Method and system for modifying a digital image taking into account it's noise

• US 20040247196A1
• Filed: 07/15/2004
• Published: 12/09/2004
• Est. Priority Date: 07/12/2001
• Status: Active Grant

• Current Assignee: Lens Correction Technologies Corporation
• Original Assignee: DXO Labs
• Inventors: Chanas, Laurent, Guichard, Frederic, Liege, Bruno, Moisan, Lionel
• Granted Patent: US 7,343,040 B2
• Time in Patent Office: Days
• US Class Current: 382/254
• CPC Class Codes: G06T 1/0007 Image acquisition
• CPC Class Codes: G06T 3/10 Selection of transformation methods according to the characteristics of the input images
• CPC Class Codes: G06T 5/00 Image enhancement or restoration
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06T 5/73 Deblurring; Sharpening
• CPC Class Codes: G06T 5/80 Geometric correction
• CPC Class Codes: H04N 1/00007 relating to particular apparatus or devices
• CPC Class Codes: H04N 1/00045 using a reference pattern designed for the purpose, e.g. a test chart
• CPC Class Codes: H04N 1/00071 characterised by the action taken
• CPC Class Codes: H04N 1/387 Composing, repositioning or otherwise geometrically modifying originals
• CPC Class Codes: H04N 1/40093 Modification of content of picture, e.g. retouching geometric modifications H04N1/387
• CPC Class Codes: H04N 1/58 Edge or detail enhancement; Noise or error suppression, e.g. colour misregistration correction H04N1/62 takes precedence

Abstract

A system and a method for calculating a transformed image from a digital image and from formatted information related to defects of an appliance chain for image capture and/or restitution. The system and method provide for automatically determining the characteristic noise data from formatted information and/or from the digital image. The transformed image can therefore be corrected such that it does not exhibit any visible or annoying defects, especially defects related to noise, as regards its subsequent use. The system and method are applicable to photographic or video image processing, in optical devices, industrial controls, robotics, metrology, etc.

Claims

1-29. -29. (Canceled).

30. A method for obtaining a transformed image from a digital image of an appliance chain, the appliance chain containing image-capture appliances and/or image-restitution appliances, the appliance chain containing at least one appliance;

the method comprising;

automatically determining characteristic data from formatted information related to defects of the appliance chain and/or from the digital image, as characteristic noise data;

calculating the transformed image from the formatted information and from the characteristic noise data;

wherein the transformed image does not exhibit any visible or annoying defect, especially defects related to noise, as regards its subsequent use.

31. A method according to claim 30, the method further comprising determining the characteristic noise data by:

selecting analysis zones over the digital image, especially as a function of the appliances of the appliance chain and/or of the formatted information,

calculating local brightness variations over the analysis zones,

deducing the characteristic noise data as a function of a statistical calculation of occurrence of the local variations over the set of the analysis zones.

32. A method according to claim 31, the method further comprising, for deducing the characteristic noise data:

constructing a histogram of occurrences of the local brightness variations,

selecting, on the histogram, at least one part of a part situated before a first local maximum, including this maximum;

such that the local brightness variations related to the noise are then obtained.

33. A method according to claim 31, the method further comprising, for selection of analysis zones over the digital image, classifying the analysis zones according to their mean brightness, to obtain classes;

the method further comprising;

deducing the characteristic noise data for the analysis zones belonging to a same class;

iterating the deducing for each of the classes;

such that characteristic noise data as a function of brightness are then obtained.

34. A method according to claim 30, wherein the formatted information contains the characteristic noise data.

35. A method according to claim 30, the method further comprising employing a transformation algorithm for constructing an intermediate digital image;

the algorithm having an advantage of making desired modifications to the digital image but suffering from a disadvantage of increasing the noise of the intermediate digital image.

36. A method according to claim 39, to calculate a transformed image from the intermediate digital image obtained from the digital image, the method further comprising:

employing a function whose purpose is to modify brightness of the digital image and which has, as arguments, at least;

the brightness of a point of the intermediate digital image,

the brightnesses of a zone around the corresponding point of the digital image,

the characteristic noise data;

such that there is then obtained a transformed image exhibiting the desired characteristics and a controlled noise level.

37. A method according to claim 36, wherein the intermediate digital image is composed of the digital image.

38. A method according to claim 30, wherein the method calculates a transformed image corrected for all or part of the blurring, the method further comprising:

selecting, within the digital image, image zones to be corrected,

constructing, for each image zone to be corrected that has been selected, an enhancement profile based on the formatted information and on the characteristic noise data,

correcting, as a function of the enhancement profile, each image zone to be corrected that has been selected, to obtain a transformed image zone,

combining the transformed image zones to obtain the transformed image of the digital image;

such that a deblurred transformed image is then obtained.

39. A method according to claim 38, wherein the formatted information makes it possible to determine, for each image zone to be corrected, an image representation and a reference representation in a base related to the image zone to be corrected, the method to construct an enhancement profile from formatted information and noise, the method further comprising:

determining a profile from the image representation and from the reference representation, while taking the noise into account as the case may be,

determining a parameterized operator with which it is possible to pass from the image representation to the profile;

such that the set of values of parameters of the parameterized operator constitutes the enhancement profile.

40. A method according to claim 39, the method further comprising, for correction of each image zone to be corrected as a function of the enhancement profile:

representing at least partly, in the base, the image zone to be corrected,

applying the parameterized operator to a representation obtained at an end of the representing to obtain a corrected representation of the image zone to be corrected,

substituting the representation of the image zone to be corrected by the corrected representation of the image zone to be corrected, so as to obtain a transformed image zone.

41. A method according to claim 38, the method further comprising:

calculating, from the transformed image, an image having a controlled noise level, by employing a function whose purpose is to modify brightness of the digital image and which has, as arguments, at least;

brightness of a point of the transformed digital image,

brightnesses of a zone around the corresponding point of the digital image,

characteristic noise data;

such that there is then obtained a deblurred image having a controlled noise level.

42. A method according to claim 30, wherein the formatted information depends on values of variable characteristics depending on the digital image, the method further comprising:

determining the value or values of the variable characteristics for the digital image.

43. A method according to claim 30, the method configured to calculate a transformed image from a digital image and from formatted information related to the defects of an appliance chain containing at least one image-restitution appliance, the restitution appliance having a dynamic range, the transformed image having a dynamic range, the method further comprising adapting the dynamic range of the transformed image to the dynamic range of the restitution appliance.

44. Application to a case of a digital image composed of color planes, the application comprising applying, to each color plane, the method according to claim 30.

45. A system for obtaining a transformed image from a digital image of an appliance chain, the appliance chain containing image-capture appliances and/or image-restitution appliances, the appliance chain containing at least one appliance, the system comprising:

data-processing means for automatically determining characteristic data from formatted information related to defects of the appliance chain and/or from the digital image as characteristic noise data;

computer processing means for calculating the transformed image from the formatted information and from the characteristic noise data;

the transformed image does not exhibit any visible or annoying defect, especially defects related to noise, as regards its subsequent use.

46. A system according to claim 45, wherein the data-processing means for determining the characteristic noise data includes:

selection means for selecting analysis zones over the digital image, especially as a function of the appliances of the appliance chain and/or of the formatted information,

calculating means for calculating local brightness variations over the analysis zones,

deducing means for deducing the characteristic noise data as a function of a statistical calculation of occurrence of the local variations over the set of the analysis zones.

47. A system according to claim 46, the deducing means comprising:

means for constructing a histogram of occurrences of the local brightness variations,

selection means for selecting, on the histogram, at least one part of a part situated before a first local maximum, including this maximum.

48. A system according to claim 46, the system further comprising:

for selection of analysis zones over the digital image, classification means for classifying the analysis zones according to their mean brightness, to obtain classes;

the system further comprising data-processing means for;

deducing the characteristic noise data for the analysis zones belonging to a same class,

iterating the deducing for each of the classes.

49. A system according to claim 45, wherein the formatted information contains the characteristic noise data.

50. A system according to claim 45, the system further comprising data-processing means employing a transformation algorithm for constructing an intermediate digital image;

the algorithm having an advantage of making the desired modifications to the digital image but suffering from a disadvantage of increasing the noise of the intermediate digital image.

51. A system according to claim 50, to calculate a transformed image from the intermediate digital image obtained from the digital image, the system further comprising:

calculating means employing a function whose purpose is to modify brightness of the digital image and which has, as arguments, at least;

the brightness of a point of the intermediate digital image,

the brightnesses of a zone around the corresponding point of the digital image,

the characteristic noise data.

52. A system according to claim 51, wherein the intermediate digital image is composed of the digital image.

53. A system according to claim 45, wherein the system calculates a transformed image corrected for all or part of blurring, the system further comprising:

selection means for selecting, within the digital image, image zones to be corrected,

calculating means for constructing, for each image zone to be corrected that has been selected, an enhancement profile based on the formatted information and on the characteristic noise data,

data-processing means for;

correcting, as a function of the enhancement profile, each image zone to be corrected that has been selected, to obtain a transformed image zone, and for

combining the transformed image zones to obtain the transformed image of the digital image.

54. A system according to claim 53, wherein the formatted information makes it possible to determine, for each image zone to be corrected, an image representation and a reference representation in a base related to the image zone to be corrected;

wherein in the system the calculating means for constructing an enhancement profile from formatted information and noise further comprises means for determining;

a profile from the image representation and from the reference representation, while taking the noise into account as the case may be,

a parameterized operator with which it is possible to pass from the image representation to the profile.

55. A system according to claim 54, wherein the data-processing means for correction of each image zone to be corrected as a function of the enhancement profile comprising calculating means for:

representing at least partly, in the base, the image zone to be corrected,

applying the parameterized operator to the representation of the image zone to be corrected, to obtain a corrected representation of the image zone to be corrected,

substituting the representation of the image zone to be corrected by the corrected representation of the image zone to be corrected, so as to obtain a transformed image zone.

56. A system according to claim 53, further comprising:

calculating means for calculating, from the transformed image, an image having a controlled noise level, by employing a function whose purpose is to modify brightness of the digital image and which has, as arguments, at least;

the brightness of a point of the transformed digital image,

the brightnesses of a zone around the corresponding point of the digital image,

the characteristic noise data.

57. A system according to claim 45, wherein the formatted information depends on values of variable characteristics depending on the digital image, the system further comprising:

calculating means for determining the value or values of the variable characteristics for the digital image.

58. A system according to claim 45, to calculate a transformed image from a digital image and from formatted information related to the defects of an appliance chain containing at least one image-restitution appliance, the restitution appliance having a dynamic range, the transformed image having a dynamic range, the system further comprising:

data-processing means for adapting the dynamic range of the transformed image to the dynamic range of the restitution appliance.