Method and system for conducting identity matching

• US 7,835,548 B1
• Filed: 08/30/2010
• Issued: 11/16/2010
• Est. Priority Date: 03/01/2010
• Status: Active Grant

• Current Assignee: Corporation DAON Technology
• Original Assignee: DAON Holding Limited
• Inventors: Langley, Richard Jay
• Primary Examiner(s): Johns; Andrew W
• Application Number:
  US12/870,870
• Time in Patent Office: 78 Days
• Field of Search: 382/115, 382/116, 382/117, 382/118, 382/119, 382/124, 382/159, 382/160, 340/5.1, 340/5, 340/52, 340/5.53, 340/5.82, 340/5.83, 340/5.84
• US Class Current: 382/116
• CPC Class Codes: G06F 18/256 of results relating to different input data, e.g. multimodal recognition
• CPC Class Codes: G06V 30/242 Division of the character sequences into groups prior to recognition; Selection of dictionaries
• CPC Class Codes: G06V 30/2504 Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches
• CPC Class Codes: G06V 40/70 Multimodal biometrics, e.g. combining information from different biometric modalities

Abstract

A method of managing large scale biometric data identity matching includes identifying a plurality of biometric modalities to be used in conducting identity matches, determining one of the biometric modalities to be a primary biometric modality and determining whether at least one matching algorithm is compatible with matrix entries. When the at least one matching algorithm is compatible with a matrix entry the method includes assigning a plurality of matching systems to the matrix entry. The matching systems are configured to conduct 1:1 or 1:few matching after conducting a 1:N identity matching. Furthermore, the method includes comparing a probe against enrollment data records stored in a corresponding matching system, and storing information regarding a matching enrollment data record in a queue when the probe matches an enrollment data record.

Claims

1. A method of conducting biometric identification matching comprising:

dividing an entire population into subpopulations;

generating a matrix including a plurality of matrix entries;

associating each of the matrix entries with one of the subpopulations and with one of a plurality of matching systems that corresponds to the one subpopulation;

obtaining biometric data, including the obtained biometric data in a probe, and determining a row in the matrix corresponding to characteristics of the probe;

conducting a 1;

N identity match between the probe and each subpopulation using the one matching system corresponding to each subpopulation;

determining a candidate match when the probe matches a matching system enrollment data record in the corresponding one matching system;

conducting a 1;

1 verification to eliminate the candidate match if the candidate match is not a true match, the corresponding one matching system being associated with a reliable one of a plurality of biometric modalities;

storing the candidate match in a queue when the candidate match is confirmed; and

communicating the queue to an entity requiring the identity match.

2. A method in accordance with claim 1, said generating a matrix operation further comprising:

determining whether each matrix entry is compatible with one of a plurality of biometric modality matching algorithms;

associating each matrix entry with a compatible one of the biometric modality matching algorithms; and

associating each matrix entry associated with a compatible one of the biometric modality algorithms with an additional biometric modality matching algorithm.

3. A method in accordance with claim 2, said associating each of the matrix entries with one of the plurality of matching systems operation comprising:

associating each matrix entry associated with a compatible one of the biometric modality matching algorithms, when the compatible one of the biometric modality matching algorithms is based on a reliable biometric modality, with one of the matching systems that includes the compatible biometric modality algorithm for executing said conducting a 1;

N identity match operation based on the reliable biometric modality; and

associating each matrix entry that is incompatible with the biometric modality matching algorithms, when the biometric matching algorithms are based on the reliable biometric modality, with one of the matching systems that includes a parallel fusion algorithm for executing said conducting a 1;

N identity match operation.

4. A method in accordance with claim 2, further comprising executing said conducting a 1:

1 verification operation with the additional biometric modality matching algorithm.

5. A method in accordance with claim 1, said dividing an entire population operation comprising:

determining the plurality of biometric modalities based on criteria;

determining one of the plurality of biometric modalities that best conforms to the criteria as a primary biometric modality;

determining another one of the plurality of biometric modalities as a secondary biometric modality based on conformance to the criteria; and

determining a third one of the plurality of biometric modalities as a tertiary biometric modality based on conformance to the criteria.

6. A method in accordance with claim 5, said dividing an entire population operation further comprising:

establishing a hierarchy of the biometric modalities; and

establishing a sub-hierarchy for each biometric modality included in the hierarchy, each sub-hierarchy being based on the quality of biometric modality data to be included in the hierarchy and including at least high and low quality biometric modality data.

7. A system for conducting biometric identification matching comprising:

a computer configured as a server, said server including at least a database, said server being operable to store within said database at least enrollment data records of an entire population, determine a plurality of biometric modalities, determine quality categories for each of the biometric modalities, and generate a matrix including a plurality of matrix entries, wherein each of the matrix entries is associated with one of a plurality of subpopulations;

at least one client system configured to communicate with said server and being operable to at least obtain biometric data; and

a plurality of matching systems, wherein each matching system

corresponds to one of the subpopulations,

is operable to store therein matching system enrollment data records of the corresponding subpopulation, and

is configured to communicate with said server,

said server being operable to

communicate with said at least one client system and said matching systems,

associate each of the matrix entries with one of said matching systems,

include obtained biometric data in a probe,

determine characteristics of the probe and a row in the matrix corresponding to the probe, and

each matching system being operable to

conduct a 1;

N identity match by comparing the probe against the matching system enrollment data records stored therein,

determine a candidate match when the probe matches one of the matching system enrollment data records stored therein, and

conduct a 1;

1 verification to eliminate the candidate match if the candidate match is not a true match when the matching system enrollment data records stored therein are associated with a reliable one of the biometric modalities,

said server being further operable to

compare the probe against the enrollment data records of the entire population by communicating with each of said matching systems associated with the row such that each of said matching systems associated with the row conducts a 1;

N identity match, for the respective subpopulation, between the probe and the matching system enrollment data records stored therein,

store at least one candidate match in a queue when the at least one candidate match is confirmed, and

communicate the queue to an entity executing the identity match.

8. A system in accordance with claim 7, said server being further operable to:

determine whether each matrix entry is compatible with one of a plurality of biometric modality matching algorithms;

associate each matrix entry with a compatible one of the biometric modality matching algorithms; and

associate each matrix entry associated with a compatible one of the biometric modality matching algorithms with an additional biometric modality matching algorithm.

9. A system in accordance with claim 8, said server system being further operable to:

associate each matrix entry associated with a compatible one of the biometric modality matching algorithms, when the compatible one of the biometric modality matching algorithms is based on a reliable biometric modality, with one of said matching systems that includes the compatible one matching algorithm such that said one matching system is operable to conduct the 1;

N identity match based on the reliable biometric modality; and

associate each matrix entry that is incompatible with biometric modality matching algorithms when the biometric matching algorithms are based on the reliable biometric modality, with one of said matching systems that includes an algorithm for conducting parallel fusion.

10. A system in accordance with claim 8, wherein each of said matching systems associated with a matrix entry associated with a compatible one of the biometric modality matching algorithms is further operable to conduct the 1:

1 verification with the additional biometric modality matching algorithm.

11. A system in accordance with claim 7, said server being further operable to:

establish a hierarchy of the biometric modalities;

determine categories of quality for biometric data, wherein the categories include high quality data, low quality data and unavailable data; and

establish a sub-hierarchy for each biometric modality included in the hierarchy based on the categories.

12. A system in accordance with claim 11, said server being further operable to subdivide the entire population into the plurality of subpopulations by associating each individual included in the entire population with a respective category included in the sub-hierarchy.

13. A method in accordance with claim 11, said server is further operable to assign subpopulation designations to each of the plurality of subpopulations.

14. A method of conducting biometric identification matching comprising:

dividing an entire population into subpopulations;

generating a matrix including a plurality of matrix entries;

associating each of the matrix entries with one of the subpopulations and with one of a plurality of matching systems that corresponds to the one subpopulation;

obtaining biometric data, including the obtained biometric data in a probe, and determining a row in the matrix corresponding to characteristics of the probe;

conducting a 1;

N identity match between the probe and each subpopulation using the one matching system corresponding to each subpopulation;

determining a candidate match when the probe matches a matching system enrollment data record in the corresponding one matching system;

storing the candidate match in a queue; and

communicating the queue to an entity requiring the identity match.

15. A method in accordance with claim 14, said storing operation comprising:

sequentially evaluating the candidate match in a 1;

1 verification to eliminate the candidate match if the candidate match is not a true match; and

storing the candidate match in the queue when the candidate match is confirmed.

16. A method in accordance with claim 14, said generating a matrix operation further comprising:

determining whether each matrix entry is compatible with one of a plurality of biometric modality matching algorithms;

associating each matrix entry with a compatible one of the biometric modality matching algorithms; and

associating each matrix entry associated with a compatible one of the biometric matching algorithms with an additional biometric modality matching algorithm.

17. A method in accordance with claim 16, said associating each of the matrix entries with one of the plurality of matching systems operation comprising:

associating each matrix entry associated with a compatible one of the biometric modality matching algorithms, when the compatible one of the biometric modality matching algorithms is based on a reliable biometric modality, with one of the matching systems that includes the compatible biometric modality matching algorithm for executing said conducting a 1;

N identity match operation based on the reliable biometric modality; and

associating each matrix entry that is incompatible with the biometric modality matching algorithms, when the biometric matching algorithms are based on the reliable biometric modality, with one of the matching systems that includes a parallel fusion algorithm for executing said conducting a 1;

N identity match operation.

18. A method in accordance with claim 16, further comprising executing said conducting a 1:

1 verification operation with the additional biometric modality matching algorithm.

19. A method in accordance with claim 14, said dividing operation comprising:

establishing a hierarchy of biometric modalities;

determining categories of quality for biometric data, wherein the categories include high quality data, low quality data and unavailable data; and

establishing a sub-hierarchy for each biometric modality included in the hierarchy based on the categories.

20. A method in accordance with claim 19, said dividing operation further comprising subdividing the entire population into the plurality of subpopulations by associating each individual included in the entire population with a respective category included in the sub-hierarchy.