Knowledge Learning System and Process for Fingerprint Verifications

• US 20100111378A1
• Filed: 11/04/2008
• Published: 05/06/2010
• Est. Priority Date: 11/04/2008
• Status: Active Grant

• Current Assignee: Signal Processing, Inc.
• Original Assignee: Signal Processing, Inc.
• Inventors: KWAN, Chiman
• Granted Patent: US 8,295,561 B2
• Time in Patent Office: Days
• US Class Current: 382/124
• CPC Class Codes: G06V 10/98 Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
• CPC Class Codes: G06V 40/1335 Combining adjacent partial images (e.g. slices) to create a composite input or reference pattern; Tracking a sweeping finger movement

Abstract

This invention relates to a novel technology of fingerprint verifications and identifications based on the accumulated knowledge base.

Claims

1. A method of knowledge learning for fingerprint verifications comprises steps of, providing a fingerprint scanner;

providing a fingerprint image database (FID);

providing a minutiae point extraction and fusion module (MPEFM);

providing a minutiae point database (MPD);

providing a master minutiae point database (MMPD) wherein the MMPD comprises a plurality of master minutiae point records; and

providing a verification module (VM).

2. The method of knowledge learning for fingerprint verifications of claim 1 further comprises steps of,

operating the fingerprint scanner to scan first set of fingerprints and convert the scanned first set of fingerprints into image data;

transmitting the first set of fingerprint image data to the FID and storing the first set of fingerprint image data in the FID;

performing identifications of first set of minutiae points by the MPEFM from the first set of fingerprint image data; and

storing information of the first set of minutiae points in the MPD, wherein the information of the first set of minutiae points is represented by a mathematical vector form.

3. The method of knowledge learning for fingerprint verifications of claim 2, wherein

the mathematical vector form includes information of a x-coordinate, a y-coordinate, and an angle in view of a selected reference Cartesian coordinate system.

4. The method of knowledge learning for fingerprint verifications of claim 2 further comprises steps of,

performing a comparison of the information of the first set of minutiae points with the plurality of master minutiae point records to determine a success of the comparison, wherein

the success is determined when quantity of same minutiae points between the first set of minutiae points and minutiae points within a specific master minutiae record exceeds a predetermined quantity requirement;

and

when the determination is a success, generating a report of success and terminating the verification.

5. The method of knowledge learning for fingerprint verifications of claim 4 further comprises, operating the fingerprint scanner to scan another set of fingerprints and convert the scanned another set of fingerprints into image data only when the comparison is not a success;

in response of the operation of the fingerprint scanner to scan another set of fingerprints, perform further steps including;

transmitting the another set of fingerprint image data to the FID and storing the another set of fingerprint image data in the FID;

performing identifications of another set of minutiae points by the MPEFM from the another set of fingerprint image data;

performing a fusion process to create a combined minutiae point record, wherein

the combined minutiae point record includes information of the minutiae points from the first set of fingerprints and the minutiae points from another set of fingerprints.

6. The method of knowledge learning for fingerprint verifications of claim 5, wherein the fusion process comprises steps of:

applying a reference mathematical vector to the mathematical vector of the first set of minutiae points and the mathematical vector of the another set of minutiae points respectively; and

identifying unique and duplicate minutiae points between the first set of minutiae points and the another set of minutiae points.

7. The method of knowledge learning for fingerprint verifications of claim 5 wherein,

the combined minutiae point record is represented by a mathematical vector form and representing all identified minutiae points from the first set of fingerprint image data and another set of fingerprint image data, wherein all the identified minutiae points are unique.

8. The method of knowledge learning for fingerprint verifications of claim 5 further comprises,

in response of the operation of the fingerprint scanner to scan another set of fingerprints, perform further steps including:

performing a comparison of the information of the minutiae points in the combined minutiae point record with the plurality of master minutiae point records to determine a success of the comparison, wherein

the success is determined when quantity of same minutiae points between the combined minutiae points and minutiae points within a specific master minutiae record exceeds a predetermined quantity requirement;

when the determination is a success, generating a report of success and terminate the verification; and

when the determination is not a success, determining quantity of scanning performed, wherein

when the quantity of scanning equals to a predetermined trial requirement, terminating the verification; and

when the quantity of scanning is less than the predetermined trial requirement, repeating all of the operations including;

i) scanning another set of fingerprint;

ii) transmitting the another set of fingerprint image data to the FID and store the another set of fingerprint image data in the FID;

iii) identifications of another set of minutiae points by the MPEFM from the another set of fingerprint image data;

iv) a fusion process to create a combined minutiae point record, wherein the combined minutiae point record includes information of the minutiae points from the first set of fingerprints and the minutiae points from another set of fingerprints;

v) applying a reference mathematical vector to the mathematical vector of the first set of minutiae points and the mathematical vector of the another set of minutiae points respectively;

vi) identifying unique and duplicate minutiae points between the first set of minutiae points and the another set of minutiae points; and

vii) performing a comparison of the information of the minutiae points in the combined minutiae point record with the plurality of master minutiae point records to determine a success of the comparison.

9. A knowledge learning system for fingerprint verifications comprises,

a fingerprint scanner;

a fingerprint image database (FID);

a minutiae point extraction and fusion module (MPEFM);

a minutiae point database (MPD);

a master minutiae point database (MMPD) wherein the MMPD comprises a plurality of master minutiae point records; and

a verification module (VM).

10. The knowledge learning system for fingerprint verifications of claim 9 wherein,

the fingerprint scanner to scans first set of fingerprints and converts the scanned first set of fingerprints into image data;

the fingerprint scanner transmits the first set of fingerprint image data to the FID and stores the first set of fingerprint image data in the FID;

the MPEFM identifies first set of minutiae points from the first set of fingerprint image data; and

the MPD stores information of the first set of minutiae points, wherein the information of the first set of minutiae points is represented by a mathematical vector form.

11. The knowledge learning system for fingerprint verifications of claim 10, wherein

the mathematical vector form includes information of a x-coordinate, a y-coordinate, and an angle in view of a selected reference Cartesian coordinate system.

12. The knowledge learning system for fingerprint verifications of claim 10, wherein

the VM performs a comparison of the information of the first set of minutiae points with the plurality of master minutiae point records to determine a success of the comparison, wherein

the success is determined when quantity of same minutiae points between the first set of minutiae points and minutiae points within a specific master minutiae record exceeds a predetermined quantity requirement;

and

when the determination is a success, generating a report of success and terminating the verification.

13. The knowledge learning system for fingerprint verifications of claim 12, wherein

the fingerprint scanner scans another set of fingerprints and converts the scanned another set of fingerprints into image data only when the comparison is not a success;

in response of the scanning another set of fingerprints, the scanner transmits the another set of fingerprint image data to the FID and stores the another set of fingerprint image data in the FID;

the MPEFM identifies another set of minutiae points from the another set of fingerprint image data;

the MPEFM performs a fusion process to create a combined minutiae point record, wherein

the combined minutiae point record includes information of the minutiae points from the first set of fingerprints and the minutiae points from another set of fingerprints.

14. The knowledge learning system for fingerprint verifications of claim 13, wherein

the MPEFM performs the fusion process by applying a reference mathematical vector to the mathematical vector of the first set of minutiae points and the mathematical vector of the another set of minutiae points respectively;

and

the MPEFM identifies unique and duplicate minutiae points between the first set of minutiae points and the another set of minutiae points.

15. The knowledge learning system for fingerprint verifications of claim 13, wherein

the combined minutiae point record is represented by a mathematical vector form and representing all identified minutiae points from the first set of fingerprint image data and another set of fingerprint image data, wherein the all identified minutiae points are unique with no duplicates.

16. The knowledge learning system for fingerprint verifications of claim 13, wherein

in response of the scanning of another set of fingerprints, the VM performs a comparison of the information of the minutiae points in the combined minutiae point record with the plurality of master minutiae point records to determine a success of the comparison, wherein

the success is determined when quantity of same minutiae points between the combined minutiae points and minutiae points within a specific master minutiae record exceeds a predetermined quantity requirement;

when the determination is a success, generating a report of success and terminate the verification; and

when the determination is not a success, determining quantity of scanning performed, wherein

when the quantity of scanning equals to a predetermined trial requirement, terminating the verification; and

when the quantity of scanning is less than the predetermined trial requirement, repeating all of the operations including;

i) scanning another set of fingerprint;

ii) transmitting the another set of fingerprint image data to the FID and store the another set of fingerprint image data in the FID;

iii) identifications of another set of minutiae points by the MPEFM from the another set of fingerprint image data;

iv) a fusion process to create a combined minutiae point record, wherein the combined minutiae point record includes information of the minutiae points from the first set of fingerprints and the minutiae points from another set of fingerprints;

v) applying a reference mathematical vector to the mathematical vector of the first set of minutiae points and the mathematical vector of the another set of minutiae points respectively;

vi) identifying unique and duplicate minutiae points between the first set of minutiae points and the another set of minutiae points; and

vii) performing a comparison of the information of the minutiae points in the combined minutiae point record with the plurality of master minutiae point records to determine a success of the comparison.