Method and systems for generating a subset of biometric representations

• US 8,031,981 B2
• Filed: 12/21/2007
• Issued: 10/04/2011
• Est. Priority Date: 12/21/2007
• Status: Active Grant

• Current Assignee: Corporation DAON Technology
• Original Assignee: Daon Holdings Limited
• Inventors: Peirce, Michael
• Primary Examiner(s): Repko; Jason M
• Assistant Examiner(s): Thomas; Mia M
• Application Number:
  US11/963,297
• Publication Number:
  US 20100312763A1
• Time in Patent Office: 1,383 Days
• Field of Search: 382/115, 382/181, 382/209, 382/305, 382/306, 340/5.52, 340/5.82, 707/821, 707/723, 358/403, 358/404
• US Class Current: 382/305
• CPC Class Codes: G06F 18/24323 Tree-organised classifiers
• CPC Class Codes: G06F 18/28 Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries
• CPC Class Codes: G06V 10/764 using classification, e.g. of video objects
• CPC Class Codes: G06V 10/772 Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries
• CPC Class Codes: G06V 40/70 Multimodal biometrics, e.g. combining information from different biometric modalities

Abstract

A system and method for enabling analysis of enrolled biometric data is presented. A plurality of vectors each having a plurality of score values representative of the relationship between individual ones of the enrolled biometrics with a plurality of biometric representations forming a filter set are described. Judicious use of the vectors enables a filtering of the enrolled biometric data on a dynamic basis.

Claims

1. A method of filtering a master database of enrolled biometric representations to generate a subset of enrolled biometric representations against which a presented biometric is authenticated, said method comprising:

matching individual enrolled biometric representations included in the master database of enrolled biometric representations against a filter set and generating a score for each of the matches, the filter set including a plurality of biometric representations;

generating vectors of the scores for each of the matched enrolled biometric representations;

associating the individual enrolled biometric representations with a corresponding generated vector; and

using the generated vectors in an authentication request to determine a subset of the enrolled biometric representations against which a presented biometric is authenticated.

2. The method of claim 1 wherein the subset of enrolled biometric representations is generated on a dynamic basis for each authentication request.

3. The method of claim 1, the authentication request comprising:

receiving a biometric representation of a user;

matching the received biometric representation against the filter set and generating a set of scores for the matches;

generating a vector of the scores for the received biometric representation;

comparing the vector for the received biometric representation with the vectors of the enrolled biometric representations; and

using the results of the comparison to determine the subset of the enrolled biometric representations against which the received biometric representation are to be tested.

4. The method of claim 3 wherein the received biometric representation is received from a biometric capture device.

5. The method of claim 3 wherein the received biometric representation is received from a third party application which effected capture of the biometric representation.

6. The method of claim 3, said comparing operation generating a plurality of vector comparison values, each of which indicates the level of similarity between the received biometric representation with an individual enrolled biometric representation.

7. The method of claim 6 further comprising applying a threshold factor to the plurality of vector comparison values to determine which of the enrolled biometric representations are used to form the subset.

8. The method of claim 1 wherein individual biometric representations relate to multiple biometrics.

9. The method of claim 8 wherein the multiple biometrics relate to at least one of:

multi-presentation whereby the same biometric instance is presented multiple times;

multi-instance whereby different instances of the same biometric modality is presented;

multi-modal wherein different modalities of biometric are presented;

multi-sensor wherein two or more different capture devices are used to effect capture of the biometric representations; and

multi-algorithm where two or more different feature extraction algorithms are used to generate the biometric representations from captured biometric images.

10. The method of claim 1 wherein each of the generated vectors define an ordered list of match scores, each of the match scores identifies the similarity of the biometric representation to an individual biometric representations of the filter set.

11. The method of claim 10 wherein the plurality of match scores are provided in a numerical format, said method further comprising translating the numerical format into a text format to provide for subsequent text searching.

12. The method of claim 1 wherein the master database includes a number of databases each being distinguished by the biometric stored.

13. The method of claim 1 further comprising storing the generated vectors separately from the master database.

14. The method of claim 1 wherein the generated vectors provide a representation of a relationship between an individual enrolled biometric representation and the filter set.

15. The method of claim 1 further comprising generating different vectors for different biometrics.

16. The method of claim 15, said generating different vectors operation comprising using multiple filters, each filter being appropriate to a specific biometric.

17. The method of claim 15 further comprising combining vectors of different biometrics into a single vector representative of a multiple biometric.

18. The method of claim 15 wherein on presentation of an authentication request containing multiple biometrics, individual multiple biometrics are matched against appropriate filter sets such that for each biometric and filter match a set of scores is created, the multiple sets of scores being used in the generation of multiple vectors, each of the multiple vectors being used in the selection of the subset.

19. The method of claim 18 wherein the multiple vectors are combined prior to comparison with combined multiple enrollment vectors.

20. The method of claim 18 wherein the multiple vectors are compared against multiple vectors of the enrolled biometric representations to provide a preliminary set of vector comparison scores.

21. The method of claim 20 further comprising weighting individual scores of the preliminary set of vector comparison scores prior to combining the preliminary set of vector comparison scores to provide a vector comparison score.

22. The method of claim 21 wherein the number of weightings used is related to the number of biometrics within the multiple biometric.

23. The method of claim 1 wherein the filter is derived exclusively from enrolled biometric representations from the master database.

24. The method of claim 1 wherein the filter is derived from enrolled biometric representations from the master database.

25. The method of claim 1 wherein the filter is derived from a brute force empirical search of a given dataset.

26. The method of claim 1 wherein the filter is derived from biometric error rate data.