Multimodal biometric platform

• US 20060104485A1
• Filed: 11/16/2004
• Published: 05/18/2006
• Est. Priority Date: 11/16/2004
• Status: Active Grant

• Current Assignee: Tech5 USA Incorporated
• Original Assignee: Imageware Systems Incorporated
• Inventors: Dlab, Daniel, Willis, William Frederic, Miller, S. James Jr., Lau, Johann H.
• Granted Patent: US 7,298,873 B2
• Time in Patent Office: Days
• US Class Current: 382/115
• CPC Class Codes: G06V 10/95 structured as a network, e.g. client-server architectures
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands

Abstract

A multimodal biometric identification or authentication system includes a plurality of biometric clients. Each of the biometric clients may include devices for capturing biometric images of a plurality of types. The system includes a router in communication with the biometric clients. The router receives biometric images from, and returns biometric scores or results to, the biometric clients. The system includes a plurality of biometric matching engines in communication with the router. Each biometric matching engine includes multiple biometric processors. Each biometric processor is adapted to process biometric data of a particular type. The biometric matching engines transmit and receive biometric data to and from the router.

Claims

1. A biometric system, which comprises:

a biometric client, said biometric client including means for capturing biometric images in a plurality of modes;

a router in communication with said biometric client to receive biometric images from said biometric client; and

,

a biometric matching engine in communication with said router to receive biometric images from said router, said biometric matching engine including means for processing biometric images in a plurality of modes.

2. The biometric system as claimed in claim 1, wherein said biometric matching engine includes:

means for creating biometric templates from biometric images received from said router.

3. The biometric system as claimed in claim 2, wherein said biometric matching engine includes:

means for sending to said router biometric templates.

4. The biometric system as claimed in claim 1, wherein said biometric matching engine includes:

means for caching biometric templates received from said router.

5. The biometric system as claimed in claim 1, wherein said biometric matching engine includes means for caching biometric templates in physical memory.

6. The biometric system as claimed in claim 1, wherein said biometric matching engine includes means for comparing cached biometric templates with probe template.

7. The biometric system as claimed in claim 1, wherein said biometric matching engine includes means for normalizing scores resulting from comparison of cached templates with probe templates.

8. The biometric system as claimed in claim 1, wherein said biometric matching engine includes means for sending to said router scores resulting from comparison probe templates with cached templates.

9. The biometric system as claimed in claim 8, wherein said router includes means for sending to said biometric client scores received from said biometric matching engine.

10. The biometric system as claimed in claim 1, wherein said biometric matching engine includes:

a first biometric processor for comparing biometric templates of a first type and returning scores based on comparison of biometric templates of said first type;

a first biometric plugin coupled to said first biometric processor;

a second biometric processor for comparing biometric templates of a second type and returning scores based on comparison of biometric templates of said second type; and

,

a second biometric plugin coupled to said second biometric processor.

11. A biometric system, which comprises:

a plurality of biometric clients, each of said biometric client including means for capturing biometric images of a plurality of types;

a router in communication with said biometric clients to receive biometric images from said biometric clients; and

,

a plurality of biometric matching engines in communication with said router to receive biometric images from said router, each said biometric matching engine including means for processing biometric images of a plurality of types.

12. The biometric system as claimed in claim 11, wherein said biometric matching engines comprise:

a first group of biometric matching engines, each biometric matching engine of said first group including mean for processing a first set of biometric image types; and

,

a second group of biometric matching engines, each biometric matching engine of said second group including mean for processing a second set of biometric image types.

13. The biometric system as claimed in claim 12, where said router includes:

means for sending to one biometric matching engine of said first group a biometric image of one of said first biometric image types.

14. The biometric system as claimed in claim 13, wherein each biometric matching engine of said first group includes:

means for creating biometric templates from biometric images of said first set of image types.

15. The biometric system as claimed in claim 14, wherein each biometric matching engine of said first group includes:

means for sending to said router biometric templates created from images of said first set of image types.

16. The biometric system as claimed in claim 15, wherein said router includes:

means for sending to each biometric matching engine biometric templates created from images of said first set of image types.

17. The biometric system as claimed in claim 16, wherein each biometric matching engine of said first group includes:

means for caching biometric templates received from said router.

18. The biometric system as claimed in claim 16, wherein each biometric matching engine of said first set includes means for caching biometric templates in physical memory.

19. The biometric system as claimed in claim 17, wherein each biometric matching engine of said first group includes:

means for comparing cached biometric templates with a probe template.

20. The biometric system as claimed in claim 15, wherein said router includes:

means for sending to a selected one of said biometric matching engines of said first group a biometric template created from a biometric image of said first type.

21. The biometric system as claimed in claim 11, wherein said router is configurable to operate in a mirrored mode, wherein a biometric template is sent to each biometric matching engine of said first group for caching.

22. The biometric system as claimed claim 21, wherein said router is configured in said mirrored mode to send to a selected one of said biometric matching engines of said first group a target biometric image of said first type.

23. The biometric system as claimed in claim 11, wherein said router is configurable to operate in a striped mode, wherein a biometric template is sent to a selected one of said biometric matching engines of said first group for caching.

24. The biometric system as claimed claim 23, wherein said router is configured in said striped mode to send to each of said biometric matching engines of said first group a target biometric image of said first type.

25. A biometric matching engine, which comprises:

means for creating biometric templates of a first type;

means for caching biometric templates of said first type;

pluging means for providing to a first biometric processor probe templates and cached templates of said first type; and

,

means for sending to a router scores.

26. The biometric matching engine as claimed in claim 25, including means for normalizing scores returned from said first biometric processor.

27. A method of processing multimodal biometric data, which comprises:

receiving at a router multimodal biometric enrollment data, said multimodal biometric enrollment data comprising a biometric image of a first type and a biometric image of a second type;

sending said biometric image of said first type and a biometric image of said second type to one of a group of multimodal biometric matching engines;

creating from said biometric image of said first type a first biometric template and from said biometric image of said second type a second biometric template; and

,

sending said first and second biometric templates to said router.

31. The method as claimed in claim 27, including:

mirroring biometric templates across the multimodal biometric matching engines of said group.

32. The method as claimed in claim 31, wherein mirroring includes:

caching each biometric template at each biometric matching engine of said group.

33. The method as claimed in claim 31, including:

sending a target image of said first type and a target image of said second type to a selected biometric matching engine of said group for processing.

28. The method as claimed in claim 37, including:

striping biometric templates across the multimodal biometric matching engines of said group.

29. The method as claimed in claim 28, wherein striping comprises:

caching a biometric template at a single biometric matching engine of said group.

30. The method as claimed in claim 28, including:

sending a target image of said first type and a target image of said second type to each biometric matching engine of said group for processing.

34. A method of producing a score in a multimodal biometric system which comprises:

forming a set of enrolled biometric data records, each biometric data record comprising a record identifier, an enrolled biometric template of a first type and an enrolled biometric template of a second type;

comparing a probe biometric template of said first type to enrolled biometric templates of said first type to produce a set of first scores;

saving in a match set biometric data records wherein the first score for the data record is greater than a first biometric threshold;

comparing a probe biometric template of said second type to biometric templates of said second type in said match set to produce a set of second scores; and

,

saving biometric data records wherein the second score for the data record is greater than a second biometric threshold.

35. A method of normalizing biometric scores, which comprises:

enrolling in a biometric database a plurality of individuals by storing in said database for each individual a plurality of biometric templates of one type;

comparing each biometric template in said database with every other biometric template of said database to obtain biometric scores;

if a biometric score is obtained by comparing one biometric template of an individual with an other biometric template for that same individual, putting that score in a matching category; and

,

analyzing the scores in the matching category to determine a probability that a particular score is a matching score.

36. The method as claimed in claim 35, including:

if a biometric score is obtained by comparing a biometric template of an individual with a biometric template for different individual, putting that score in a non-matching category; and

,

analyzing the scores in the non-matching category to determine a probability that a particular score is not a matching score.