System and method for fast biometric pattern matching

• US 20060104493A1
• Filed: 11/09/2005
• Published: 05/18/2006
• Est. Priority Date: 11/12/2004
• Status: Active Grant

• Current Assignee: Thales DIS France SAS (Thales SA)
• Original Assignee: Cogent Systems, Inc. (Thales SA)
• Inventors: Chen, Ping, Wang, Jing, Xue, Huansheng, Hsieh, Ming
• Granted Patent: US 7,616,788 B2
• Time in Patent Office: Days
• US Class Current: 382/125
• CPC Class Codes: G06V 40/1371 Matching features related to minutiae or pores

Abstract

A method and system for matching two biometric images including receiving an input biometric image; generating an index table for the input biometric image, wherein the index table includes a quality quantity for each minutia of the input biometric image; receiving a second biometric image; generating a number of patterns for a first minutia of the second biometric image; associatively accessing the index table by the generated number of patterns; accumulating quality quantities accessed from the index table for each minutia of the input biometric image for the number of patterns of the first minutia of the second biometric image; and selecting a minutia candidate of the input biometric image responsive to the accumulated quality quantities.

Claims

1. A method for matching two biometric images comprising:

receiving an input biometric image;

generating an index table for the input biometric image, wherein the index table includes a quality quantity for each minutia of the input biometric image;

receiving a second biometric image;

generating a number of patterns for a first minutia of the second biometric image;

associatively accessing the index table by the generated number of patterns;

accumulating quality quantities accessed from the index table for each minutia of the input biometric image for the number of patterns of the first minutia of the second biometric image; and

selecting a minutia candidate of the input biometric image responsive to the accumulated quality quantities.

2. The method of claim 1, wherein the second biometric image is selected from a plurality of biometric images stored in a database.

3. The method of claim 1, wherein the number of generated patterns for the first minutia of the second biometric image is user programmable.

4. The method of claim 1, wherein the biometric images are fingerprint images.

5. The method of claim 1, wherein the biometric images are one or more of the group eye iris images, retina images, and facial images.

6. The method of claim 1, further comprising repeating the generating, associatively accessing, accumulating, and selecting steps for each minutia of the second biometric image.

7. The method of claim 6, further comprising generating a match score for the second biometric image.

8. The method of claim 1, wherein the generating a number of patterns comprises generating a number of minutiae pair relation patterns for a first minutia of the second biometric image.

9. The method of claim 1, wherein the generating a number of patterns comprises generating a number of minutiae triple relations patterns for a first minutia of the second biometric image.

10. The method of claim 1, wherein the index table is a two dimensional matrix, each column is a minutia index number, each row represents a pattern of the input biometric image, and each cell of the matrix represents a quality of a respective pattern.

11. The method of claim 1, wherein each entry of the index table represents all minutia points of the input biometric image which have same patterns as the address to that entry.

12. The method of claim 1, wherein the generating a number of patterns comprises generating only patterns that are within a predetermined distance from the first minutia of the second biometric image.

13. The method of claim 1, wherein the quality quantity includes redundant patterns in a predetermined error range.

14. The method of claim 7, wherein the generating a match score for the second biometric image comprises generating a plurality of matched minutia/quality pairs for each minutia of the second biometric image, selecting a best matched pair, and determining a match score for the selected best matched pair.

15. A system for matching two biometric images comprising:

an input buffer for receiving an input biometric image;

a memory or storing an index table for the input biometric image, wherein the index table includes a quality quantity for each minutia of the input biometric image;

a database for storing a second biometric image;

means for generating a number of patterns for a first minutia of the second biometric image and associatively accessing the index table by the generated number of patterns;

a plurality of accumulators for accumulating quality quantities accessed from the index table for each minutia of the input biometric image for the number of patterns of the first minutia of the second biometric image; and

a plurality of comparators for selecting a minutia candidate of the input biometric image responsive to the accumulated quality quantities.

16. The system of claim 15, wherein the number of generated patterns for the first minutia of the second biometric image is user programmable.

17. The system of claim 15, wherein the biometric images are one or more of the group fingerprint images, eye iris images, retina images, and facial images.

18. The system of claim 15, further comprising means for repeating the generating, associatively accessing, accumulating, and selecting steps for each minutia of the second biometric image.

19. The system of claim 18, further comprising means for generating a match score for the second biometric image.

20. The system of claim 15, wherein the means for generating a number of patterns comprises generating a number of minutiae pair relation patterns for a first minutia of the second biometric image.

21. The system of claim 15, wherein the index table is a two dimensional matrix, each column is a minutia index number, each row represents a pattern of the input biometric image, and each cell of the matrix represents a quality of a respective pattern.

22. The system of claim 15, wherein each entry of the index table represents all minutia points of the input biometric image which have same patterns as the address to that entry.

23. The system of claim 15, wherein the means for generating a number of patterns comprises means for generating only patterns that are within a predetermined distance from the first minutia of the second biometric image.

24. The system of claim 15, wherein the quality quantity includes redundant patterns in a predetermined error range.

25. A system for identifying a biometric image candidate comprising:

means for receiving an input biometric image;

means for generating an index table for the input biometric image, wherein the index table includes a quality quantity for each minutia of the input biometric image;

means for receiving a second biometric image;

means for generating a number of patterns for a first minutia of the second biometric image;

means for associatively accessing the index table by the generated number of patterns;

means for accumulating quality quantities accessed from the index table for each minutia of the input biometric image for the number of patterns of the first minutia of the second biometric image; and

means for selecting a minutia candidate of the input biometric image responsive to the accumulated quality quantities.

26. The system of claim 25, wherein the biometric images are one or more of the group fingerprint images, eye iris images, retina images, and facial images.

27. The system of claim 25, further comprising means for repeating the generating, associatively accessing, accumulating, and selecting steps for each minutia of the second biometric image.

28. The system of claim 27, further comprising means for generating a match score for the second biometric image.