Dynamic process for identifying objects in multi-dimensional data

• US 6,850,627 B2
• Filed: 03/07/2003
• Issued: 02/01/2005
• Est. Priority Date: 10/22/1999
• Status: Expired due to Fees

• Current Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Original Assignee: Lockheed Martin Corporation (Martin Marietta Corporation)
• Inventors: Phillips, Ronald William
• Primary Examiner(s): Johns, Andrew W.
• Application Number:
  US10/384,312
• Publication Number:
  US 20040022414A1
• Time in Patent Office: 697 Days
• Field of Search: 382/103, 382/190, 382/191, 706/52, 706/59, 706/900
• US Class Current: 382/103
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes

Abstract

A method for identifying a detected object in multi-dimensional imagery data is disclosed. The method includes, (a) providing (1) a set of comparative feature values associated with one or more known objects in one or more specified observation conditions and (2) a set of observation-related values including (i) a set of feature values for one or more features of an object as observed in an environment, (ii) a set of condition values for one or more conditions of the observation, including one or more of an orientation value, a depression value, a resolution value, and an obscuration value; and (b) generating one or more identification values for the object by comparing the set of feature values to the set of comparative feature values, including compensating for the set of condition values.

Claims

1. A method for identifying an object in observation-related data of an object in a surrounding environment comprising:

(a) providing;

(1) a set of comparative feature values associated with one or more known objects in or more specified observation conditions; and

(2) a set of observation-related values including;

(i) a set of observed feature values for one or more features of an object as observed in an environment; and

(ii) a set of condition values for one or more conditions of the observation; and

(b) generating one or more identification values for the observed object by comparing the set of observed feature values to the set of comparative feature values, including compensating for the set of condition values.

2. The method of claim 1, wherein compensating for the set of condition values includes modifying at least one feature value to an expected value to compensate for the obscuration value.

3. The method of claim 1, wherein the compensation correlates the comparative feature values to the set of condition values.

4. The method of claim 1, wherein the compensation modifies the set of feature values to a set of expected feature values.

5. The method of claim 4, wherein the expected feature values comprise a range of values.

6. The method of claim 1, wherein generating one or more identification values further includes establishing a confidence in the accuracy of one or more of the feature values in view of one or more of the condition values and weighting the importance of the comparison based on the confidence.

7. The method of claim 1, wherein the observation-related data includes three dimensional LADAR pixel data.

8. The method of claim 1, wherein the condition values include at least one of an orientation value, a depression value, a resolution value, and an obscuration value.

9. A method for identifying an object in multi-dimensional imagery data comprising:

(a) providing multi-dimensional imagery data, including;

(1) a feature value of an object,

(2) a condition value, and

(3) a feature membership function for a potential identification; and

(b) comparing the feature value to the feature membership function to identify the object, including compensating for the condition value by modifying either or both the feature value and the feature membership function.

10. The method of claim 9, wherein the compensation includes establishing comparative feature values in view of one or more condition values and modifying the feature membership function based on these comparative feature values.

11. The method of claim 9, wherein the compensation includes establishing a confidence in the accuracy of the feature value in view of one or more condition values and weighting the importance of the comparison based on the confidence.

12. The method of claim 9, wherein the compensation includes executing a fuzzy logic process utilizing the modified feature membership function.

13. The method of claim 9, wherein the multi-dimensional imagery data includes three dimensional LADAR pixel data.

14. A method for identifying an object in multi-dimensional imagery data comprising:

(a) providing multi-dimensional imagery data, including;

(1) a feature value of an object,

(2) a condition value, and

(3) feature membership parameters for a potential identification; and

(b) comparing the feature value to the feature membership function to identify the object, including selecting the feature membership parameters based on the condition value.

15. The method of claim 14, wherein comparing the feature value to the feature membership function to identify the object further includes weighting the importance of the comparison based on the condition value.

16. The method of claim 14, wherein comparing the feature value to the feature membership function to identify the object further includes executing a fuzzy logic process utilizing the feature membership parameters.

17. The method of claim 14, wherein the multi-dimensional imagery data includes three dimensional LADAR pixel data.

18. A method employing fuzzy logic for identifying an object in LADAR data, the method comprising:

(a) providing;

(1) feature value of the object,

(2) a condition value, and

(3) a membership function for a corresponding feature of a potential identification;

(b) comparing the feature value to the membership function using a fuzzy logic process, the comparison including;

(1) weighting at least one rule in a rulebase according to the condition value; and

(2) modifying the membership function responsive to the condition value;

(c) executing the fuzzy logic process to obtain an indication of whether the object corresponds to the potential identification; and

(d) identifying the object.

19. The method of claim 18, wherein the condition value is one of an object orientation value, a depression value, a resolution value, and an obscuration value.

20. A method for identifying an object in observation-related data of an object in a surrounding environment comprising:

providing;

a set of comparative feature values associated with at least one known object in at least one specified observation condition, and

a set of observation-related values including;

a set of observed feature values for at least one feature of an observed object in an environment, and

a set of condition values for at least one condition of the observation; and

modifying a fuzzy logic membership function for at least one of the comparative feature values responsive to at least one condition value.

21. The method of claim 20, wherein the set of condition values includes at least one of an orientation value, a resolution value, and an obscuration value.

22. The method of claim 20, wherein the comparative feature values comprise a range of values.

23. The method of claim 20, further comprising generating at least one identification value for the observed object from the observed feature values, the observed feature values including the modified observed feature value.

24. The method of claim 23, wherein generating the identification value includes a comparing the set of observed feature values to the comparative feature value.

25. The method of claim 23, wherein generating the identification value includes executing a fuzzy logic process.

26. The method of claim 25, further comprising shifting the membership function for a corresponding one of the comparative feature values.

27. The method of claim 23, wherein generating the identification value includes:

establishing a confidence in the accuracy of at least one of the observed feature values in view of at least one of the condition values; and

weighting the importance of the comparison based on the confidence.

28. The method of claim 20, wherein the observation-related data includes three dimensional LADAR pixel data.

29. A method for identifying an observed object in multi-dimensional imagery data comprising:

(a) providing multi-dimensional imagery data, including;

(1) an observed feature value of an object;

(2) a condition value; and

(3) a feature membership function for a potential identification;

(b) compensating for the condition value by modifying the feature membership function; and

(c) comparing the observed feature value to the modified feature membership function to identify the observed object.

30. The method of claim 29, wherein the compensation includes:

(i) establishing a plurality of comparative feature values in view of at least one condition value; and

(ii) modifying the feature membership function based on the comparative feature values.

31. The method of claim 29, wherein the compensation includes:

(i) establishing a confidence in the accuracy of the observed feature value in view of at least one condition value; and

(ii) weighting the comparison based on the confidence.

32. The method of claim 29, wherein the compensation includes executing a process utilizing the modified feature membership function.

33. The method of claim 29, wherein the multi-dimensional imagery data includes three dimensional LADAR pixel data.

34. A program storage medium encoded with instructions that, when executed by a computer, perform a method for identifying an object in observation-related data of an object in a surrounding environment comprising:

(a) providing;

(1) a set of comparative feature values associated with one or more known objects in or more specified observation conditions; and

(2) a set of observation-related values including;

(i) a set of feature values for one or more features of an object as observed in an environment; and

(ii) a set of condition values for one or more conditions of the observation, including one or more of an orientation value, a depression value, a resolution value, and an obscuration value; and

(b) generating one or more identification values for the object by comparing the set of feature values to the set of comparative feature values, including compensating for the set of condition values.

35. The program storage medium of claim 34, wherein compensating for the set of condition values includes modifying at least one feature value to an expected value to compensate for the obscuration value.

36. The program storage medium of claim 34, wherein the compensation correlates the comparative feature values to the set of condition values.

37. The program storage medium of claim 34, wherein the compensation modifies the set of feature values to a set of expected feature values.

38. The program storage medium of claim 37, wherein the expected feature values comprise a range of values.

39. The program storage medium of claim 34, wherein generating one or more identification values further includes establishing a confidence in the accuracy of one or more of the feature values in view of one or more of the condition values and weighting the importance of the comparison based on the confidence.

40. The program storage medium of claim 34, wherein the observation-related data includes three dimensional LADAR pixel data.

41. A program storage medium encoded with instructions that, when executed by a computer, perform a method for identifying an object in multi-dimensional imagery data comprising:

(a) providing multi-dimensional imagery data, including;

(1) a feature value of an object,

(2) a condition value, and

(3) a feature membership function for a potential identification; and

(b) comparing the feature value to the feature membership function to identify the object, including compensating for the condition value by modifying either or both the feature value and the feature membership function.

42. The program storage medium of claim 41, wherein the compensation includes establishing comparative feature value in view of one or more condition values and modifying the feature membership function based on these comparative feature values.

43. The program storage medium of claim 41, wherein the compensation includes establishing a confidence in the accuracy of the feature value in view of one or more condition values and weighting the importance of the comparison based on the confidence.

44. The program storage medium of claim 41, wherein the compensaton includes executing a fuzzy logic process utilizing the modified feature membership function.

45. The program storage medium of claim 41, wherein the multi-dimensional imagery data includes three dimensional LADAR pixel data.

46. A program storage medium encoded with instructions that, when executed by a computer, perform a method for identifying an object in multi-dimensional imagery data comprising:

(a) providing multi-dimensional imagery data, including;

(1) a feature value of an object,

(2) a condition value, and

(3) feature membership parameters for a potential identification; and

(b) comparing the feature value to the feature membership function to identify the object, including selecting the feature membership parameters based on the condition value.

47. The program storage medium of claim 46, wherein comparing the feature value to the feature membership function to identify the object further includes weighting the importance of the comparison based on the condition value.

48. The program storage medium of claim 46, wherein comparing the feature value to the feature membership function to identify the object further includes executing a fuzzy logic process utilizing the feature membership parameters.

49. The program storage medium of claim 46, wherein the multi-dimensional imagery data includes three dimensional LADAR pixel data.

50. A program storage medium encoded with instructions that, when executed by a computer, perform a method employing fuzzy logic for identifying an object in LADAR data, the method comprising:

(a) providing;

(1) feature value of the object,

(2) a condition value including one or more of object orientation value, depression value, resolution value, and obscuration value, and

(3) a membership function for a corresponding feature of a potential identification;

(b) comparing the feature value to the membership function using a fuzzy logic process including a rulebase, the comparison including;

(1) weighting at least one rule in the rulebase according to the condition value; and

(2) modifying the membership function responsive to the condition value;

(c) executing the fuzzy logic process to obtain an indication of whether the object corresponds to the potential identification; and

(d) identifying the object.

51. A program storage medium encoded with instructions that, when executed by a computer, perform a method for identifying an object in observation-related data of an object in a surrounding environment comprising:

providing;

a set of comparative feature values associated with at least one known object in at least one specified observation condition, and

a set of observation-related values including;

a set of observed feature values for at least one feature of an observed object in an environment, and

a set of condition values for at least one condition of the observation; and

modifying a fuzzy iogic membership function for at least one of the comparative feature values responsive to at least one condition value.

52. The program storage medium of claim 51, wherein the set of condition values includes at least one of an orientation value, a resolution value, and an obscuration value.

53. The program storage medium of claim 51, wherein the comparative feature values comprise a range of values.

54. The program storage medium of claim 51, further comprising generating at least one identification value for the observed object from the observed feature values, the observed feature values including the modified observed feature value.

55. The program storage medium of claim 54, wherein generating the identification value includes a comparing the set of observed feature values to the comparative feature value.

56. The program storage medium of claim 54, wherein generating the identification value includes executing a fuzzy logic process.

57. The program storage medium of claim 56, further comprising shifting the membership function for a corresponding one of the comparative feature values.

58. The program storage medium or claim 54, wherein generating the identification value includes:

establishing a confidence in the accuracy of at least one of the observed feature values in view of at least one of the condition values; and

weighting the importance of the comparison based on the confidence.

59. The program storage medium of claim 51, wherein the observation-related data includes three dimensional LADAR pixel data.

60. A program storage medium encoded with instructions that, when executed by a computer, perform a method for identifying an observed object in multi-dimensional imagery data comprising:

(a) providing multi-dimensional imagery data, including;

(1) an observed feature value of an object;

(2) a condition value; and

(3) a feature membership function for a potential identification;

(b) compensating for the condition value by modifying the feature membership function; and

(c) comparing the observed feature value to the modified feature membership function to identify the observed object.

61. The program storage medium of claim 60, wherein the compensation includes:

(i) establishing a plurality of comparative feature values in view of at least one condition value; and

(ii) modifying the feature membership function based on the comparative feature values.

62. The program storage medium of claim 60, wherein the compensation includes:

(i) establishing a confidence in the accuracy of the observed feature value in view of at least one condition value; and

(ii) weighting the comparison based on the confidence.

63. The program storage medium of claim 60, wherein the compensation includes executing a process utilizing the modified feature membership function.

64. The program storage medium of claim 60, wherein the multi-dimensional imagery data includes three dimensional LADAR pixel data.