Lighting correction for the outdoor environment with extension to the self adjusting algorithm for general lighting conditions

1. A method for identifying an object, the method comprising

acquiring an image of an object exposed to direct light and to disperse light;

calculating a first parameter indicative of the direct light and a second parameter indicative of the disperse light;

obtaining an albedo by using the first parameter, and the second parameter; and

utilizing the albedo to identify the object.

2. The method of claim 1 , wherein the step of calculating comprises the steps of, obtaining a first albedo from the image of the object; and

utilizing the first albedo to calculate the first parameter and the second parameter.

3. The method of claim 2 , wherein the step of obtaining a first albedo comprises,

calculating a first amount of light that irradiates the object; and

calculating a second amount of light that radiates from the object.

4. The method of claim 3 , wherein the step of calculating a first amount of light that irradiates the object comprises,

obtaining a mean shape of the object;

calculating a normal at one or more locations on the mean shape of the object; and

obtaining a direction of illumination of the object, wherein the first amount of light is calculated using at least the normal and the direction of illumination.

5. The method of claim 4 , further comprising the steps of:

representing the first albedo as an albedo matrix; and

smoothing the albedo matrix by averaging matrix elements of the albedo matrix.

6. The method of claim 5 , further comprising the step of calibrating the smoothed albedo matrix.

7. The method of claim 6 , wherein the calibrated albedo matrix is substantially equal to kA+(1−k)S, where A is the albedo matrix, S is the smoothed albedo matrix, and k is a coefficient that is substantially equal to about 0.7.

8. The method of claim 4 , wherein the step of utilizing the first albedo comprises minimizing a merit function of the first parameter and the second parameter.

9. The method of claim 8 , wherein the merit function is given by ∥Γ−(αC+βD)∥, where Γ is a vector that represents the image, C is a vector that represents the first albedo, D is a vector that depends on the mean shape and the direction of illumination, α denotes the first parameter, and β denotes the second parameter.

10. The method of claim 3 , wherein the step of calculating a first amount of light that irradiates the object comprises,

calculating a normal at one or more locations on the object; and

obtaining a direction of illumination of the object, wherein the normal and the direction of illumination are utilized to calculate the first amount of light.

11. The method of claim 10 , wherein, in the step of calculating a normal, each said normal is determined from shading of the image utilizing a photoclinometry process.

12. The method of claim 10 , further comprising the steps of

representing the first albedo as an albedo matrix; and

smoothing the albedo matrix by averaging matrix elements of the albedo matrix.

13. The method of claim 12 , further comprising the step of calibrating the smoothed albedo matrix.

14. The method of claim 13 , wherein the calibrated albedo matrix is substantially equal to kA+(1−k)S, where A is the albedo matrix, S is the smoothed albedo matrix, and k is a coefficient that is substantially equal to about 0.7.

15. The method of claim 3 , wherein the second amount of light that radiates from the object is obtained from the image of the object.

16. The method of claim 2 , wherein the step of utilizing the albedo comprises

associating a vector with the albedo;

calculating eigenvectors of a correlation matrix from a training set of albedos, the eigenvectors spanning a subspace;

projecting the vector onto the subspace to obtain an albedo representation of the object; and

comparing the albedo representation of the object to reference representations to find a match, thereby identifying the object.

17. A system for identifying an object, the system comprising

an image acquisition device for acquiring an image of an object exposed to direct light and to disperse light;

a merit function manipulator for calculating a first parameter indicative of the direct light and a second parameter indicative of the disperse light;

an albedo developer for obtaining an albedo associated with the image by using the first parameter, and the second parameter; and

a discrimination stage for utilizing the albedo to identify the object.

18. The system of claim 17 , further comprising an albedo recovery module for obtaining a first albedo from the image of the object, wherein the merit function manipulator utilizes the first albedo to calculate the first parameter and the second parameter.

19. The system of claim 18 , wherein the albedo recovery module calculates a first amount of light that irradiates the object, and a second amount of light that radiates from the object.

20. The system of claim 19 , wherein the albedo recovery module includes

a normal module for calculating a normal at one or more locations on a mean shape of the object; and

a light direction module for obtaining a direction of illumination of the object, wherein the first amount of light is calculated using at least the normal and the direction of illumination.

21. The system of claim 20 , wherein the albedo recovery module represents the first albedo as an albedo matrix, and smoothes the albedo matrix by averaging matrix elements of the albedo matrix.

22. The system of claim 21 , wherein the albedo recovery module calibrates the smoothed albedo matrix.

23. The system of claim 22 , wherein the calibrated albedo matrix is substantially equal to kA+(1−k)S, where A is the albedo matrix, S is the smoothed albedo matrix, and k is a coefficient that is substantially equal to about 0.7.

24. The system of claim 19 , wherein the second amount of light that radiates from the object is obtained from the image of the object.

25. The system of claim 18 , wherein the albedo recovery module comprises,

a normal module for calculating a normal at one or more locations on the object; and

a light direction module for obtaining a direction of illumination of the object, wherein the normal and the direction of illumination are utilized to calculate the first amount of light.

26. The system of claim 25 , wherein the normal module calculates the normal from shading of the image utilizing a photoclinometry process.

27. The system of claim 25 , wherein the albedo recovery module represents the first albedo as an albedo matrix, and smoothes the albedo matrix by averaging matrix elements of the albedo matrix.

28. The system of claim 27 , wherein the albedo recovery module calibrates the smoothed albedo matrix.

29. The system of claim 28 , wherein the calibrated albedo matrix is substantially equal to kA+(1−k)S, where A is the albedo matrix, S is the smoothed albedo matrix, and k is a coefficient that is substantially equal to about 0.7.

30. The system of claim 29 , wherein the merit function is given by ∥Γ−(αC+βD)∥, where Γ is a vector that represents the image, C is a vector that represents the first albedo, D is a vector that depends on the mean shape and the direction of illumination, α denotes the first parameter, and β denotes the second parameter.

31. The system of claim 18 , wherein the merit function manipulator utilizes the first albedo to minimize a merit function of the first parameter and the second parameter.

32. The system of claim 17 , further comprising

an image manipulation stage for associating a vector with the albedo, and for calculating eigenvectors, which span a subspace, of a correlation matrix from a training set of albedos; and

a compression stage for projecting the vector onto the subspace to obtain an albedo representation of the object, wherein the discrimination stage compares the albedo representation of the object to reference representations to find a match, thereby identifying the object.

33. A system for identifying an object in an image, the system comprising

an object development unit for producing a representation of the object using a first parameter indicative of direct light illuminating the image and a second parameter indicative of disperse light illuminating the image; and

a discrimination stage for matching the representation of the object to a second representation of the object held in a storage device to identify the object;

wherein the object development unit comprises an albedo recovery module for producing a first albedo by determining a first amount of light that irradiates the object, and a second amount of light that radiates from the object, the first albedo is used to produce the first parameter and the second parameter.

34. The system of claim 33 , wherein a ratio of the second amount of light to the first amount of light determines the first albedo.

35. The system of claim 33 , wherein the albedo recovery module determines the first amount of light by utilizing a direction of illumination of the object and by utilizing a normal at one or more locations on the object.

36. The system of claim 35 , wherein the normal is calculated from a mean shape of the object.

37. The system of claim 35 , wherein the normals are calculated from shading of the image utilizing a photoclinometry process.

38. The system of claim 33 , wherein the albedo recovery module represents the first albedo as an albedo matrix, and smooths the albedo matrix by averaging matrix elements of the albedo matrix.

39. The system of claim 38 , wherein the albedo recovery module calibrates the smoothed albedo matrix to yield a calibrated albedo matrix given substantially by kA+(1−k)S, where A is the albedo matrix, S is the smoothed albedo matrix, and k is a coefficient that is substantially equal to about 0.7.

40. The system of claim 33 , wherein the object development unit comprises an albedo recovery module for determining a first albedo of the object, the first albedo used to calculate the first parameter and the second parameter.

41. The system of claim 40 , wherein the object development unit further comprises at least one of

a shape development module for determining at least an approximate shape of the object from the first albedo generated by the albedo recovery module, and

a surface development module for determining at least one normal of the approximate shape of the object determined by the shape development module.

42. The system of claim 40 , wherein the albedo recovery module comprises at least one of

a normal module for determining at least one normal from a shape of the object,

a light direction module for determining a direction of light irradiating the object, and

an albedo calculator for calculating the first albedo of the object as a function of at least the light direction and the normal.

43. The system of claim 42 , wherein the albedo calculator calculates the first albedo as a function of the light direction, the normal, and image data introduced to the albedo calculator.

44. The system of claim 43 , further comprising

a head find stage for finding the head of the object, and

an eye find stage for finding the eyes of the object.

45. In a system capable of acquiring an image of an individual, a method for identifying the individual from an acquired image of the individual, the method comprising the steps of:

acquiring an image of the individual with an image acquisition device;

determining from the acquired image a first parameter and a second parameter;

generating a first representation of the individual in the acquired image;

generating a second representation of the individual in the acquired image using the first parameter, the second parameter and the first representation; and

identifying the individual in the acquired image using the second representation of the individual;

wherein the step of generating a first representation comprises the step of, determining one or more areas in the acquired image of the individual that are disregarded when generating the first representation.

46. The method of claim 45 , wherein the step of generating a first representation further comprises the step of, determining a shape of an object in the acquired image.

47. The method of claim 46 , further comprising the steps of,

determining from the shape of the object a shape of a surface of the object; and

determining a perpendicular element extending the surface of the object wherein, the perpendicular element indicates a direction of illumination in the acquired image.

48. The method of claim 45 , wherein the first representation comprises a first albedo.

49. The method of claim 45 , wherein the second representation comprises a second albedo.

50. The method of claim 45 , wherein the first parameter represents dispersed light in the acquired image.

51. The method of claim 45 , wherein the second parameter represents direct light in the acquired image.

52. A computer readable medium holding device executable instructions for performing a method to identify an individual using an amount of light reflected by the individual, the method comprising the steps of,

capturing an image of the individual with an image capturing device;

generating a first representation of the individual from the captured image using the amount of light reflected by the individual in the captured image;

generating a second representation of the individual from the first representation; and

identifying the individual using the second representation;

wherein the first representation comprises a first albedo of the individual in the captured image.

53. The computer readable medium of claim 52 , further comprising the step of, compensating for shadow and glare in the captured image.

54. The computer readable medium of claim 52 , further comprising the step of, determining a first parameter and second parameter from the captured image, wherein the first parameter and the second parameter are used in generating the first representation.

55. The computer readable medium of claim 54 , wherein the first parameter comprises dispersed light data in the captured image.

56. The computer readable medium of claim 54 , wherein the second parameter comprises direct light data in the captured image.

57. The computer readable medium of claim 52 , wherein the second representation comprises a second albedo of the individual in the captured image.