Systems and methods with identity verification by comparison & interpretation of skin patterns such as fingerprints

• US 20020041700A1
• Filed: 06/27/2001
• Published: 04/11/2002
• Est. Priority Date: 09/09/1996
• Status: Active Grant

• Current Assignee: Bioscrypt, Inc. (Advent International Corporation)
• Original Assignee: Areté Associates, Inc.
• Inventors: Therbaud, Lawrence R.
• Granted Patent: US 6,636,621 B2
• Time in Patent Office: Days
• US Class Current: 382/124
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction
• CPC Class Codes: G06V 40/1365 Matching; Classification
• CPC Class Codes: G07C 9/37 using biometric data, e.g. fingerprints, iris scans or voice recognition

Abstract

Preferably a sensor receives a print image from an authorized person to form a template, and from a candidate to form test data. Noise variance is estimated from the test data as a function of position in the image, and used to weight the importance of comparison with the template at each position. Test data are multilevel, and are bandpassed and normalized—and expressed as local sinusoids—for comparison. A ridge spacing and direction map of the template is stored as vector wavenumber fields, which are later used to refine comparison. Global dilation—and also differential distortions—of the test image are estimated, and taken into account in the comparison. Comparison yields a test statistic that is the ratio, or log of the ratio, of the likelihoods of obtaining the test image assuming that it respectively was, and was not, formed by an authorized user. The test statistic is compared with a threshold value, preselected for a desired level of certainty, to make the verification decision—which controls access to a utilization system such as facilities, equipment, a financial service, or a system for providing or receiving information. Certain forms of the invention also encompass the utilization system. Nonvolatile memory holds instructions for automatic operation as described above.

Claims

1. Apparatus for verifying the identity of a person by comparing test data representing a two-dimensional test image of that person'"'"'s skin-pattern print with reference data derived from a two-dimensional reference skin-pattern print image obtained during a prior enrollment procedure, said apparatus comprising:

means for extracting from the test data an estimate of noise variance in the test data as a function of position in the test image;

means for comparing portions of the test and reference data, for corresponding positions in the two images;

means for weighting the importance of comparison for each portion, in accordance with the noise-variance estimate for the corresponding position;

means, responsive to the weighting means, for making an identity-verification decision; and

nonvolatile memory means for holding instructions for automatic operation of the foregoing means.

2. The apparatus of claim 1, further comprising:

sensor means for acquiring such test data; and

means, responsive to the decision-making means, for operating a switch.

3. Apparatus for verifying the identity of a person by comparing test data representing a two-dimensional test image of that person'"'"'s skin-pattern print with reference data derived from a two-dimensional reference skin-pattern print image obtained during a prior enrollment procedure, said apparatus comprising:

means for deriving from the test data corresponding multilevel test data that are bandpassed and normalized;

means for comparing portions of the bandpassed and normalized multilevel test data with the reference data;

means, responsive to the comparing means, for making an identity-verification decision; and

nonvolatile memory means for holding instructions for automatic operation of the foregoing means.

4. The apparatus of claim 3, further comprising:

means for preparing the reference data for storage or export, said preparing means comprising;

means for at least downsampling and normalizing the reference data, and then expressing the thus-prepared reference data in two- or one-bit form before storage or export.

5. The apparatus of claim 4, further comprising:

means for extracting the reference data from storage or from an imported data set for use in verification, said extracting means comprising;

means for bandpassing, normalizing and smoothing the stored or imported data before use in verification.

6. Apparatus for verifying the identity of a person by comparing test data representing a two-dimensional test image of that person'"'"'s skin-pattern print with reference data derived from a two-dimensional reference skin-pattern print image obtained during a prior enrollment procedure, said apparatus comprising:

means for expressing the test data in the form of local sinusoids;

means for expressing the reference data in the form of local sinusoids;

means for comparing portions of the sinusoidally expressed test data with the sinusoidally expressed reference data;

means, responsive to the comparing means, for making an identity-verification decision; and

nonvolatile memory means for holding instructions for automatic operation of the foregoing means.

7. The apparatus of claim 6, further comprising:

means for deriving from the reference data a map of ridge spacing and direction, and for storing the map as one or more vector wavenumber fields; and

wherein the comparing means comprise means for using the vector wavenumber fields to refine said comparing; and

means, responsive to the comparing means, for making an identity-verification decision.

8. Apparatus for verifying the identity of a person by comparing test data representing a two-dimensional test image of that person'"'"'s skin-pattern print with reference data derived from a two-dimensional reference skin-pattern print image obtained during a prior enrollment procedure, said apparatus comprising:

means for deriving from the reference data a map of ridge spacing and direction, and for storing the map as one or more vector wavenumber fields;

means for comparing portions of the test data with the reference data, said comparing means comprising means for using the vector wavenumber fields to refine said comparing;

means, responsive to the comparing means, for making an identity-verification decision; and

nonvolatile memory means for holding instructions for automatic operation of the foregoing means.

9. Apparatus for verifying the identity of a person by comparing test data representing a two-dimensional test image of that person'"'"'s skin-pattern print with reference data derived from one or more two-dimensional reference skin-pattern print images obtained during a prior enrollment procedure;

said apparatus being for use in the presence of an assumed dilation of the test image relative to the reference image;

said apparatus comprising;

means for estimating the assumed dilation of the test image relative to a reference image;

means for comparing the test data with the reference data, taking into account the estimated dilation;

means, responsive to the comparing means, for making an identity-verification decision; and

nonvolatile memory means for holding instructions for automatic operation of the foregoing means.

10. The apparatus of claim 9:

further comprising means for extracting from the reference data a subset thereof; and

wherein the estimating means comprise means for trial-matching the subset of the reference data with successive portions of the test data, to find a portion of the test data that best fits the reference data subset.

11. The apparatus of claim 10:

wherein the estimating means further comprise means for taking possible dilation into account in gauging the test data portions against the reference-data subset, to determine relative dilation;

further comprising means for applying the determined dilation to roughly equalize the test and reference data with respect to the assumed dilation; and

wherein the comparing means compare the test and reference data after said rough equalization.

12. The apparatus of claim 9, wherein:

the extracting means comprise means for extracting from the reference data plural subsets thereof; and

the estimating means comprise means for trial-matching each of plural subsets of the reference data with successive portions of the test data.

15. The apparatus of claim 14, wherein:

the applying and comparing means comprise means for using the estimated distortion to generate a matched filter for use in forming a test statistic.

17. The apparatus of claim 16, wherein the taking-into-account means comprise:

means for weighting the importance of the estimating or comparing, or both, as a function of position in accordance with the noise-variance estimate for the corresponding position.

18. The apparatus of claim 13, wherein:

the estimating means comprise means for using a maximum-likelihood method to estimate the distortion field.

19. The apparatus of claim 18, wherein:

said estimating means further comprise means for using the maximum-likelihood-estimated distortion field to generate a matched filter for use in forming a test statistic.

20. The apparatus of claim 18, wherein:

said using means comprise means for taking account of a priori statistics in estimating the distortion field.

21. The apparatus of claim 20, wherein:

said using means comprise means for using a Fourier-space model for taking account of the a priori statistics.

22. The apparatus of claim 13, wherein:

the estimating means comprise means for demodulating the test data preliminary to extracting a distortion field therefrom.

23. The apparatus of claim 20, wherein:

the estimating means comprise means for applying information developed from the reference data as a carrier field, in demodulating the test data.

24. The apparatus of claim 13, wherein the estimating means comprise:

the estimating means comprise means for avoiding errors due to phase shifts greater than half the local periodicity of the assumed carrier field.

25. The apparatus of claim 24, wherein:

the estimating means comprise means for beginning the estimating with a first approximation that is localized and of small magnitude, and iterating so that progressive approximations propagate in the frequency or spatial domain, or both; and

the avoiding means comprise means for controlling rate of propagation to minimize likelihood of introducing a phase shift greater than half the local periodicity in any single iteration.

26. The apparatus of claim 25, wherein the controlling means comprise:

means for monitoring a measure of the rate at which the distortion is changing; and

means, responsive to the monitoring means, for limiting the rate of propagation as a function of position, in accordance with said measure.

27. The apparatus of claim 13, wherein the estimating means comprise:

means for providing an assumed carrier field for use in demodulating the test data;

means for analyzing the demodulated test data to approximately ascertain distortion associated with the demodulated test data.

28. The apparatus of claim 13, wherein:

the estimating means comprise means for downsampling the test data according to anticipated degree of abruptness of at least one type of distortion, in preparation for estimating the distortion.

29. The apparatus of claim 28, further comprising:

means for extracting from prior statistics an estimate of the power spectral density for distortion fields generally; and

means for setting sampling intervals as a function of position, in accordance with the estimate of power spectral density.

30. The apparatus of claim 28:

wherein the estimating means comprise means for downsampling the test data before estimating the distortion;

further comprising means for applying the estimated distortion to approximately equalize the test and reference data with respect to the assumed distortion;

further comprising means for interpolating the estimated distortion, to upsample the estimated distortion for use in the applying means; and

wherein the comparing means compare the thus-approximately-equalized test and reference data.

31. The apparatus of claim 13, wherein:

the estimating means comprise means for using a gradient search to estimate the distortion.

13. Apparatus for verifying the identity of a person by comparing test data representing a two-dimensional test image of that person'"'"'s skin-pattern print with reference data derived from a two-dimensional reference skin-pattern print image obtained during a prior enrollment procedure;

said apparatus being for use in the presence of an assumed distortion of the test image relative to the reference image; and

said apparatus comprising;

means for estimating the assumed distortion of the test image relative to a reference image;

means for comparing the test data with the reference data, taking into account the estimated distortion;

means responsive to the comparing means for making an identity-verification decision; and

nonvolatile memory means for holding instructions for automatic operation of the foregoing means.

14. The apparatus of claim 13:

further comprising means for applying the estimated distortion to approximately equalize the test and reference data with respect to the assumed distortion; and

wherein the comparing means compare the thus-approximately-equalized test and reference data.

16. The apparatus of claim 13:

further comprising means for extracting from the test data an estimate of noise variance in the test data as a function of position in the test image; and

wherein the estimating means and the comparing means perform said estimating and said comparing, respectively, as a function of position;

wherein the estimating means or the comparing means, or both, comprise means for taking into account the estimated noise variance.

32. Apparatus for verifying the identity of a person by comparing test data representing a two-dimensional test image of that person'"'"'s skin-pattern print with reference data derived from a two-dimensional reference skin-pattern print image obtained during a prior enrollment procedure, said apparatus comprising:

means for comparing the test data with the reference data to form a test statistic as the ratio, or logarithm of the ratio, of;

likelihood of obtaining the test image, assuming that said person also formed the reference fingerprint image, and

likelihood of obtaining the test image, assuming that a different person formed the reference fingerprint image;

means, responsive to the test statistic, for making an identity-verification decision; and

nonvolatile memory means for holding instructions for automatic operation of the foregoing means.

33. The apparatus of claim 32, wherein:

the decision-making means comprise means for comparing the test statistic with a threshold value preselected to impose a desired level of certainty in verification.

37. The method of claim 36, further comprising:

operating a sensor to acquire the test data; and

responsive to the decision-making step, operating a switch if identity is verified.

34. Apparatus for receiving surface-relief data from a sensor that acquires surface-relief data from a relieved surface such as a finger, and in response controlling access to facilities, equipment, a financial service, or a system for providing or receiving information;

said apparatus being for use in presence of an assumed dilation of the surface, and comprising;

a system for processing the received data to determine identity of the relieved surface, said system including;

means for analyzing the data to estimate the assumed dilation,

means for comparing the test data with reference data, taking into account the estimated dilation, and

means, responsive to the comparing means, for making an identity-verification decision;

means for applying the determined identity to control access to such facilities, equipment, financial service, or source or reception of information; and

nonvolatile memory means for holding instructions for automatic operation of the foregoing means.

35. A secured system subject to access control based upon surface-relief data from a relieved surface such as a finger;

said system being for use in the presence of an assumed distortion of the relieved surface; and

said system comprising;

utilization means, susceptible to misuse in the absence of a particular such relieved surface that is related to an authorized user, said utilization means being selected from the group consisting of;

a facility,

apparatus,

means for providing a financial service, and

means for providing or receiving information;

sensor means for acquiring surface-relief data from such a relieved surface;

means for processing the data to determine identity of the relieved surface, and for applying the determined identity to control access to the utilization means, said processing and applying means including;

means for analyzing the data to estimate the assumed distortion,

means for comparing the test data with reference data related to said particular relieved surface related to the authorized user, taking into account the estimated distortion, and

means, responsive to the comparing means, for making an identity-verification decision; and

nonvolatile memory means for holding instructions for automatic operation of the foregoing means.

36. A method for verifying the identity of a person by comparing test data representing a two-dimensional test image of that person'"'"'s skin-pattern print with reference data derived from a two-dimensional reference skin-pattern print image obtained during a prior enrollment procedure, said method comprising the steps of:

extracting from the test data an estimate of noise variance in the test data as a function of position in the test image;

comparing portions of the test and reference data, for corresponding positions in the two images;

weighting the importance of comparison for each portion, in accordance with the noise-variance estimate for the corresponding position;

responsive to the weighting means, making an identity-verification decision are; and

in nonvolatile memory holding instructions for automatic operation of the foregoing steps.