Fingerprint recognition for low computing power applications

• US 8,908,934 B2
• Filed: 03/13/2008
• Issued: 12/09/2014
• Est. Priority Date: 03/14/2007
• Status: Expired due to Fees

• Current Assignee: IVI Holdings Limited
• Original Assignee: IVI Holdings Limited
• Inventors: Saito, Tamio
• Primary Examiner(s): Bella, Matthew
• Assistant Examiner(s): Rosario, Dennis
• Application Number:
  US12/048,162
• Publication Number:
  US 20100215224A1
• Time in Patent Office: 2,462 Days
• Field of Search: 382/124, 382/243
• US Class Current: 382/124
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction
• CPC Class Codes: G06V 40/1353 Extracting features related to minutiae or pores
• CPC Class Codes: G06V 40/1371 Matching features related to minutiae or pores

Abstract

A method and apparatus are provided for improved fingerprint recognition. A fingerprint image associated with an authorized user can be captured with a sensor. Direction vectors for a plurality of pixels of the first fingerprint image are then calculated. Next, characteristic points of the fingerprint image are located in a first coordinate system. The fingerprint central nucleus is determined based on the calculated direction vectors, and is expressed in the first coordinate system. Subsequently, a second coordinate system having an origin located relative to the fingerprint central nucleus is determined. Thereafter, the characteristic points of the fingerprint image are mapped to the second coordinate system. Finally, the fingerprint central nucleus and fingerprint characteristic points expressed in the second coordinate system are saved as a fingerprint template.

Claims

1. A method for fingerprint recognition, comprising:

capturing a first fingerprint image with a sensor;

calculating direction vectors for a plurality of pixels of the first fingerprint image;

locating characteristic points of the first fingerprint image in a first coordinate system;

determining that a fingerprint central nucleus of the first fingerprint image is a center point of adjacent pixels by calculating a minimum radius from a plurality of radii of curvature based on the direction vectors, the adjacent pixels being adjacent to the minimum radius,

wherein the fingerprint central nucleus is expressed in the first coordinate system;

determining a second coordinate system so that its origin is located relative to the fingerprint central nucleus of the first fingerprint image;

mapping the characteristic points of the first fingerprint image to the second coordinate system; and

saving the fingerprint central nucleus and fingerprint characteristic points of the first fingerprint image expressed in the second coordinate system as a fingerprint template.

2. The method of claim 1, wherein the first fingerprint image is associated with an authorized user.

3. The method of claim 1, wherein the sensor is a biometric sensor configured to read a fingerprint of a user.

4. The method of claim 1, wherein after capturing the fingerprint image with the sensor, the quality of the fingerprint image is checked, and, if the quality is poor, image noise is automatically removed.

5. The method of claim 1, wherein the second coordinate system is a polar coordinate system representing an angle and a distance from its origin relative to the fingerprint central nucleus.

6. The method of claim 1, further comprising:

capturing a second fingerprint image with a sensor;

calculating direction vectors for a plurality of pixels of the second fingerprint image;

locating characteristic points of the second fingerprint image in the first coordinate system;

determining a fingerprint central nucleus of the second fingerprint image based on the direction vectors of the second fingerprint image, wherein the fingerprint central nucleus of the second fingerprint image is expressed in the first coordinate system;

mapping the characteristic points and the fingerprint central nucleus of the second fingerprint image to the second coordinate system; and

comparing the characteristic points and the fingerprint central nucleus of the second fingerprint image to the stored fingerprint template.

7. The method of claim 6, wherein a fingerprint match is determined if the characteristic points and the fingerprint central nucleus of the second fingerprint image match the stored fingerprint template.

8. The method of claim 1, wherein the second coordinate system is a multi-dimensional coordinate system having at least three dimensions.

9. The method of claim 8, wherein one of the at least three dimensions indicates a pixel gradient intensity.

10. A system for fingerprint recognition, the system comprising:

a fingerprint sensor configured to capture a first fingerprint image;

a processor configured to;

(a) calculate direction vectors for a plurality of pixels of the first fingerprint image;

(b) locate characteristic points of the first fingerprint image in a first coordinate system;

(c) determine that a fingerprint central nucleus is a center point of adjacent pixels by calculating a minimum radius from a plurality of radii of curvature based on the direction vectors the adjacent pixels being adjacent to the minimum radius;

wherein the fingerprint central nucleus is expressed in the first coordinate system;

(d) determine a second coordinate system whose so that its origin is located relative to the first central nucleus; and

(e) map the characteristic points to the second coordinate system; and

a

temporary storage unit for storing a fingerprint template, wherein the fingerprint template comprises the fingerprint central nucleus and fingerprint characteristic points of the first fingerprint image expressed in the second coordinate system.

11. The system of claim 10 wherein the first fingerprint image is associated with an authorized user.

12. The system of claim 10 wherein the sensor is a biometric sensor configured to read a fingerprint of a user.

13. The system of claim 10 wherein the processor is configured to check the quality of the fingerprint image, and if the quality is poor, image noise is automatically removed.

14. The system of claim 10 wherein the sensor, the processing unit, and the temporary storage unit are implemented on an IC card.

15. The system of claim 10 wherein the first coordinate system is a Cartesian coordinate system.

16. The system of claim 10 wherein the second coordinate system is a polar coordinate system, a three dimensional coordinate system, a spherical coordinate system, or a cylindrical coordinate system.

17. The system of claim 10 wherein the processor is further configured to:

calculate direction vectors for a plurality of pixels of a second fingerprint image;

locate characteristic points of the second fingerprint image in the first coordinate system;

determine a fingerprint central nucleus of the second fingerprint image based on the direction vectors of the second fingerprint image, wherein the fingerprint central nucleus of the second fingerprint image is expressed in the first coordinate system;

map the characteristic points and the fingerprint central nucleus of the second fingerprint image to the second coordinate system; and

compare the characteristic points and the fingerprint central nucleus of the second fingerprint image to the fingerprint template stored in the temporary storage unit.

18. A method for fingerprint recognition, comprising:

capturing a first fingerprint image with a sensor;

calculating direction vectors for a plurality of pixels of the first fingerprint image;

locating characteristic points of the first fingerprint image in a first coordinate system;

determining that a fingerprint central nucleus is a center point of adjacent pixels by calculating a minimum radius from a plurality of radii of curvature based on the direction vectors, the adjacent pixels being adjacent to the minimum radius;

wherein the fingerprint central nucleus is expressed in the first coordinate system;

determining a second coordinate system whose so that its origin is located relative to the first central nucleus;

mapping the characteristic points of the first fingerprint image to the second coordinate system;

saving the fingerprint central nucleus and fingerprint characteristic points of the first fingerprint image expressed in the second coordinate system as fingerprint template information;

capturing a second fingerprint image with a sensor;

calculating direction vectors for a plurality of pixels of the second fingerprint image;

locating characteristic points of the second fingerprint image in the first coordinate system;

determining a fingerprint central nucleus of the second fingerprint image based on the direction vectors of the second fingerprint image;

wherein the fingerprint central nucleus of the second fingerprint image is expressed in the first coordinate system;

mapping the characteristic points and the fingerprint central nucleus of the second fingerprint image to the second coordinate system; and

comparing the characteristic points and the fingerprint central nucleus of the second fingerprint image to the stored fingerprint template information.