Systems and methods for improving spoof detection based on matcher alignment information

Patent metadata

Field	Value
Publication number	US-10121054-B2
Application number	US-201615388908-A
Country	US
Kind code	B2
Filing date	Dec 22, 2016
Priority date	Nov 10, 2016
Publication date	Nov 6, 2018
Grant date	Nov 6, 2018

Abstract

Disclosed is a system and method for performing spoof detection. The method includes: receiving, by a processor from a biometric sensor, an input image of a biometric; obtaining, by the processor, alignment information that aligns the input image to an enrollment image; determining, by the processor, an overlap region and a non-overlap region of the input image relative to the enrollment image; extracting, by the processor, one or more anti-spoof features from the input image based on one or more of the overlap region and the non-overlap region; and, determining, by the processor, whether the input image is a replica of the biometric based on the one or more anti-spoof features.

First claim

What is claimed is:

1. An electronic device, comprising:
a biometric sensor; and
a processor configured to:
receive, from the biometric sensor, an input image of a biometric;
obtaining alignment information that aligns the input image to an enrollment image of the biometric;
determine an overlap region and a non-overlap region of the input image relative to the enrollment image based on the alignment information that aligns the input image to the enrollment image;
extract one or more anti-spoof features from the input image based on one or more of the overlap region and the non-overlap region; and
determine whether the input image is a replica of the biometric based on the one or more anti-spoof features.

2. The electronic device of claim 1 , wherein the alignment information comprises transformation information that, when applied to the input image, translates and rotates the input image to align the input image with an enrollment image.

3. The electronic device of claim 1 , wherein the alignment information comprises a mask or coordinate information that identifies the overlap region and the non-overlap region.

4. The electronic device of claim 1 , wherein the one or more anti-spoof features comprise at least one of a number of keypoints and a number of minutiae in the non-overlap region.

5. The electronic device of claim 1 , wherein the one or more anti-spoof features comprise at least one of a number of keypoints and a number of minutiae in the overlap region.

6. The electronic device of claim 1 , wherein determining whether the input image is a replica of the biometric is based on applying a first weight to the anti-spoof features from the overlap region and applying a second weight to the anti-spoof features from the non-overlap region.

7. The electronic device of claim 6 , wherein the second weight is greater than the first weight.

8. The electronic device of claim 1 , wherein extracting the one or more anti-spoof features from the input image is based on a differential between anti-spoof features in the overlap region of the input image and anti-spoof features in a corresponding portion of an anti-spoof template.

9. The electronic device of claim 1 , wherein extracting the one or more anti-spoof features from the input image is based on a differential between anti-spoof features in the non-overlap region of the input image and anti-spoof features in a corresponding portion of an anti-spoof template.

10. The electronic device of claim 1 , wherein the biometric comprises a fingerprint of a finger, and the replica comprises a gelatin mold, a graphite mold, or a wood glue mold of the fingerprint of the finger.

11. A method for performing spoof detection, comprising:
receiving, by a processor from a biometric sensor, an input image of a biometric;
obtaining, by the processor, alignment information that aligns the input image to an enrollment image of the biometric;
determining, by the processor, an overlap region and a non-overlap region of the input image relative to the enrollment image based on the alignment information that aligns the input image to the enrollment image;
extracting, by the processor, one or more anti-spoof features from the input image based on one or more of the overlap region and the non-overlap region; and
determining, by the processor, whether the input image is a replica of the biometric based on the one or more anti-spoof features.

12. The method of claim 11 , wherein the alignment information comprises transformation information that, when applied to the input image, translates and rotates the input image to align the input image with an enrollment image.

13. The method of claim 11 , wherein the alignment information comprises a mask or coordinate information that identifies the overlap region and the non-overlap region.

14. The method of claim 11 , wherein the one or more anti-spoof features comprise at least one of a number of keypoints and a number of minutiae in the non-overlap region.

15. The method of claim 11 , wherein the one or more anti-spoof features comprise at least one of a number of keypoints and a number of minutiae in the overlap region.

16. The method of claim 11 , wherein determining whether the input image is a replica of the biometric is based on applying a first weight to the anti-spoof features from the overlap region and applying a second weight to the anti-spoof features from the non-overlap region.

17. The method of claim 16 , wherein the second weight is greater than the first weight.

18. The method of claim 11 , wherein extracting the one or more anti-spoof features from the input image is based on a differential between anti-spoof features in the overlap region of the input image and anti-spoof features in a corresponding portion of an anti-spoof template.

19. The method of claim 11 , wherein extracting the one or more anti-spoof features from the input image is based on a differential between anti-spoof features in the non-overlap region of the input image and anti-spoof features in a corresponding portion of an anti-spoof template.

20. The method of claim 11 , wherein the biometric comprises a fingerprint of a finger, and the replica comprises a gelatin mold, a graphite mold, or a wood glue mold of the fingerprint of the finger.

21. A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, causes a computing device to perform spoof detection, by performing steps comprising:
receiving, from a biometric sensor, an input image of a biometric;
obtaining alignment information that aligns the input image to an enrollment image of the biometric;
determining an overlap region and a non-overlap region of the input image relative to the enrollment image based on the alignment information that aligns the input image to the enrollment image;
extracting one or more anti-spoof features from the input image based on one or more of the overlap region and the non-overlap region; and
determining whether the input image is a replica of the biometric based on the one or more anti-spoof features.

22. The computer-readable storage medium of claim 21 , wherein obtaining the alignment information that aligns the input image to the enrollment image comprises receiving the alignment information from a matcher module.

Assignees

• Synaptics Inc

Inventors

• Krishnapura Rohini
• Russo Anthony P

Classifications

• G06K9/00899

  Physics · mapped topic

• G06K9/00087 Primary

  Physics · mapped topic

• G06V40/40

  Spoof detection, e.g. liveness detection · CPC title

• G06V40/1365 Primary

  Matching; Classification · CPC title