Method and system for producing formatted data related to geometric distortions

• US 7,532,766 B2
• Filed: 06/05/2002
• Issued: 05/12/2009
• Est. Priority Date: 07/12/2001
• Status: Expired due to Term

• Current Assignee: Lens Correction Technologies Corporation
• Original Assignee: DXO Labs
• Inventors: Guichard, Frederic, Lavest, Jean-Marc, Chauville, Benoit, Liege, Bruno
• Primary Examiner(s): Bella, Matthew C
• Assistant Examiner(s): Rush, Eric
• Application Number:
  US10/482,419
• Publication Number:
  US 20040240750A1
• Time in Patent Office: 2,533 Days
• Field of Search: 382/275, 382/274, 382/269
• US Class Current: 382/254
• CPC Class Codes: G06T 1/0007 Image acquisition
• CPC Class Codes: H04N 1/387 Composing, repositioning or otherwise geometrically modifying originals
• 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 processing system and method for producing formatted information related to appliances of an appliance chain. This appliance chain includes at least one image-capture appliance and/or at least one image-restitution appliance for capturing and/or restituting an image on a medium. The system and method produce formatted information related to geometric distortions of at least one appliance of the chain. Fixed characteristics of the appliances and/or variable characteristics depending on the image can be taken into account. The fixed characteristics and/or the variable characteristics can be associated with one or more characteristic values, especially the focal length and/or the focusing. Then, measured formatted information related to the geometric distortions of the appliance are produced from a measured field. The system and method are applicable to photographic or video image processing, in optical devices, industrial controls, robotics, metrology, etc.

Claims

1. A method for producing formatted information related to appliances of an appliances chain, the appliance chain including at least one image-capture appliance and/or at least one image-restitution appliance, in which the appliance contains at least one variable characteristic depending on the image, each variable characteristic being configured to be associated with one or more values of characteristic, the method comprising:

capturing or restituting a reference scene by the appliance, to produce, on a medium, characteristic image points that are images of reference points of the reference scene;

establishing a bijection between the characteristic image points and the reference points, so as to obtain a measured field composed of set of pairs composed of one of the reference points, and of the characteristic image point associated by the bijection;

selecting one or more variable characteristics among the set of variable characteristics and including, in the measured field, the value for each image of each of the selected variable characteristic;

selecting on the medium, only four characteristic image points such that a quadrilateral defined by the four characteristic image points has a maximum area and a center of gravity situated in proximity of the geometric center of the image;

choosing a mathematical projection between the medium and a reference surface, the mathematical projection being a bilinear transformation that transforms the four characteristic image points into four specific reference points among said reference points associated by the bijection with the four characteristic image points,

wherein the measured field includes the value for the image of each of the four specific reference points and the mathematical projection onto the reference surface of the four characteristic image points, and/or of the pair composed of the four characteristic image points and the mathematical projection of the four specific reference points onto the medium; and

producing, from the measured field, said formatted information related to geometric distortions of the appliance.

2. A method according to claim 1, wherein the formatted information comprises data related to a technical characteristic of an image capture appliance at the moment of the image capture, and/or data related to preferences of a user of the image-capture appliance.

3. A method according to claim 1, wherein the formatted information produced from the formatted information is represented by parameters of a parameterizable model chosen from among a set of parameterizable models, the method further comprising:

selecting the parameterizable model within the set of parameterizable models by;

defining a maximum deviation,

ordering the parameterizable models of the set of parameterizable models in accordance with their degree of complexity of employment, and

choosing a first of the parameterizable models of the ordered set of parameterizable models such that the deviation is smaller than the maximum deviation.

4. A method according to claim 1, further comprising:

obtaining, from the formatted information, extended formatted information related to an arbitrary reference point on the reference surface and/or related to an arbitrary characteristic image point of the medium, by deducing the formatted information related to the arbitrary reference point or to the arbitrary characteristic image point.

5. A method according to claim 1, wherein

the appliance of the appliance chain is provided with at least one variable characteristic depending on the image, each variable characteristic configured to be associated with a value to form a combination composed of the set of the variable characteristics and of the values, the method further comprising:

selecting predetermined combinations; and

calculating, including employing a first calculation algorithm for each of the selected predetermined combinations.

6. A method according to claim 5, wherein

an argument is defined, depending on a case, as an arbitrary reference point on the reference surface and a combination, or an arbitrary characteristic image point of the medium and a combination, the method further comprising:

deducing, from the formatted information, extended formatted information related to an arbitrary argument.

7. A method according to claim 4, wherein to deduce the extended formatted information from the formatted information a first threshold is defined and the extended formatted information is selected such that the deviation is below the first threshold.

8. A method according to claim 1, wherein the image is a color image composed of a plurality of color planes, the method further comprising:

producing the formatted information by employing a first calculation algorithm for at least two of the color planes, by using a same mathematical projection for each of the color planes.

9. A method according to claim 1, wherein the image is a color image composed of a plurality of color planes, the method further comprising:

producing the formatted information by employing a first calculation algorithm for at least one of the color planes, by using a same virtual reference for each of the color planes, wherein the formatted information can correct chromatic aberrations of the appliance.

10. A system for producing formatted information related to appliances of an appliances chain, the appliance chain including at least one image-capture appliance and/or at least one image-restitution appliance, in which the appliance contains at least one variable characteristic depending on the image, each variable characteristic configured to be associated with one or more values of characteristic, the system comprising:

means for capturing or restituting a reference scene by the appliance, to produce, on a medium, characteristic image points that are images of reference points of the reference scene;

means for establishing a bijection between the characteristic image points and the reference points, so as to obtain a measured field composed of set of pairs composed of one of the reference points, and of the characteristic image point associated by the bijection;

means for selecting one or more variable characteristics among the set of variable characteristics and including, in the measured field, the value for each image of each of the selected variable characteristic;

means for selecting on the medium, only four characteristic image points such that a quadrilateral defined by the four characteristic image points has a maximum area and a center of gravity situated in proximity of the geometric center of the image;

means for choosing a mathematical projection between the medium and a reference surface, the mathematical projection being a bilinear transformation that transforms the four characteristic image points into four specific reference points among said reference points associated by the bijection with the four characteristic image points,

wherein the measured field includes the value for the image of each of the four specific reference points and the mathematical projection onto the reference surface of the four characteristic image points, and/or of the pair composed of the four characteristic image points and the mathematical projection of the four specific reference points onto the medium; and

means for producing, from the measured field, said formatted information related to geometric distortions of the appliance.

11. The system of claim 10, wherein the formatted information comprises data related to a technical characteristic of an image capture appliance at the moment of the image capture, and/or data related to preferences of a user of the image-capture appliance.

12. A system according to claim 10, wherein

the formatted information produced from the formatted information is represented by parameters of a parameterizable model chosen from among a set of parameterizable models, the system further comprising:

selection means for selecting the parameterizable model within the set of parameterizable models, the selection means including data-processing means for defining a maximum deviation, ordering the parameterizable models of the set of parameterizable models in accordance with their degree of complexity of employment, choosing a first of the parameterizable models of the ordered set of parameterizable models such that the deviation is smaller than the maximum deviation.

13. A system according to claim 10, further comprising:

data-processing means for obtaining, from the formatted information, extended formatted information related to an arbitrary reference point on the reference surface and/or related to an arbitrary characteristic image point of the medium, by deducing the formatted information related to the arbitrary reference point or to the arbitrary characteristic image point.

14. A system according to claim 10, wherein

the appliance of the appliance chain is provided with at least one variable characteristic depending on the image, each variable characteristic configured to be associated with a value to form a combination composed of the set of the variable characteristics and of the values, the system further comprising:

selection means for selecting predetermined combinations; and

calculating means for calculating formatted information, by employing a first calculation algorithm for each of the selected predetermined combinations.

15. A system according to claim 14, wherein

an argument designates, depending on a case, an arbitrary reference point on the reference surface and a combination, or an arbitrary characteristic image point of the medium and a combination, the system further comprising:

data-processing means for deducing, from the formatted information, extended formatted information related to an arbitrary argument.

16. A system according to claim 13, wherein to deduce the extended formatted information from the formatted information, the data-processing means includes selection means for selecting the extended formatted information such that the deviation is below a first threshold.

17. A system according to claim 10, wherein

the image is a color image composed of a plurality of color planes, the system further comprising:

data-processing means for producing the formatted information by employing a first calculation algorithm for at least two of the color planes, by using a same mathematical projection for each of the color planes, wherein the formatted information can correct distortions and/or chromatic aberrations of the appliance.

18. A system according to claim 10, wherein

the image is a color image composed of a plurality of color planes, the system further comprising:

data-processing means for producing the formatted information by employing a first calculation algorithm for at least one of the color planes, by using a same virtual reference for each of the color planes, wherein the formatted information can correct chromatic aberrations of the appliance.