Apparatus and method for biometric database management system

• US 20050223016A1
• Filed: 02/22/2005
• Published: 10/06/2005
• Est. Priority Date: 03/10/2004
• Status: Active Grant

• Current Assignee: InteliTrac, Inc.
• Original Assignee: Tianlong Chen, Yingbin Wang
• Inventors: Chen, Tianlong, Wang, Yingbin
• Granted Patent: US 7,689,005 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2462 Approximate or statistical queries
• CPC Class Codes: G07C 9/37 using biometric data, e.g. fingerprints, iris scans or voice recognition

Abstract

A method and system for constructing a database management system for managing biometric data is disclosed. The disclosed system receives data from another database or from and enrollment process, encodes the data with an encoding plug-in, and stores the encoded data in a biometric data storage. The data may be enhanced before being stored. Incoming target data likewise is encoded using an encoding plug-in and may be pre-processed, and is sent to a matching algorithm that is either built-in or a plug-in algorithm. Further processing may occur after application of the matching algorithm. The disclosed database management system can be used not only for biometric database, but also for other similar types of data management.

Claims

1. A method for implementing a biometric database management system, comprising the steps of:

providing a biometric data storage for storing and managing biometric data;

providing an enrollment process;

providing a searching process;

providing an encoding plug-in interface in said enrollment process to convert a biometric data into an encoded biometric data;

providing an encoding plug-in interface in said searching process to convert a biometric data into an encoded biometric data;

providing a searching algorithm management for said searching process, said searching algorithm having at least functionality of one-to-one similarity comparing of said biometric data;

providing configurable dataflow for said enrollment and searching processes; and

providing a unique ID for a set of related biometric data in said biometric data storage.

2. A method according to claim 1 wherein said enrollment process comprises a pre-encoding process before said encoding plug-in interface.

3. A method according to claim 1 wherein said enrollment process comprises a pro-encoding process after said encoding plug-in interface.

4. A method according to claim 3 wherein said pro-encoding process provides functionality of indexing or index grouping.

5. A method according to claim 4 wherein said indexing or indexing group is created by neural network methods.

6. A method according to claim 2 wherein said pre-encoding process comprises a plug-in interface for user to insert customized functionality.

7. A method according to claim 2 wherein said pre-encoding process comprises a built-in procedure provided by said biometric database management system.

8. A method according to claim 3 wherein said pro-encoding process comprises a plug-in interface for user to insert customized functionality.

9. A method according to claim 3 wherein said pro-encoding process comprises a built-in procedure provided by said biometric database management system.

10. A method according to claim 1 wherein said search process comprises a plug-in matching algorithm interface to handle matching between two encoded biometric data, and said matching algorithm generates a similarity score as a comparison result.

11. A method according to claim 1 wherein said encoding plug-in interface in said enrollment process can do nothing.

12. A method according to claim 1 wherein said encoding plug-in interface in said searching process can do nothing.

13. A method according to claim 1 wherein said searching process has a built-in matching algorithm as part of the said searching algorithm management.

14. A method according to claim 1 wherein said searching process has a plug-in matching algorithm interface as part of the said searching algorithm management.

15. A method according to claim 1 wherein said searching process includes a pre-matching procedure before the said searching algorithm management, and said pre-matching procedure can access said biometric data storage.

16. A method according to claim 15 wherein said pre-procedure is a plug-in interface for any process happening before said matching algorithm.

17. A method according to claim 15 wherein said pre-matching procedure is a built-in procedure for any process happening before said matching algorithm.

18. A method according to claim 15 wherein said pre-matching procedure includes indexing or index group calculation.

19. A method according to claim 1 wherein said searching process includes a pro-matching procedure after the said searching algorithm management, and said pro-matching procedure can access said biometric data storage.

20. A method according to claim 19 wherein said pro-matching procedure is a plug-in interface for any process happening after said matching algorithm.

21. A method according to claim 19 wherein said pro-procedure is a built-in procedure for any process happening after said matching algorithm.

22. A method according to claim 19 wherein said pro-matching procedure includes a group voting procedure.

23. A method according to claim 19 wherein said pro-matching procedure generates a combined similarity score.

24. A method according to claim 19 wherein said pro-matching procedure generates a list of similarity scores.

25. A method according to claim 1 wherein said biometric database management system is an in-memory database management.

26. A method according to claim 25 wherein said biometric database management system uses a relative virtual address as internal memory address, and said relative virtual address includes an offset ID of a memory block in a computer system in which said biometric database management system is hosted.

27. A method according to claim 1 wherein said encoding plug-in interface in said enrollment process is a dynamic, linked library.

28. A method according to claim 1 wherein said encoding plug-in interface in said searching process is a dynamic linked library.

29. A method according to claim 6 wherein said plug-in pre-encoding process is a dynamic linked library.

30. A method according to claim 8 wherein said plug-in matching algorithm is a dynamic linked library.

31. A method according to claim 15 wherein said pre-matching process can be disabled or enabled by configuration.

32. A method according to claim 19 wherein said pro-matching process can be disabled or enabled by configuration.

33. A method according to claim 1 wherein said searching algorithm has one-to-many similarity comparing said biometric data with each resulting in a similarity score.

34. A biometric database management system comprising:

means for encoding reference biometric data;

means for enhancing said reference biometric data;

a biometric data storage for storing said encoded reference biometric data and said enhanced encoded reference biometric data;

means for encoding target biometric data;

means for pre-processing said encoded target biometric data;

a matching algorithm for comparing said pre-processed encoded target biometric data to said reference biometric data; and

means for pro-processing a result of said matching algorithm.