Systems and methods for spoof detection relative to a template instead of on an absolute scale

• US 10,430,638 B2
• Filed: 12/22/2016
• Issued: 10/01/2019
• Est. Priority Date: 11/10/2016
• Status: Active Grant

• Current Assignee: Synaptics Incorporated
• Original Assignee: Synaptics Incorporated
• Inventors: Russo, Anthony P.
• Primary Examiner(s): Patel, Jayesh A
• Application Number:
  US15/388,586
• Publication Number:
  US 20180129858A1
• Time in Patent Office: 1,013 Days
• Field of Search: None
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
• CPC Class Codes: G06V 40/1365 Matching; Classification
• 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 40/40 Spoof detection, e.g. liveness detection
• CPC Class Codes: G06V 40/50 Maintenance of biometric data or enrolment thereof

Abstract

Disclosed are 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; extracting, by the processor, one or more anti-spoof metrics from the input image; receiving, by the processor, an anti-spoof template corresponding to the biometric; for a first anti-spoof metric, computing, by the processor, a differential value between a value of the first anti-spoof metric extracted from the input image and a value of the first anti-spoof metric in the anti-spoof template; and determining, by the processor, whether the input image is a replica of the biometric based on the differential value.

Claims

1. A device, comprising:

a sensor configured to capture an input image of a biometric; and

a processor configured to;

receive, from the sensor, the input image;

extract one or more anti-spoof metrics from the input image;

receive an anti-spoof template corresponding to a biometric of a first user;

for a first anti-spoof metric, compute a first amount of a difference between a value of the first anti-spoof metric extracted from the input image and a value of the first anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user;

for a second anti-spoof metric extracted from the input image, compute a second amount of a difference between a value of the second anti-spoof metric extracted from the input image and a value of the second anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user;

determine whether the input image is a replica of the biometric of the first user based on the first amount of the difference between the value of the first anti-spoof metric extracted from the input image and the value of the first anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user and the second amount of the difference between the value of the second anti-spoof metric extracted from the input image and the value of the second anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user; and

prevent successful authentication on the device in response to determining that the input image is a replica of the biometric of the first user.

2. The device of claim 1, wherein the processor is further configured to:

receive, from the sensor, one or more enrollment images of the biometric of the first user;

extract one or more anti-spoof metrics from each of the one or more enrollment images; and

generate the anti-spoof template based on the one or more anti-spoof metrics extracted from each of the one or more enrollment images.

3. The device of claim 2, wherein the one or more enrollment images comprises at least two enrollment images, and wherein generating the anti-spoof template comprises combining corresponding anti-spoof metrics extracted from each of the at least two enrollment images.

4. The device of claim 3, wherein the combining comprises computing, for each anti-spoof metric, an average value for the anti-spoof metric.

5. The device of claim 3, wherein the combining comprises computing, for each anti-spoof metric, a median value for the anti-spoof metric.

6. The device of claim 3, wherein the combining comprises computing, for each anti-spoof metric, a range of values for the anti-spoof metric extracted from each of the at least two enrollment images.

7. The device of claim 2, wherein the one or more enrollment images of the biometric of the first user are further stored in a matcher template storage, wherein the processor is configured to determine whether the input image is a match to one or more of the enrollment images separate from determining whether the input image is a replica of the biometric of the first user.

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

9. A method for performing spoof detection, comprising:

receiving, by a processor from a sensor, an input image of a biometric;

extracting, by the processor, one or more anti-spoof metrics from the input image;

receiving, by the processor, an anti-spoof template corresponding to a biometric of a first user;

for a first anti-spoof metric, computing, by the processor, a first amount of a difference between a value of the first anti-spoof metric extracted from the input image and a value of the first anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user;

for a second anti-spoof metric extracted from the input image, computing, by the processor, a second amount of a difference between a value of the second anti-spoof metric extracted from the input image and a value of the second anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user;

determining, by the processor, whether the input image is a replica of the biometric of the first user based on the first amount of the difference between the value of the first anti-spoof metric extracted from the input image and the value of the first anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user and the second amount of the difference between the value of the second anti-spoof metric extracted from the input image and the value of the second anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user; and

preventing, by the processor, successful authentication on the device in response to determining that the input image is a replica of the biometric of the first user.

10. The method of claim 9, further comprising:

receiving, by the processor from the sensor, one or more enrollment images of the biometric of the first user;

extracting, by the processor, one or more anti-spoof metrics from each of the one or more enrollment images; and

generating, by the processor, the anti-spoof template based on the one or more anti-spoof metrics extracted from each of the one or more enrollment images.

11. The method of claim 10, wherein the one or more enrollment images comprises at least two enrollment images, and wherein generating the anti-spoof template comprises combining corresponding anti-spoof metrics extracted from each of the at least two enrollment images.

12. The method of claim 11, wherein the combining comprises computing, for each anti-spoof metric, an average value for the anti-spoof metric.

13. The method of claim 11, wherein the combining comprises computing, for each anti-spoof metric, a median value for the anti-spoof metric.

14. The method of claim 11, wherein the combining comprises computing, for each anti-spoof metric, a range of values for the anti-spoof metric extracted from each of the at least two enrollment images.

15. The method of claim 10, wherein the one or more enrollment images of the biometric of the first user are further stored in a matcher template storage, wherein the processor is configured to determine whether the input image is a match to one or more of the enrollment images separate from determining whether the input image is a replica of the biometric of the first user.

16. The method of claim 9, wherein the biometric of the first user comprises a fingerprint of a finger of the first user, and the replica comprises a gelatin mold, a graphite mold, or a wood glue mold of the fingerprint of the finger.

17. 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 sensor, an input image of a biometric;

extracting one or more anti-spoof metrics from the input image;

receiving an anti-spoof template corresponding to a biometric of a first user;

for a first anti-spoof metric, computing a first amount of a difference between a value of the first anti-spoof metric extracted from the input image and a value of the first anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user;

for a second anti-spoof metric extracted from the input image, computing a second amount of a difference between a value of the second anti-spoof metric extracted from the input image and a value of the second anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user;

determining whether the input image is a replica of the biometric of the first user based on the first amount of the difference between the value of the first anti-spoof metric extracted from the input image and the value of the first anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user and the second amount of the difference between the value of the second anti-spoof metric extracted from the input image and the value of the second anti-spoof metric in the anti-spoof template that corresponds to the biometric of the first user; and

preventing successful authentication on the device in response to determining that the input image is a replica of the biometric of the first user.

18. The computer-readable storage medium of claim 17, wherein one or more enrollment images of the biometric of the first user are further stored in a matcher template storage, and wherein the processor is configured determine whether the input image is a match to one or more of the enrollment images separate from determining whether the input image is a replica of the biometric of the first user.