The method of claim 1, wherein the digital image comprises a single channel and represents a monochrome image.

4. The method of claim 1, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of a plurality of pixels in the defect channel.

5. The method of claim 1, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of a pixel in the defect channel which spatially corresponds to the first pixel.

6. The method of claim 1, wherein the weighting of the at least one of the pixels used in filtering the first pixel is further responsive to red leakage in the defect channel.

7. The method of claim 1, wherein the weighting of the at least one of the pixels used in filtering the first pixel is further responsive to a clear film intensity measure.

8. The method of claim 1, wherein the filtering is performed in log space.

9. The method of claim 8, wherein the corrected digital image is further responsive to the difference between the digital image and a filtered version of the digital image containing the filtered pixel.

10. The method of claim 8, wherein the corrected digital image is further responsive to the difference between (a) the difference between the digital image and a filtered version of the digital image containing the filtered pixel and (b) the difference between the original defect channel and a filtered version of the defect channel.

11. The method of claim 1, wherein the intensity of pixels in the defect channel is responsive to the difference between at least two reflected scans of an image from which the digital image was derived, wherein the image is illuminated by at least one light source at a different angle relative to the surface of the image during each scan.

12. The method of claim 1, wherein filtering the first pixel further comprises:

filtering the first pixel a plurality of times, wherein ones of the plurality of filtering operations cover a different bandwidth, are responsive to the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel, and wherein at least one of the pixels use in filtering the first pixel is weighted in response to the intensity value of at least one pixel in the defect channel, each of the plurality of filtering operations producing a filtered pixel, collectively comprising a plurality of filtered pixels; and

wherein the corrected digital image is further responsive to the plurality of filtered pixels.

13. The method of claim 1,

wherein the digital image comprises three channels, each channel comprising a plurality of pixels;

wherein the digital image represents a color image;

wherein the at least one channel comprises one of the three channels; and

wherein the corrected digital image is produced in response to the filtered pixel by applying a correction in all three channels in response to the filtered pixel determined for the at least one channel.

14. The method of claim 1,

wherein the digital image comprises three channels, each channel comprising a plurality of pixels;

wherein the digital image represents a color image;

wherein the method further comprises forming a black and white channel based upon the three channels, the black and white channel comprising the at least one channel; and

wherein the corrected digital image is produced in response to the filtered pixel by applying a correction in all three channels in response to the filtered pixel determined for the black and white channel.

15. The method of claim 1, wherein the intensity of pixels in the defect channel is proportional to infrared transmission through film from which the digital image was derived.

16. The method of claim 15, further comprising:

assigning a low weight value to a pixel for the filtering operation if a spatially corresponding pixel in the defect channel has an intensity less than a first threshold.

17. The method of claim 15, further comprising:

assigning a low weight value to a pixel for the filtering operation if the average intensity of a spatially corresponding pixel in the defect channel and other pixels in the defect channel in the neighborhood of the spatially corresponding pixel is less than a first threshold.

18. The method of claim 1, further comprising:

dividing the digital image into a plurality of segments;

wherein the first pixel is in a first segment comprising one of the plurality of segments of the at least one channel of the digital image and wherein the plurality of additional pixels in the neighborhood of the first pixel are in the first segment;

filtering a second pixel of the at least one channel of the digital image to produce a second filtered pixel the filtering of the second pixel responsive to the intensity of the second pixel and a plurality of additional pixels in the neighborhood of the second pixel, wherein at least one of the pixels used in filtering the second pixel is weighted in response to the intensity value of at least one pixel in the defect channel;

wherein the second pixel is in a second segment comprising one of the plurality of segments of the at least one channel of the digital image and wherein the plurality of additional pixels in the neighborhood of the second pixel are in the second segment; and

wherein the corrected digital image is produced in response the filtered pixel, the second filtered pixel, and the at least one channel.

19. The method of claim 1, wherein the first pixel is filtered using a square filter.

20. The method of claim 1, wherein the first pixel is filtered using a circular filter.

62. The method of claim 1, wherein the first pixel is filtered using a window filter with feathered edges.

63. The method of claim 1, further comprising:

filtering a second pixel comprising a part of the defect channel to produce a second filtered pixel the filtering of the second pixel responsive to the intensity of the second pixel and a plurality of additional pixels in the neighborhood of the second pixel, wherein at least one of the pixels used in filtering the second pixel is weighted in response to the intensity value of at least one pixel in the defect channel.

64. The method of claim 63, wherein the corrected digital image is produced in response to the filtered pixel, the second filtered pixel, the at least one channel, and the defect channel.

65. The method of claim 1, further comprising:

prior to filtering the first pixel, altering at least a second pixel of the at least one channel by subtracting a first value responsive to the intensity value of a pixel in the defect channel spatially corresponding to the second pixel'"'"'s position in the at least one channel from a second value responsive to the intensity value of the second pixel.

21. A method for altering defects in a digital image, comprising:

filtering, using digital circuitry, a first pixel of a first channel of the digital image to produce a filtered pixel by averaging the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel, wherein at least one of the pixels used for averaging is weighted in response to the intensity value of at least one pixel in a defect channel associated with the digital image; and

producing a corrected digital image, using digital circuitry, in response to the filtered pixel and the first channel.

22. The method of claim 21, wherein the digital circuitry comprises dedicated digital hardware.

23. The method of claim 21, wherein the digital circuitry comprises a microprocessor executing computer software.

24. The method of claim 21,

wherein the digital image comprises three channels, each channel comprising a plurality of pixels;

wherein the digital image represents a color image; and

wherein the method further comprises

filtering at least a second pixel at the same location in each of the three channels to produce a second filtered pixel for each of the three channels, the filtering of the second pixel in each channel, for a particular channel, responsive to the intensity of the second pixel in the particular channel and a plurality of additional pixels in the neighborhood of the second pixel in the particular channel, wherein at least one of the pixels used in filtering the second pixel in the particular channel is weighted in response to the intensity value of at least one pixel in the defect channel;

producing a corrected digital image in response to the filtered pixel, the second filtered pixel for each channel, and the three channels.

25. The method of claim 21, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of a plurality of pixels in the defect channel.

26. The method of claim 21, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of a pixel in the defect channel which spatially corresponds to the first pixel.

27. The method of claim 21, further comprising:

filtering the first pixel a plurality of times, wherein ones of the plurality of filtering operations cover a different bandwidth, are responsive to the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel, and wherein at least one of the pixels use in filtering the first pixel is weighted in response to the intensity value of at least one pixel in the defect channel, each of the plurality of filtering operations producing a filtered pixel, collectively comprising a plurality of filtered pixels; and

wherein the corrected digital image is further responsive to the plurality of filtered pixels.

28. The method of claim 21, further comprising:

assigning a low weight value to a pixel for the filtering operation if a spatially corresponding pixel in the defect channel has an intensity indicating a high probability that the pixel is defective.

29. The method of claim 21, further comprising:

assigning a low weight value to a pixel for the filtering operation if the average intensity of a spatially corresponding pixel in the defect channel and other pixels in the defect channel in the neighborhood of the spatially corresponding pixel indicates a high probability that the pixel is defective.

30. The method of claim 21, further comprising:

performing a filtering operation on each pixel in the at least one channel.

31. The method of claim 21, further comprising:

dividing the digital image into a plurality of segments;

wherein the first pixel is in a first segment comprising one of the plurality of segments of the at least one channel of the digital image and wherein the plurality of additional pixels in the neighborhood of the first pixel are in the first segment;

filtering a second pixel of the at least one channel of the digital image to produce a second filtered pixel the filtering of the second pixel responsive to the intensity of the second pixel and a plurality of additional pixels in the neighborhood of the second pixel, wherein at least one of the pixels used in filtering the second pixel is weighted in response to the intensity value of at least one pixel in the defect channel;

wherein the second pixel is in a second segment comprising one of the plurality of segments of the at least one channel of the digital image and wherein the plurality of additional pixels in the neighborhood of the second pixel are in the second segment; and

wherein the corrected digital image is produced in response the filtered pixel, the second filtered pixel, and the at least one channel.

32. A digital image scanning system comprising:

scanning hardware operable to scan an image and convert the image into a digital image having at least one channel and to produce a defect channel responsive to defects in the image; and

computer software associated with the scanning hardware and operable to;

filter a first pixel of the at least one channel of the digital image to produce a filtered pixel the filtering of the first pixel responsive to the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of at least one pixel in the defect channel; and

produce a corrected digital image in response to the filtered pixel and the at least one channel.

33. The digital image scanning system of claim 32, wherein the digital image comprises three channels, each channel comprising a plurality of pixels;

wherein the digital image represents a color image; and

wherein the computer software is further operable to

filter at least a second pixel at the same location in each of the three channels to produce a second filtered pixel for each of the three channels, the filtering of the second pixel in each channel, for a particular channel, responsive to the intensity of the second pixel in the particular channel and a plurality of additional pixels in the neighborhood of the second pixel in the particular channel, wherein at least one of the pixels used in filtering the second pixel in the particular channel is weighted in response to the intensity value of at least one pixel in the defect channel;

produce a corrected digital image in response to the filtered pixel, the second filtered pixel for each channel, and the three channels.

34. The digital image scanning system of claim 32, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of a plurality of pixels in the defect channel.

35. The digital image scanning system of claim 32, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of a pixel in the defect channel which spatially corresponds to the first pixel.

36. The digital image scanning system of claim 32, wherein the computer software is further operable to:

filter the first pixel a plurality of times, wherein ones of the plurality of filtering operations cover a different bandwidth, are responsive to the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel, and wherein at least one of the pixels use in filtering the first pixel is weighted in response to the intensity value of at least one pixel in the defect channel, each of the plurality of filtering operations producing a filtered pixel, collectively comprising a plurality of filtered pixels; and

wherein the corrected digital image is further responsive to the plurality of filtered pixels.

37. The digital image scanning system of claim 32, wherein the computer software is further operable to:

assign a low weight value to a pixel for the filtering operation if a spatially corresponding pixel in the defect channel has an intensity indicating a high probability that the pixel is defective.

38. The digital image scanning system of claim 32, wherein the computer software is further operable to:

assign a low weight value to a pixel for the filtering operation if the average intensity of a spatially corresponding pixel in the defect channel and other pixels in the defect channel in the neighborhood of the spatially corresponding pixel indicates a high probability that the pixel is defective.

39. The digital image scanning system of claim 32, wherein the computer software is further operable to:

divide the digital image into a plurality of segments;

wherein the first pixel is in a first segment comprising one of the plurality of segments of the at least one channel of the digital image and wherein the plurality of additional pixels in the neighborhood of the first pixel are in the first segment;

filter a second pixel of the at least one channel of the digital image to produce a second filtered pixel the filtering of the second pixel responsive to the intensity of the second pixel and a plurality of additional pixels in the neighborhood of the second pixel, wherein at least one of the pixels used in filtering the second pixel is weighted in response to the intensity value of at least one pixel in the defect channel;

wherein the second pixel is in a second segment comprising one of the plurality of segments of the at least one channel of the digital image and wherein the plurality of additional pixels in the neighborhood of the second pixel are in the second segment; and

wherein the corrected digital image is produced in response the filtered pixel, the second filtered pixel, and the at least one channel.

40. A digital image processing system comprising:

a computer readable storage medium;

computer software stored on the computer readable storage medium and operable to;

filter a first pixel of the at least one channel of the digital image to produce a filtered pixel the filtering of the first pixel responsive to the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of at least one pixel in the defect channel; and

produce a corrected digital image in response to the filtered pixel and the at least one channel.

41. The digital image processing system of claim 40, wherein the digital image comprises three channels, each channel comprising a plurality of pixels;

wherein the digital image represents a color image; and

wherein the computer software is further operable to

filter at least a second pixel at the same location in each of the three channels to produce a second filtered pixel for each of the three channels, the filtering of the second pixel in each channel, for a particular channel, responsive to the intensity of the second pixel in the particular channel and a plurality of additional pixels in the neighborhood of the second pixel in the particular channel, wherein at least one of the pixels used in filtering the second pixel in the particular channel is weighted in response to the intensity value of at least one pixel in the defect channel;

produce a corrected digital image in response to the filtered pixel, the second filtered pixel for each channel, and the three channels.

42. The digital image processing system of claim 40, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of a plurality of pixels in the defect channel.

43. The digital image processing system of claim 40, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of a pixel in the defect channel which spatially corresponds to the first pixel.

44. The digital image scanning system of claim 40, wherein the computer software is further operable to:

filter the first pixel a plurality of times, wherein ones of the plurality of filtering operations cover a different bandwidth, are responsive to the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel, and wherein at least one of the pixels use in filtering the first pixel is weighted in response to the intensity value of at least one pixel in the defect channel, each of the plurality of filtering operations producing a filtered pixel, collectively comprising a plurality of filtered pixels; and

wherein the corrected digital image is further responsive to the plurality of filtered pixels.

45. The digital image scanning system of claim 40, wherein the computer software is further operable to:

assign a low weight value to a pixel for the filtering operation if a spatially corresponding pixel in the defect channel has an intensity indicating a high probability that the pixel is defective.

46. The digital image scanning system of claim 40, wherein the computer software is further operable to:

assign a low weight value to a pixel for the filtering operation if the average intensity of a spatially corresponding pixel in the defect channel and other pixels in the defect channel in the neighborhood of the spatially corresponding pixel indicates a high probability that the pixel is defective.

47. The digital image scanning system of claim 40, wherein the computer software is further operable to:

divide the digital image into a plurality of segments;

wherein the first pixel is in a first segment comprising one of the plurality of segments of the at least one channel of the digital image and wherein the plurality of additional pixels in the neighborhood of the first pixel are in the first segment;

filter a second pixel of the at least one channel of the digital image to produce a second filtered pixel the filtering of the second pixel responsive to the intensity of the second pixel and a plurality of additional pixels in the neighborhood of the second pixel, wherein at least one of the pixels used in filtering the second pixel is weighted in response to the intensity value of at least one pixel in the defect channel;

wherein the second pixel is in a second segment comprising one of the plurality of segments of the at least one channel of the digital image and wherein the plurality of additional pixels in the neighborhood of the second pixel are in the second segment; and

wherein the corrected digital image is produced in response the filtered pixel, the second filtered pixel, and the at least one channel.

48. An altered digital image derived from a digital image having at least one channel, comprising:

a computer readable storage medium;

an altered digital image stored on the computer readable storage medium wherein the altered digital image was created by;

filtering a first pixel of the at least one channel of the digital image to produce a filtered pixel the filtering of the first pixel responsive to the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel, wherein at least one of the pixels used in filtering the first pixel is weighted in response to the intensity value of at least one pixel in the defect channel; and

producing a corrected digital image in response to the filtered pixel and the at least one channel.

49. The altered digital image of claim 48, wherein the digital image comprises three channels, each channel comprising a plurality of pixels;

and

wherein the digital image represents a color image and wherein the altered digital image was further created by;

filtering at least a second pixel at the same location in each of the three channels to produce a second filtered pixel for each of the three channels, the filtering of the second pixel in each channel, for a particular channel, responsive to the intensity of the second pixel in the particular channel and a plurality of additional pixels in the neighborhood of the second pixel in the particular channel, wherein at least one of the pixels used in filtering the second pixel in the particular channel is weighted in response to the intensity value of at least one pixel in the defect channel;

producing a corrected digital image in response to the filtered pixel, the second filtered pixel for each channel, and the three channels.

50. The altered digital image of claim 48, wherein the altered digital image was further created by:

filtering the first pixel a plurality of times, wherein ones of the plurality of filtering operations cover a different bandwidth, are responsive to the intensity of the first pixel and a plurality of additional pixels in the neighborhood of the first pixel, and wherein at least one of the pixels use in filtering the first pixel is weighted in response to the intensity value of at least one pixel in the defect channel, each of the plurality of filtering operations producing a filtered pixel, collectively comprising a plurality of filtered pixels; and

wherein the corrected digital image is further responsive to the plurality of filtered pixels.

51. A method for altering defects in a digital image, comprising:

storing on a computer readable storage medium a digital image comprising at least one channel, the at least one channel comprising a plurality of pixels each having an intensity value;

storing on the computer readable storage medium a defect channel comprising a plurality of pixels each having an intensity value, at least some of the plurality of pixels in the defect channel having an intensity value proportional to defects in the digital image;

creating, using digital circuitry, a plurality of filtered versions of the at least one channel, wherein pixels of the filtered versions of the at least one channel are determined by, for a particular pixel in a particular filtered version, computing a weighted average of the intensity values of the particular pixel and pixels within a region adjacent to the particular pixel in the at least one channel, wherein at least some of the intensity values used in the computation of the weighted average are weighted in response to the intensity value of at least one pixel in the defect channel; and

producing a corrected digital image in response to the plurality of filtered versions of the at least one channel, the original at least one channel, and the defect channel.

52. The method of claim 51, further comprising:

creating, using digital circuitry, a plurality of filtered versions of the defect channel, wherein pixels of the filtered versions of the defect channel are determined by, for a particular pixel in a particular filtered version, computing a weighted average of the intensity values of the particular pixel and pixels within a region adjacent to the particular pixel, wherein at least some of the intensity values used in the computation of the weighted average are weighted in response to the intensity value of at least one pixel in the defect channel; and

wherein the corrected digital image is further produced in response to the plurality of filtered versions of the defect channel.

53. The method of claim 52, wherein the filtered versions of the at least one channel are created in log space.

54. The method of claim 51, further comprising:

dividing the at least one channel into segments;

creating, using digital circuitry, a plurality of filtered versions of each segment of the at least one channel wherein pixels of the filtered versions of each segment of the at least one channel are determined by, for a particular pixel in a particular filtered version, computing a weighted average of the intensity values of the particular pixel and pixels within a region adjacent to the particular pixel and within the same segment as the particular pixel, and wherein at least some of the intensity values used in the computation of the weighted average are weighted in response to the intensity value of at least one pixel in the defect channel; and

wherein the the corrected digital image is further produced in response to the plurality of filtered versions of each segment of the at least one channel.

55. The method of claim 51, further comprising:

prior to creating the plurality of filtered versions of the at least one channel, altering a plurality of pixels of the at least one channel by subtracting, for a particular pixel, a first value responsive to the intensity value of a pixel in the defect channel spatially corresponding to the particular pixel'"'"'s position in the at least one channel from a second value responsive to the intensity value of the particular pixel.

56. The method of claim 51, wherein the corrected digital image is produced by:

creating a series of frequency bands, the frequency bands essentially contiguous and responsive to the at least one channel and the plurality of filtered versions of the at least one channel; and

adding the frequency bands together to form the corrected digital image.

57. The method of claim 55, wherein the corrected digital image is produced by:

creating a series of frequency bands, the frequency bands essentially contiguous and responsive to the altered version of the at least one channel and the plurality of filtered versions of the altered version of the at least one channel; and

adding the frequency bands together to form the corrected digital image.

58. The method of claim 52, wherein the corrected digital image is produced by:

creating a first series of frequency bands, the first series of frequency bands essentially contiguous and responsive to the at least one channel and the plurality of filtered versions of the at least one channel;

creating a second series of frequency bands, the second series of frequency bands essentially contiguous and responsive to the defect channel and the plurality of filtered versions of the defect channel and wherein each of the second series of frequency bands corresponds to one of the frequency bands of the first series of frequency bands;

subtracting each corresponding frequency band in the second series of frequency bands from a frequency band in the first series of frequency bands to form a third series of frequency bands;

adding the third series of frequency bands together to form the corrected digital image.

59. The method of claim 58, wherein the subtracting comprises a bounded subtraction wherein the bounded subtraction, for a particular pixel in a particular frequency band, comprises the selection of a result of a plurality of subtractions including a first subtraction proportional to the difference between a first value proportional to the intensity of the particular pixel in the particular frequency band in the first series of frequency bands and a second value proportional to the intensity of the particular pixel in the particular frequency band in the second series of frequency bands and a second subtraction proportional to the difference between the first value and a third value proportional to the intensity of the particular pixel in the particular frequency band in the second series of frequency bands.

60. The method of claim 58, wherein the second series of frequency bands is further responsive to red residue in the defect channel such that the second series of frequency bands is adjusted to reduce red residue in the defect channel.

61. The method of claim 58, wherein the weighted average comprise an average selected from the group consisting of a median average, an arithmetic average, a geometric average, a mean average, and a mode average.