Method and system for hash string extraction from biometric data

• US 7,233,686 B2
• Filed: 12/21/2005
• Issued: 06/19/2007
• Est. Priority Date: 07/25/2001
• Status: Expired due to Term

• Current Assignee: Activcard Ireland, Limited
• Original Assignee: Activcard Ireland, Limited
• Inventors: Hamid, Larry
• Primary Examiner(s): Mehta; Bhavesh M.
• Assistant Examiner(s): Strege; John B.
• Application Number:
  US11/315,446
• Publication Number:
  US 20060098850A1
• Time in Patent Office: 545 Days
• Field of Search: 382/124, 382/125, 340/5.53, 340/5.83, 356/71, 713/186
• US Class Current: 382/124
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction
• CPC Class Codes: H04L 9/3231 Biological data, e.g. fingerprint, voice or retina network architectures or network communication protocols for supporting authentication of entities using biometrical features in a packet data network H04L63/0861
• CPC Class Codes: H04L 9/3236 using cryptographic hash functions

Abstract

A method of hash string extraction from biometric information is disclosed. The method comprises the steps of providing a biometric information sample in the form of a fingerprint for example, extracting features from the biometric information sample and encoding the features based on their location within the biometric information sample; and, generating a string of values based on the extracted features and their determined locations. The method further comprises the steps of hashing the string of symbols to produce a plurality of hash values for comparing the plurality of hash values against a stored hash value for identifying a user.

Claims

1. A method of hash string extraction from biometric information comprising the steps of:

providing a biometric information sample;

extracting features from the biometric information sample and encoding symbols indicative of characteristics of the features;

generating a string of the symbols indicative of the extracted features;

determining a plurality of error strings of symbols each representing a set of error features within a predetermined error range from the extracted features; and

hashing strings from the plurality of error strings of symbols to produce a hash value for comparison with a stored hash value.

2. A method of string extraction from biometric information according to claim 1 wherein the step of extracting features from the biometric information sample comprises the step of identifying a feature type of an extracted feature and encoding the feature as a symbol based on its type.

3. A method of string extraction from biometric information according to claim 2, wherein a symbol is encoded as an n-bit value.

4. A method of string extraction from biometric information according to claim 2 wherein the step of extracting features from the biometric information sample comprises the step of identifying a feature location of an extracted feature and encoding the feature as a symbol based on its location.

5. A method of string extraction from biometric information according to claim 2 wherein the step of extracting features from the biometric information sample comprises the step of identifying a feature location of an extracted feature and encoding the feature as a symbol based on its orientation.

6. A method of string extraction from biometric information according to claim 4 wherein the step of encoding the features based on their location within the biometric information sample comprises the steps of:

providing a plurality of sectors within the biometric information sample, the sectors determined in dependence upon an identifiable and common feature within the biometric information;

determining a sector wherein a feature is located; and

,

encoding a value indicative of the determined sector for the extracted feature location.

7. A method of string extraction from biometric information according to claim 6 wherein the biometric information is a fingerprint and wherein the common feature is a core of the fingerprint.

8. A method of string extraction from biometric information according to claim 1 wherein the step of generating a string of values results in an unordered string.

9. A method of string extraction from biometric information according to claim 7 wherein the step of encoding the features based on their location comprises the steps of

identifying a feature representative of the core within the biometric information sample;

orienting the sectors relative to the identified core; and

encoding a symbol indicative of a location of a feature other than the core relative to the core.

10. A method of string extraction from biometric information according to claim 1 further comprising the step of:

comparing the string against a previously stored string to determine a comparison value.

11. A method of string extraction from biometric information according to claim 10 further comprising the steps of:

generating the previously stored string by the steps of;

providing a number N of biometric information samples from a same biometric information source;

extracting a number r of features from the biometric information sample and encoding r symbols, one per feature; and

,

extracting a number δ

of extra features, and encoding δ

extra symbols, one per extra feature, and

wherein the step of comparing the string includes the step of determining a number of symbols within the string that are absent from the previously stored string and deleting those symbols.

12. A method of string extraction from biometric information according to claim 4 wherein the step of encoding the features based on their location within a sector of a plurality of sectors of the biometric information sample further comprises the steps of:

providing a number of quantization errors tolerated; and

providing a number of sector re-quantization possibilities representative of a tolerance in defining quantization of a sector.

13. A method of string extraction from biometric information according to claim 12 wherein the tolerance is relative to a small rotation of the features relative to the sectors.

14. A method of string extraction from biometric information according to claim 12 wherein the tolerance is relative to a small translation of the features relative to the sectors.

15. A method of string extraction from biometric information according to claim 12 wherein the tolerance is relative to a small diagonal translation of the features relative to the sectors.

16. A method of string extraction from biometric information according to claim 1 comprising the additional steps of:

comparing the hashed strings from the plurality of error strings against the stored hashed value; and

,

when a hashed string and the stored hashed value are a same value, providing the error string as a password.

17. A method of string extraction from biometric information for authorizing access comprising the steps of:

providing a biometric information sample to a contact imager for imaging the biometric information sample wherein the contact imager comprises a platen disposed for accepting a biometric information source thereon;

processing the imaged information by extracting features therefrom and encoding symbols based on at least an extracted feature location within the biometric sample;

determining a plurality of error strings of symbols each representing a set of error features within a predetermined error range from the extracted features;

hashing strings from the plurality of error strings of symbols to produce a hash value for comparison with a stored hash value; and

comparing the hash value determined from the error strings of symbols with the stored hash value to determine a match, wherein upon a match access is provided.

18. A method of string extraction from biometric information for authorizing access as defined in claim 17 wherein the step of comparing a string of symbols generated from an enrollment sample and a string of symbols generated from the imaged information of the biometric sample comprises the step of comparing symbols within predetermined limits wherein symbols need not match exactly.

19. A method of string extraction from biometric information for authorizing access as defined in claim 18 wherein the step of comparing symbols within predetermined limits includes the step of masking the symbols.

20. A method of string extraction from biometric information for authorizing access as defined in claim 18 wherein the step of comparing symbols within predetermined limits includes the step of comparing symbols indicative of feature locations wherein variations in symbol locations that are consistent across all symbols are not indicative of other than a match.

21. A method of string extraction from biometric information according to claim 17 wherein the step of comparing a string of symbols generated from an enrolment sample and a string of symbols generated from the imaged information of the biometric sample comprises the step of comparing symbols relating to features wherein the symbols must be indicative of a same feature type to be considered a match.

22. A method of string extraction from biometric information according to claim 17 wherein the step of extracting features therefrom and encoding symbols based on at least an extracted feature location comprises the steps of:

determining a sector wherein the extracted feature is located, the sectors disposed relative to a known feature location; and

,

encoding the sector indicative of an extracted feature location.

23. A system that provides hash string extraction from biometric information, comprising:

a contact imager that images a biometric information sample to provide biometric image data; and

a processor that processes the biometric image data to extract features therefrom, to encode symbols indicative of characteristics of the features, to generate a string of the symbols indicative of the extracted features, to determine a plurality of error strings of symbols each representing a set of error features within a predetermined error range from the extracted features, and to hash strings from the plurality of error strings of symbols to produce a hash value for comparison with a stored hash value.

24. A system according to claim 23 wherein the processor farther identifies a feature type of an extracted feature and encodes the feature as a symbol based on its type.

25. A system according to claim 24 wherein the processor identifies a feature location of an extracted feature and encodes the feature as a symbol based on its location.

26. A system according to claim 24 wherein the processor identifies a feature location of an extracted feature and encodes the feature as a symbol based on its orientation.

27. A system according to claim 25 wherein the processor is programmed to encode the features based on their location within the biometric information sample by performing the steps of:

providing a plurality of sectors within the biometric information sample, the sectors determined in dependence upon an identifiable and common feature within the biometric information;

determining a sector wherein a feature is located; and

encoding a value indicative of the determined sector for the extracted feature location.

28. A system according to claim 27 wherein the biometric information is a fingerprint and wherein the common feature is a core of the fingerprint.

29. A system according to claim 23 wherein the processor generates the hash value as an unordered string.

30. A system according to claim 28 wherein the processor is programmed to encode the features based on their location by performing the steps of:

identifying a feature representative of the core within the biometric information sample;

orienting the sectors relative to the identified core; and

encoding a symbol indicative of a location of a feature other than the core relative to the core.

31. A system according to claim 23 wherein the processor is further programmed to perform the steps of:

generating a stored error string by providing a number N of biometric information samples from a same biometric information source, extracting a number r of features from the biometric information sample and encoding r symbols, one per feature, extracting a number δ

of extra features, and encoding δ

extra symbols, one per extra feature, and

comparing an error string with the stored error string by determining a number of symbols within the error string that are absent from the stored error string and deleting those symbols.

32. A system according to claim 25 wherein the processor is programmed to encode the features based on their location within a sector of a plurality of sectors of the biometric information sample by providing a number of quantization errors tolerated and providing a number of sector re-quantization possibilities representative of a tolerance in defining quantization of a sector.

33. A system according to claim 32 wherein the tolerance is relative to a small rotation of the features relative to the sectors.

34. A system according to claim 32 wherein the tolerance is relative to a small translation of the features relative to the sectors.

35. A system according to claim 32 wherein the tolerance is relative to a small diagonal translation of the features relative to the sectors.

36. A system according to claim 23 wherein the processor further compares the hashed strings from the plurality of error strings against the stored hashed value and, when a hashed string and the stored hashed value are a same value, provides the error string as a password.

37. A system that provides string extraction from biometric information for authorizing access, comprising:

a contact imager having a platen disposed for accepting a biometric information source thereon for imaging the biometric information sample to provide biometric image data; and

a processor that processes the biometric image data to extract features therefrom, to encode symbols based on at least an extracted feature location within the biometric sample, to determine a plurality of error strings of symbols each representing a set of error features within a predetermined error range from the extracted features, to hash strings from the plurality of error strings of symbols to produce a hash value, and to compare the hash value determined from the error strings of symbols with a stored hash value to determine a match, wherein upon a match access is provided.

38. A system according to claim 37 wherein the processor compares a string of symbols generated from an enrollment sample and a string of symbols generated from the imaged information of the biometric sample within predetermined limits wherein symbols need not match exactly.

39. A system according to claim 38 wherein the processor compares the symbols within predetermined limits by masking the symbols.

40. A system according to claim 38 wherein the processor compares the symbols within predetermined limits by comparing symbols indicative of feature locations wherein variations in symbol locations that are consistent across all symbols are not indicative of other than a match.

41. A system according to claim 37 wherein the processor compares a string of symbols generated from an enrollment sample and a string of symbols generated from the biometric image data by comparing symbols relating to features wherein the symbols must be indicative of a same feature type to be considered a match.

42. A system according to claim 37 wherein the processor is programmed to extract features and to encode symbols based on at least an extracted feature location by determining a sector wherein the extracted feature is located, the sectors disposed relative to a known feature location and encoding the sector indicative of an extracted feature location.

43. A computer readable medium having computer executable instructions stored thereon that when processed by a processor causes the processor to provide hash string extraction from biometric information, the computer executable instructions comprising:

computer executable instructions that image a biometric information sample to provide biometric image data; and

computer executable instructions that process the biometric image data to extract features therefrom, encode symbols indicative of characteristics of the features, generate a string of the symbols indicative of the extracted features, determine a plurality of error strings of symbols each representing a set of error features within a predetermined error range from the extracted features, and hash strings from the plurality of error strings of symbols to produce a hash value for comparison with a stored hash value.

44. A computer readable medium according to claim 43 wherein the executable instructions further identify a feature type of an extracted feature and encode the feature as a symbol based on its type.

45. A computer readable medium according to claim 44 wherein the executable instructions further identify a feature location of an extracted feature and encode the feature as a symbol based on its location.

46. A computer readable medium according to claim 44 wherein the executable instructions further identify a feature location of an extracted feature and encode the feature as a symbol based on its orientation.

47. A computer readable medium according to claim 45 wherein the executable instructions further encode the features based on their location within the biometric information sample by causing the processor to perform the steps of:

providing a plurality of sectors within the biometric information sample, the sectors determined in dependence upon an identifiable and common feature within the biometric information;

determining a sector wherein a feature is located; and

encoding a value indicative of the determined sector for the extracted feature location.

48. A computer readable medium according to claim 47 wherein the biometric information is a fingerprint and wherein the common feature is a core of the fingerprint.

49. A computer readable medium according to claim 43 wherein the executable instructions generate the hash value as an unordered string.

50. A computer readable medium according to claim 48 wherein the executable instructions farther encode the features based on their location by causing the processor to perform the steps of:

identifying a feature representative of the core within the biometric information sample;

orienting the sectors relative to the identified core; and

encoding a symbol indicative of a location of a feature other than the core relative to the core.

51. A computer readable medium according to claim 43 wherein the executable instructions further cause the processor to perform the steps of:

generating a stored error string by providing a number N of biometric information samples from a same biometric information source, extracting a number r of features from the biometric information sample and encoding r symbols, one per feature, extracting a number δ

of extra features, and encoding δ

extra symbols, one per extra feature, and

comparing an error string with the stored error string by determining a number of symbols within the error string that are absent from the stored error string and deleting those symbols.

52. A computer readable medium according to claim 45 wherein the executable instructions encode the features based on their location within a sector of a plurality of sectors of the biometric information sample by providing a number of quantization errors tolerated and providing a number of sector re-quantization possibilities representative of a tolerance in defining quantization of a sector.

53. A computer readable medium according to claim 52 wherein the tolerance is relative to a small rotation of the features relative to the sectors.

54. A computer readable medium according to claim 52 wherein the tolerance is relative to a small translation of the features relative to the sectors.

55. A computer readable medium according to claim 52 wherein the tolerance is relative to a small diagonal translation of the features relative to the sectors.

56. A computer readable medium according to claim 43 wherein the executable instructions further compare the hashed strings from the plurality of error strings against the stored hashed value and, when a hashed string and the stored hashed value are a same value, provide the error string as a password.

57. A computer readable medium having computer executable instructions stored thereon that when processed by a processor causes the processor to provide hash string extraction from biometric information, the computer executable instructions comprising:

computer executable instructions that image a biometric information sample to provide biometric image data; and

computer executable instructions that process the biometric image data to extract features therefrom, encode symbols based on at least an extracted feature location within the biometric sample, determine a plurality of error strings of symbols each representing a set of error features within a predetermined error range from the extracted features, hash strings from the plurality of error strings of symbols to produce a hash value, and compare the hash value determined from the error strings of symbols with a stored hash value to determine a match, wherein upon a match access is provided.

58. A computer readable medium according to claim 57 wherein the executable instructions compare a string of symbols generated from an enrollment sample and a string of symbols generated from the imaged information of the biometric sample within predetermined limits wherein symbols need not match exactly.

59. A computer readable medium according to claim 58 wherein the executable instructions compare the symbols within predetermined limits by masking the symbols.

60. A computer readable medium according to claim 58 wherein the executable instructions compare the symbols within predetermined limits by comparing symbols indicative of feature locations wherein variations in symbol locations that are consistent across all symbols are not indicative of other than a match.

61. A computer readable medium according to claim 57 wherein the executable instructions compare a string of symbols generated from an enrollment sample and a string of symbols generated from the biometric image data by comparing symbols relating to features wherein the symbols must be indicative of a same feature type to be considered a match.

62. A computer readable medium according to claim 57 wherein the executable instructions extract features and encode symbols based on at least an extracted feature location by determining a sector wherein the extracted feature is located, the sectors disposed relative to a known feature location and encoding the sector indicative of an extracted feature location.