Multi-Resolutional Texture Analysis Fingerprint Liveness Systems and Methods

• US 20080253625A1
• Filed: 10/10/2007
• Published: 10/16/2008
• Est. Priority Date: 10/10/2006
• Status: Active Grant

• Current Assignee: Precise Biometrics AB (Precise Biometrics Incorporated)
• Original Assignee: West Virginia University Research Corp., Wvu Office Of Technology Transfer & Wvu Business Incubator
• Inventors: Hornak, Lawrence, Schuckers, Stephanie, Abhyankar, Aditya
• Granted Patent: US 8,090,163 B2
• Time in Patent Office: Days
• US Class Current: 382/125
• CPC Class Codes: G06V 40/1376 Matching features related to ridge properties or fingerprint texture
• CPC Class Codes: G06V 40/1382 Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger
• CPC Class Codes: G06V 40/1388 using image processing

Abstract

Various systems, methods, and programs embodied in computer-readable mediums are provided for fingerprint liveness detection. In one embodiment, a method for determining fingerprint liveness is provided that comprises receiving a plurality of image analysis data of a fingerprint image; condensing the plurality of image analysis data; and determining liveness of the fingerprint image based upon the condensed data.

Claims

1. A method for determining fingerprint liveness, comprising the steps of:

receiving a plurality of image analysis data of a fingerprint image;

condensing the plurality of image analysis data; and

determining liveness of the fingerprint image based upon the condensed data.

2. The method of claim 1, wherein the plurality of image analysis data comprises multi-resolution texture analysis data.

3. The method of claim 1, wherein condensing the plurality of image analysis data comprises principal component analysis.

4. The method of claim 1, wherein the step of determining liveness of the fingerprint image further comprises the steps of:

processing the condensed data to determine a location within a cluster;

determining the distance between the condensed data location and a centroid of the cluster; and

determining liveness of the fingerprint image based upon the distance and a predefined threshold.

5. The method of claim 4, wherein the step of processing the condensed data to determine a location within a cluster is performed by fussy c-means clustering.

6. The method of claim 1, further comprising the steps of:

receiving a fingerprint image;

generating a histogram of the fingerprint image; and

determining a multi-resolution texture analysis data from the histogram.

7. The method of claim 6, wherein multi-resolution texture analysis data comprises a first order function.

8. The method of claim 1, further comprising the step of:

receiving a fingerprint image;

generating a co-occurrence matrix of the fingerprint image; and

determining a multi-resolution texture analysis data from the co-occurrence matrix.

9. The method of claim 1, further comprising the step of:

receiving a fingerprint image;

determining a local-ridge frequency of the fingerprint image; and

determining a local-ridge statistical data from the local-ridge frequency.

10. A system, comprising:

a processor circuit having a processor and a memory;

a detection system stored in the memory and executable by the processor, the detection system comprising;

logic that receives a plurality of image analysis data of a fingerprint image;

logic that condenses the plurality of image analysis data; and

logic that determines liveness of the fingerprint image based upon the condensed data.

11. The system of claim 10, wherein the logic that determines liveness of the fingerprint image further comprises:

logic that processes the condensed data to determine a location within a cluster;

logic that determines the distance between the condensed data location and a centroid of the cluster; and

logic that determines liveness of the fingerprint image based upon the distance and a predefined threshold.

12. The system of claim 10, wherein the detection system further comprises:

logic that receives a fingerprint image;

logic that generates a histogram of the fingerprint image;

logic that generates a co-occurrence matrix of the fingerprint image; and

logic that determines a multi-resolution texture analysis data from the histogram and the co-occurrence matrix.

13. The system of claim 10, wherein the detection system further comprises:

logic that receives a fingerprint image;

logic that determines a local-ridge frequency of the fingerprint image; and

logic that determines a local-ridge statistical data from the local-ridge frequency.

14. A system, comprising:

means for receiving a plurality of image analysis data of a fingerprint image;

means for condensing the plurality of image analysis data; and

means for determining liveness of the fingerprint image based upon the condensed data.

15. The system of claim 14, wherein the means for determining liveness of the fingerprint image further comprises:

means for processing the condensed data to determine a location within a cluster;

means for determining the distance between the condensed data location and a centroid of the cluster; and

means for determining liveness of the fingerprint image based upon the distance and a predefined threshold.

16. The system of claim 14, wherein the system further comprises:

means for receiving a fingerprint image;

means for generating a co-occurrence matrix of the fingerprint image; and

means for determining a multi-resolution texture analysis data from the co-occurrence matrix.

17. A method for determining fingerprint liveness, comprising the steps of:

receiving first and second images of a fingerprint, wherein the second image is obtained a predetermined time after the first image;

detecting a singularity point;

determining a time domain signal for the singularity point using data from the first and second images;

decomposing the time domain signal using a wavelet packet transform;

determining an entropy of the decomposed signals;

sifting the decomposed signals; and

determining liveness of the fingerprint images based upon the entropy of the sifted signals.

18. The method of claim 17, further comprising the steps of:

detecting a plurality of singularity points;

determining the number of singularity points in the plurality; and

determining that the fingerprint images are not live based upon the number of singularity points and a predetermined threshold.

19. The method of claim 17, wherein the step of determining a singularity point further comprises the steps of:

detecting a first singularity point of the first image, wherein the first singularity point is at a fixed pixel location;

detecting a second singularity point of the second image, wherein the second singularity point is at the fixed pixel location; and

matching the first singularity point to the second singularity point.

20. The method of claim 17, wherein the step of detecting a singularity point comprises performing wavelet analysis of fingerprint image pixels.

21. The method of claim 17, wherein the step of determining a time domain signal comprises B-spline interpolation of pixel data from the first and second images.

22. The method of claim 17, wherein the step of determining liveness of the fingerprint image comprises determining that the fingerprint images are live based upon the entropy of the sifted signals and a predetermined threshold.

23. A system, comprising:

a processor circuit having a processor and a memory;

a detection system stored in the memory and executable by the processor, the detection system comprising;

logic that receives first and second images of a fingerprint, wherein the second image is obtained a predetermined time after the first image;

logic that detects a singularity point;

logic that determines a time domain signal for the singularity point using data from the first and second images;

logic that decomposes the time domain signal using a wavelet packet transform;

logic that determines an entropy of the decomposed signals;

logic that sifts the decomposed signals; and

logic that determines liveness of the fingerprint images based upon the entropy of the sifted signals.

24. The method of claim 23, wherein the logic that detects a singularity point further comprises:

logic that detects a first singularity point of the first image, wherein the first singularity point is at a fixed pixel location;

logic that detects a second singularity point of the second image, wherein the second singularity point is at the fixed pixel location; and

logic that matches the first singularity point to the second singularity point.

25. The system of claim 23, wherein the detection system further comprises:

logic that detects a plurality of singularity points;

logic that determines the number of singularity points in the plurality; and

logic that determines that the fingerprint images are not live based upon the number of singularity points and a predetermined threshold.