Apparatus and method of biometric identification or verification of individuals using optical spectroscopy

• US 6,560,352 B2
• Filed: 04/11/2001
• Issued: 05/06/2003
• Est. Priority Date: 10/08/1999
• Status: Expired due to Term

• Current Assignee: HID Global Corporation (Assa Abloy AB)
• Original Assignee: Lumidigm Incorporated (Assa Abloy AB)
• Inventors: Robinson, Mark Ries, Perella, Steve L., Landesman, Roxanne N., Rowe, Robert K.
• Primary Examiner(s): WERNER, BRIAN P
• Application Number:
  US09/832,534
• Publication Number:
  US 20020009213A1
• Time in Patent Office: 755 Days
• Field of Search: 382/115, 382/124, 382/125, 382/126, 382/127, 73/73, 250/559.44, 250/316.1, 250/339.07, 250/339.1, 250/341.8, 340/5.52, 340/5.82, 356/31, 356/303, 356/445, 356/939, 356/71, 902/3, 707/6, 713/186, 702/19, 702/76
• US Class Current: 382/115
• CPC Class Codes: A61B 5/0059 using light, e.g. diagnosis by transillumination, diascopy, fluorescence A61B5/0093, A61B5/1455, A61B5/02416 and A61B5/0261 take precedence
• CPC Class Codes: A61B 5/117 Identification of persons methods or arrangements for recognising patterns, e.g. fingerprints, G06F18/00, G06V40/00; identification of persons by analysing their voice or speech G10L17/00
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
• CPC Class Codes: G06V 40/12 Fingerprints or palmprints
• CPC Class Codes: G06V 40/40 Spoof detection, e.g. liveness detection
• CPC Class Codes: G07C 9/37 using biometric data, e.g. fingerprints, iris scans or voice recognition
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

Methods and apparatus for performing biometric identification or verification of identities using optical spectroscopy of tissue. Tissue optical spectra can be obtained by projecting optical radiation into skin and capturing the light transmitted or reflected back and out through the tissue. The tissue spectra collected preferably includes primarily light that has passed through skin tissue below the epidermis. Multiple tissue spectra and identities can be collected from one or more individuals. These tissue spectra can be analyzed on a computer, and the spectral features that are most important for classifying person-to-person differences can be established using principle component analysis, linear discriminant analysis, or a variety of other related techniques. One or more tissue spectra and identities can be collected from individuals for whom identification or verification may later be desired. A target individual seeking identification or verification can submit a suitable tissue site for spectroscopic measurement. In addition a target individual seeking identity verification can submit a purported identity using some means such as a typed user name, PIN code, magnetic card, transponder, etc. Similarity between the target spectrum and the spectrum or spectra in the enrolled spectral database with respect to the inter-person classification spectral features is determined and identification or verification is granted based on the degree of similarity.

Claims

1. A system for verifying the purported identity of a target individual comprising:

an enrollment database including sub-epidermal tissue optical spectra collected from at least one enrolled person, said enrolled person'"'"'s sub-epidermal tissue optical spectra having a plurality of measurement wavelengths obtained by applying a spectral band of optical wavelengths to said sub-epidermal tissue;

means for obtaining at least one sub-epidermal tissue optical spectra and purported identity from said target individual, said target individual'"'"'s sub-epidermal tissue optical spectra having a plurality of measurement wavelengths obtained by applying said spectral band of optical wavelengths to sub-epidermal tissue of said target individual;

means for comparing said target individual sub-epidermal tissue optical spectra and said enrolled person'"'"'s sub-epidermal tissue optical spectra by selecting for comparison a combination of individual wavelengths of said band emanating from said sub-epidermal tissue that are unique to said person and insensitive to environmental and instrumental effects, said enrolled person'"'"'s sub-epidermal tissue optical spectra corresponding to the purported identity of the target individual, said comparison providing a measure of the degree of similarity between said target individual sub-epidermal tissue optical spectra and said enrolled person'"'"'s sub-epidermal tissue optical spectra; and

means for positively verifying said target individual'"'"'s identity by confirming that said target individual'"'"'s measure of spectral similarity is at least as similar as an established threshold value.

2. The system as recited in claim 1, wherein said means for obtaining said target individual sub-epidermal tissue optical spectra includes a spectrometer.

3. The system as recited in claim 2, wherein said spectrometer is an FTIR spectrometer.

4. The system as recited in claim 2, wherein said spectrometer is a grating array spectrometer.

5. The system as recited in claim 1, wherein said optical spectra include near-infrared wavelengths.

6. The system as recited in claim 1, wherein said optical spectra include visible wavelengths.

7. The system as recited in claim 1, wherein said optical spectra include near-ultraviolet wavelengths.

8. The system as recited in claim 1, wherein said comparison and similarity determination utilizes a classification algorithm.

9. A system for identifying a target individual comprising:

an enrollment database including sub-epidermal tissue optical spectral data collected from one or more enrolled persons, said enrolled persons'"'"' sub-epidermal tissue optical spectra having a plurality of measurement wavelengths obtained by applying a spectral band of optical wavelengths to said sub-epidermal tissue;

means for obtaining at least one sub-epidermal tissue optical spectra from said target individual, wherein said means for obtaining said target individual sub-epidermal tissue optical spectra includes means for measuring optical radiation reflected from sub-epidermal tissue of said target individual, said target individual'"'"'s sub-epidermal tissue optical spectra having a plurality of measurement wavelengths obtained by applying said spectral band of optical wavelengths to sub-epidermal tissue of said target individual;

means for comparing said target individual sub-epidermal tissue optical spectra and said all enrolled persons'"'"' sub-epidermal tissue optical spectra by selecting for comparison a combination of individual wavelengths of said band emanating from said sub-epidermal tissue that are unique to each said person and insensitive to environmental and instrumental effects, said comparison providing a measure of the degree of similarity between said target individual'"'"'s sub-epidermal tissue optical spectra and each said enrolled person'"'"'s sub-epidermal tissue optical spectra; and

means for indicating identity as at least one of the said enrolled persons if the corresponding measure of degree of similarity is at least as similar as an established threshold value.

10. The system as recited in claim 9, wherein said means for obtaining said target individual'"'"'s sub-epidermal tissue optical spectra includes a spectrometer.

11. The system as recited in claim 10, wherein said spectrometer is an FTIR spectrometer.

12. The system as recited in claim 10, wherein said spectrometer is a grating array spectrometer.

13. The system as recited in claim 9, wherein said optical spectra include near-infrared wavelengths.

14. The system as recited in claim 9, wherein said optical spectra include visible wavelengths.

15. The system as recited in claim 9, wherein said optical spectra include near-ultraviolet wavelengths.

16. The system as recited in claim 9, wherein said comparison and similarity determination utilizes a classification algorithm.

17. A system for verifying the purported identity of a target individual comprising:

a computer including an input device and an output device;

an enrollment database including sub-epidermal tissue optical spectra for at least one enrolled person, said enrolled person'"'"'s sub-epidermal tissue optical spectra having a plurality of measurement wavelengths obtained by applying a band of optical wavelengths to said sub-epidermal tissue;

means for obtaining at least one sub-epidermal tissue optical spectrum from said target individual, including an optical radiation source, an optical sampler for projecting optical radiation into the sub-epidermal tissue and for collecting radiation that substantially passed through sub-epidermal sub-epidermal tissue, an optical spectrometer for measuring the sub-epidermal optical intensity over a plurality of wavelengths, wherein said target individual'"'"'s sub-epidermal tissue optical spectrum has a plurality of measurement wavelengths obtained by applying said spectral band of optical wavelengths to sub-epidermal tissue of said target individual;

means for obtaining said target individual'"'"'s purported identity; and

a program running in said computer for comparing said target individual sub-epidermal tissue optical spectrum and said enrolled person'"'"'s sub-epidermal tissue optical spectra corresponding to said target individual'"'"'s purported identity by selecting for comparison a combination of individual wavelengths of said band emanating from said sub-epidermal tissue that are unique to said person and insensitive to environmental and instrumental effects.

18. A system for identifying a target individual comprising:

a computer including an input device and an output device;

an enrollment database including sub-epidermal tissue optical spectra for at least one enrolled person, said enrolled persons'"'"' sub-epidermal tissue optical spectra having a plurality of measurement wavelengths obtained by applying a spectral band of optical wavelengths to said sub-epidermal tissue;

means for obtaining at least one sub-epidermal tissue optical spectrum from said target individual, including an optical radiation source, an optical sampler for projecting optical radiation into the sub-epidermal tissue and for collecting radiation that substantially passed through sub-epidermal sub-epidermal tissue, an optical spectrometer for measuring the sub-epidermal optical intensity over a plurality of wavelengths, wherein said target individual'"'"'s sub-epidermal tissue optical spectrum has a plurality of measurement wavelengths obtained by applying said spectral band of optical wavelengths to sub-epidermal tissue of said target individual; and

a program running in said computer for comparing said target individual sub-epidermal tissue optical spectrum and all said enrolled persons'"'"' sub-epidermal tissue optical spectra by selecting for comparison a combination of individual wavelengths of said band emanating from said sub-epidermal tissue that are unique to each said person and insensitive to environmental and instrumental effects.

19. A method for verifying the purported identity of a target individual utilizing an enrollment database including sub-epidermal tissue optical spectra collected from a number of enrolled individuals having known identities, said sub-epidermal tissue optical spectra having a plurality of measurement wavelengths obtained by applying a spectral band of optical wavelengths to said sub-epidermal tissue, comprising the steps of:

obtaining optical target sub-epidermal tissue spectral data from said target individual, said optical target sub-epidermal tissue spectral data having a number of measurement wavelengths obtained by applying said spectral band of optical wavelengths to sub-epidermal tissue of said target individual;

obtaining said purported identity from said target individual;

comparing said optical target sub-epidermal tissue spectra and an enrolled individual'"'"'s sub-epidermal tissue optical spectra by selecting for comparison a combination of individual wavelengths of said band emanating from said sub-epidermal tissue that are unique to said enrolled individual and insensitive to environmental and instrumental effects, said enrolled individual'"'"'s sub-epidermal tissue optical spectra corresponding to the purported identity of the target individual, said comparison providing a measure of the degree of similarity between said optical target sub-epidermal tissue spectra and said enrolled individual'"'"'s sub-epidermal tissue optical spectra; and

positively verifying said target individual'"'"'s identity by confirming that said target individual'"'"'s measure of spectral similarity is at least as similar as an established threshold value.

20. The method for verifying the identity of a target individual as recited in claim 19, wherein the method further includes a classification algorithm to perform said comparison between said target individual'"'"'s optical spectral data and said enrolled individual'"'"'s optical spectral data.

21. The method for verifying the identity of a target individual as recited in claim 20, wherein the method further includes classification features that are determined from a set of calibration optical spectral data collected on at least one individual measured more than one time.

22. The method for verifying the identity of a target individual as recited in claim 21, wherein said classification features are applied to the said comparison between the target optical spectral data and the enrollment spectral data to determine the similarity with respect to the said classification features.

23. The method for verifying the identity of a target individual as recited in claim 22, wherein said verification occurs when said comparison of said target optical spectral data and said enrollment spectral data using said classification features is at least as good a predetermined measure of similarity.

24. The method for identifying a target individual as recited in claim 19, further comprising an enrollment database with optical spectral data collected from a number of enrolled individuals, wherein said number is greater than one.

25. The method for identifying a target individual as recited in claim 19, further comprising an enrollment database with optical spectral data collected from a number of enrolled individuals, wherein said number is equal to one.

26. The method for identifying a target individual as recited in claim 19, wherein said target spectrum is added to said enrollment optical spectral data after said verification of identity.

27. The method for identifying a target individual as recited in claim 19, wherein said sub-epidermal tissue optical spectra include near-ultraviolet wavelengths.

28. The method for identifying a target individual as recited in claim 19, wherein said sub-epidermal tissue optical spectra include visible wavelengths.

29. The method for identifying a target individual as recited in claim 19, wherein said sub-epidermal tissue optical spectra include near-infrared wavelengths.

30. A method for identifying a target individual utilizing an enrollment database including sub-epidermal tissue optical spectra collected from a number of enrolled persons, said sub-epidermal tissue optical spectra having a plurality of measurement wavelengths obtained by applying a spectral band of optical wavelengths to said sub-epidermal tissue, comprising the steps of:

obtaining optical target sub-epidermal tissue spectral data from said target individual, said optical target sub-epidermal tissue spectral data having a number of measurement wavelengths and including a substantial spectral contribution from sub-epidermal tissue, wherein said measurement wavelengths are obtained by applying said spectral band of optical wavelengths to sub-epidermal tissue of said target individual;

comparing said optical target sub-epidermal tissue spectral data and said enrolled persons'"'"' sub-epidermal tissue optical spectra by selecting for comparison a combination of individual wavelengths of said band emanating from said sub-epidermal tissue that are unique to each said person and insensitive to environmental and instrumental effects, said comparison providing a measure of the degree of similarity between said optical target sub-epidermal tissue spectral data and each of said enrolled person'"'"'s sub-epidermal tissue optical spectra; and

positively establishing said target individual'"'"'s identity by confirming that said target individual'"'"'s measure of spectral similarity is at least as similar to one of the enrolled person'"'"'s optical spectral data as an established threshold value.

31. The method for identifying a target individual as recited in claim 30, wherein the method further includes a classification algorithm to perform said comparison between said target individual'"'"'s optical spectral data and said enrolled persons'"'"' optical spectral data.

32. The method for identifying a target individual as recited in claim 31, wherein the method further includes classification features that are determined from a set of calibration optical spectral data collected on at least one individual measured more than one time.

33. The method for identifying a target individual as recited in claim 32, wherein said classification features are applied to the said comparison between the target optical spectral data and the enrollment spectral data to determine the similarity with respect to the said classification features.

34. The method for identifying a target individual as recited in claim 33, wherein said identification occurs when said comparison of said target optical spectral data and said enrollment spectral data using said classification features is at least as similar as a predetermined measure of similarity for a number of enrolled persons'"'"' optical spectral data.

35. The method for identifying a target individual as recited in claim 34, wherein the target identity is chosen as the most similar of all said enrolled persons whose enrollment spectral data are at least as similar to the said target spectral data as a predetermined measure of similarity.

36. The method for identifying a target individual as recited in claim 30, further comprising an enrollment database with optical spectral data collected from a number of enrolled persons, wherein said number is greater than one.

37. The method for identifying a target individual as recited in claim 30, further comprising an enrollment database with optical spectral data collected from a number of enrolled persons, wherein said number is equal to one.

38. The method for identifying a target individual as recited in claim 30, wherein said target spectrum is added to said enrollment optical spectral data after said identification.

39. The method for identifying a target individual as recited in claim 30, wherein said sub-epidermal tissue optical spectra include near-ultraviolet wavelengths.

40. The method for identifying a target individual as recited in claim 30, wherein said sub-epidermal tissue optical spectra include visible wavelengths.

41. The method for identifying a target individual as recited in claim 30, wherein said sub-epidermal tissue optical spectra include near-infrared wavelengths.

42. A system for identifying a target individual comprising:

an enrollment database including sub-epidermal tissue optical spectra collected from at least one enrolled person, each of said enrolled person'"'"'s sub-epidermal tissue optical spectrum having a plurality of measurement wavelengths obtained by applying a spectral band of optical wavelengths to said sub-epidermal tissue;

a spectrometer adapted to obtain at least one sub-epidermal tissue optical spectrum from said target individual by measuring optical radiation reflected from sub-epidermal tissue of said target individual, said target individual'"'"'s sub-epidermal tissue optical spectrum having a plurality of measurement wavelengths obtained by applying said spectral band of optical wavelengths to sub-epidermal tissue of said target individual;

a comparator adapted to compare said target individual sub-epidermal tissue optical spectrum and each said enrolled person'"'"'s sub-epidermal tissue optical spectrum by selecting for comparison a combination of individual wavelengths of said band emanating from said sub-epidermal tissue that are unique to each said person and insensitive to environmental and instrumental effects and provide a measure of a degree of similarity between said target individual'"'"'s sub-epidermal tissue optical spectrum and said enrolled person'"'"'s sub-epidermal tissue optical spectrum; and

an indicator adapted to identify at least one of said enrolled persons if the corresponding measure of the degree of similarity meets an established threshold value.

43. The system recited in claim 42, wherein said optical spectra include near-infrared wavelengths.

44. The system recited in claim 42, wherein said optical spectra include visible wavelengths.

45. The system recited in claim 42, wherein said optical spectra include near-ultraviolet wavelengths.