Biometric print quality assurance

• US 20060117188A1
• Filed: 11/17/2005
• Published: 06/01/2006
• Est. Priority Date: 11/18/2004
• Status: Active Grant

• Current Assignee: Biogy, Inc.
• Original Assignee: Bionopoly LLC
• Inventors: Fiske, Michael Stephen, Louis, Jon
• Granted Patent: US 7,565,548 B2
• Time in Patent Office: Days
• US Class Current: 713/186
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G07C 9/37 using biometric data, e.g. fingerprints, iris scans or voice recognition

Abstract

Quality of biometric prints is enhanced by any one of several different methods. In one embodiment, if a biometric print does not have a high enough quality, the biometric print is discarded. In another embodiment, a matching score is associated with the degree to which the two biometric prints match. In another embodiment, a ranking is associated with a biometric print, and the ranking is determined based upon two or more matching scores associated with the biometric print. In another embodiment, a variety of distinguishing features are derived from a given biometric print and are stored for authenticating a biometric print of a user requesting access to a secure entity. In an embodiment, a transformation is applied to at least a portion of a newly acquired biometric print to improve the match of the newly acquired biometric print to a stored biometric print.

Claims

1. A machine-implemented method comprising:

determining a value associated with how many unique aspects are discernable for at least one biometric print that was obtained during a current usage; and

determining whether to accept or reject the at least one biometric print based on a comparison of the value to a predetermined threshold.

2. The method of claim 1, further comprising enrolling a user by storing an accepted biometric print for later comparison.

3. The method of claim 1, further comprising determining whether to grant access based on a comparison of an accepted biometric print to a stored biometric print.

29. A machine-readable medium storing one or more instructions, which when implemented, cause a processor to execute the method of claim 1.

30. A machine-readable medium storing one or more instructions, which when implemented, cause a processor to execute the method of claim 2.

31. A machine-readable medium storing one or more instructions, which when implemented, cause a processor to execute the method of claim 3.

4. A machine-implemented method comprising:

determining a value associated with a match between two or more biometric prints obtained during a current session; and

determining whether to accept or reject at least one of the two or more biometric prints based on a comparison of the value with a predetermined threshold.

5. The method of claim 4, further comprising enrolling a user by at least storing an accepted biometric print for later comparison.

6. The method of claim 4, further comprising determining whether to grant access based on a comparison of an accepted biometric print to a stored biometric print.

32. A machine-readable medium storing one or more instructions, which when implemented, cause a processor to execute the method of claim 4.

33. A machine-readable medium storing one or more instructions, which when implemented, cause a processor to execute the method of claim 5.

34. A machine-readable medium storing one or more instructions, which when implemented, cause a processor to execute the method of claim 6.

7. A machine-implemented method comprising:

comparing a plurality of biometric prints to one another; and

determining one or more unique aspects associated with the biometric prints to be false based on the comparing.

35. A machine-readable medium storing one or more instructions, which when implemented, cause a processor to execute the method of claim 7.

8. A machine-implemented method comprising:

identifying distinguishing characteristics associated with a biometric print; and

at least temporarily storing the distinguishing characteristics as a representation of the biometric print.

9. The machine-implemented method of claim 8, wherein the biometric print is a fingerprint.

10. The machine-implemented method of claim 8, wherein the distinguishing characteristics include at least a sweat pore.

11. The machine-implemented method of claim 8, wherein the distinguishing characteristics include at least a termination of a ridge.

12. The machine-implemented method of claim 8, wherein the distinguishing characteristics include at least a bifurcation of a ridge.

13. The machine-implemented method of claim 8, wherein the distinguishing characteristics include at least a ridge that forms a loop that is at least partially encircled within one or more other loops.

14. The machine-implemented method of claim 8, wherein the distinguishing characteristics include at least a cut including at least a break in one or more ridges.

15. The machine-implemented method of claim 8, wherein the distinguishing characteristics include at least a first distinguishing characteristic, which is at least a count of how many ridges are located between a second distinguishing characteristic and a third distinguishing characteristic.

16. The machine-implemented method of claim 8, further comprising

acquiring a new biometric print prior to the identifying;

and

discarding the new biometric print if the biometric print is not of a quality that is at least as great as a threshold quality.

17. The machine-implemented method of claim 16, wherein the quality is determined at least in-part by how many distinguishing characteristics were identified.

18. The machine implemented method of claim 8, further comprising:

acquiring multiple biometric prints from a user during a registration process;

comparing each of the multiple biometric prints to one another; and

discarding one or more of the plurality of the biometric prints based on the comparing.

19. The machine-implemented method of claim 8, wherein the biometric print is a first biometric print and the distinguishing characteristics are a first set of distinguishing characteristics, and the method further comprises:

identifying a second set of distinguishing characteristics associated with a second biometric print;

at least one distinguishing characteristic being found in only one of the first set and the second set; and

removing the at least one distinguishing characteristic from the only one of the first set and the second set in which the at least one distinguishing characteristic was found, based on the at least one distinguishing characteristic not being found in another of the first set and the second set prior to the removing.

20. The machine-implemented method of claim 8, wherein the at least temporarily storing includes at least storing the biometric print in long-term storage, wherein the biometric print is available as a basis for authenticating a user requesting access to a secure entity.

21. The machine-implemented method of claim 8, further comprising:

determining whether the distinguishing characteristics match distinguishing characteristics stored in long-term memory.

22. The machine implemented method of claim 8, wherein

the biometric print is a first fingerprint;

the identifying includes at least

searching for sweat pores,

searching for bifurcations of ridges,

searching for terminations of ridges,

searching for a core including at least a ridge that forms a loop that is at least partially encircled within one or more other loops, and

searching for a cut including at least a break in one or more ridges.

23. The machine implemented method of claim 8, wherein

the biometric print is a first fingerprint;

the identifying includes at least

searching for bifurcations of ridges, and

searching for terminations of ridges; and

the method further comprising;

(a) collecting a plurality of fingerprints, which includes at least collecting the first fingerprint prior to the identifying;

(b) performing the identifying and the storing for each of the plurality of fingerprints;

(c) for each of the plurality of fingerprints, discarding distinguishing characteristics not found in at least one other of the plurality of fingerprints;

(d) for each of the plurality of fingerprints, determining whether the distinguishing characteristics are fewer in number than a threshold number;

(e) discarding those of the plurality of fingerprints having fewer distinguishing characteristics than the threshold number;

(f) computing matching scores for each of a plurality of pairs of fingerprints, wherein

each fingerprint of each of the plurality of pairs is one of the plurality of fingerprints,

each fingerprint of the each of the plurality of pairs is a member of more than one pair, and

each matching score is a value based upon a degree to which distinguishing characteristics of one fingerprint of the pair matched distinguishing characteristics of another fingerprint of the pair;

(g) computing at least one transformation including at least computing a rotation and computing a translation, wherein

the transformation computed is at least in part based on whether an improvement in matching indicated by the matching score results from the transformation

a value of the translation in any direction could be positive, negative, or zero,

a value of the rotation could be positive, negative, or zero, and

the value of each matching score that is retained is a value resulting after the transformation is finished being computed;

(h) for each of the plurality of fingerprints not discarded computing a ranking based on each matching score associated with that fingerprint; and

(i) discarding one or more of the plurality of fingerprints having rankings indicating poorer matches with others of the plurality of fingerprints than those of the plurality of fingerprints not discarded.

24. The method of claim 23, wherein remaining fingerprints from the plurality of fingerprints will be referred to as registration fingerprints, the method further comprising:

storing the registration fingerprints in long-term memory;

as part of a request for access to a secure entity, acquiring a second plurality of fingerprints;

repeating the identifying;

repeating (a)-(i), wherein remaining fingerprints from the second plurality of fingerprints will be referred to as newly acquired fingerprints;

computing at least one matching score, which will be referred to as at least one access matching score, wherein the at least one access matching score is computed for a pair of fingerprints in which one fingerprint of the pair is one the registration fingerprint and another fingerprint of the pair is one of the newly acquired fingerprints; and

determining whether to grant access to the secure entity based on whether the at least one access matching score at least meets a threshold value.

25. The method of claim 24, wherein the at least one access matching score includes a plurality of access matching scores and the method further comprises:

computing a ranking based on the plurality of access matching scores; and

determining whether to grant access based on whether the ranking was greater than a threshold.

26. A system comprising a processor configured to implement the method of claim 8.

27. The system of claim 26, further comprising a sensor for sensing the biometric prints.

28. The system of claim 27, further comprising one or more machine-readable media storing at least

the distinguishing characteristics;

and

one or more instructions, which when executed cause the processor to implement the method.

36. A machine-readable medium storing one or more instructions, which when implemented, cause a processor to execute the method of claim 8.

37. A machine-readable medium storing one or more instructions, which when implemented, cause a processor to execute the method of claim 9.

38. A machine-readable medium storing one or more instructions, which when implemented, cause a processor to execute the method of claim 10.

39. A method comprising:

acquiring a representation of a biometric print;

applying a first one-way function to the representation of the biometric print, therein forming an encryption of the biometric print having a first length;

storing the encryption as a registration code;

applying a second one-way function to a portion of the registration code, therein forming a passcode of a second length that is different from the first length; and

sending a request for access to a secure entity, wherein the passcode is sent to the secure entity in conjunction with the request.

40. The method of claim 39, wherein the first one-way function may be a function that is used as the second one-way function.