PREVENTING FACE-BASED AUTHENTICATION SPOOFING

• US 20160140405A1
• Filed: 09/25/2015
• Published: 05/19/2016
• Est. Priority Date: 11/13/2014
• Status: Active Grant

• Current Assignee: Tahoe Research Limited (f/k/a Learndale Limited) (Vector Capital Corporation)
• Original Assignee: Intel Corporation
• Inventors: Graumann, David L., Ghosh, Rahuldeva, Narjala, Ranjit S.
• Granted Patent: US 10,198,645 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/20182 Noise reduction or smoothing in the temporal domain; Spatio-temporal filtering
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06V 40/171 Local features and components; Facial parts eye characteristics G06V40/18; Occluding parts, e.g. glasses; Geometrical relationships
• CPC Class Codes: G06V 40/172 Classification, e.g. identification
• CPC Class Codes: G06V 40/40 Spoof detection, e.g. liveness detection
• CPC Class Codes: G09C 5/00 Ciphering apparatus or methods not provided for in the preceding groups, e.g. involving the concealment or deformation of graphic data such as designs, written or printed messages
• CPC Class Codes: H04L 9/3226 using a predetermined code, e.g. password, passphrase or PIN network architectures or network communication protocols for supporting authentication of entities using passwords in a packet data network H04L63/083
• CPC Class Codes: H04L 9/3231 Biological data, e.g. fingerprint, voice or retina network architectures or network communication protocols for supporting authentication of entities using biometrical features in a packet data network H04L63/0861

Abstract

System and techniques for preventing face-based authentication spoofing are described herein. A visible light emitter may be controlled to project a pattern into a camera'"'"'s field of view during an authentication attempt. An image may be obtained from the camera for the authentication attempt. A potential spoofing region on image may be identified by finding the pattern. An authentication attempt based on a face found in the potential spoofing region may be prevented.

Claims

1. A system to prevent face-based authentication spoofing, the system comprising:

an emitter controller to control a visible light emitter to project a pattern into a field of view for a camera during an authentication attempt using a face;

an image sampler to obtain an image from the camera for the authentication attempt; and

a spoofing detection controller to;

identify a potential spoofing region of the image by finding the pattern within predetermined tolerances in the image; and

prevent the authentication attempt based on a face found in the potential spoofing region.

2. The system of claim 1, wherein the pattern is moving.

3. The system of claim 1, wherein the pattern includes a foreground and a background.

4. The system of claim 3, wherein the background is selected to illuminate the face for the authentication attempt.

5. The system of claim 1, comprising a library controller to:

receive a user submitted pattern; and

test the user submitted pattern to determine whether it has features to distinguish it from organic features.

6. The system of claim 5, wherein the spoofing controller is arranged to include the user submitted pattern in a library of patterns when it does have features to distinguish the user submitted pattern from organic features, and rejecting the user submitted pattern otherwise.

7. The system of claim 5, wherein the library controller is arranged to:

add features that are distinguishable from organic features to the user submitted pattern to create a modified user submitted pattern; and

add the modified user submitted pattern to the library of patterns.

8. The system of claim 1, wherein to identify a potential spoofing region of the image by finding the pattern within predetermined tolerances in the image includes the spoofing controller arranged to:

find the pattern across a plurality of images, including the image;

measure a signal-to-noise ratio for a region in which the pattern was previously found, across the plurality of images; and

make the region the potential spoofing region when the signal-to-noise ratio is beyond a threshold.

9. A machine-implemented method to prevent face-based authentication spoofing, the method comprising:

controlling a visible light emitter to project a pattern into a field of view for a camera during an authentication attempt using a face;

obtaining an image from the camera for the authentication attempt;

identifying a potential spoofing region of the image by finding the pattern within predetermined tolerances in the image; and

preventing an authentication attempt based on a face found in the potential spoofing region.

10. The method of claim 9, wherein controlling the visible light emitter to project the pattern comprises projecting the pattern with motion.

11. The method of claim 9, wherein the pattern includes a foreground and a background.

12. The method of claim 11, wherein the background is selected to illuminate the face for the authentication attempt.

13. The method of claim 9, comprising:

receiving a user submitted pattern; and

testing the user submitted pattern to determine whether it has features to distinguish it from organic features.

14. The method of claim 13, comprising including the user submitted pattern in a library of patterns when it does have features to distinguish the user submitted pattern from organic features, and rejecting the user submitted pattern otherwise.

15. The method of claim 13, comprising:

adding features that are distinguishable from organic features to the user submitted pattern to create a modified user submitted pattern; and

adding the modified user submitted pattern to the library of patterns.

16. The method of claim 9, wherein identifying a potential spoofing region of the image by finding the pattern within predetermined tolerances in the image includes:

finding the pattern across a plurality of images, including the image;

measuring a signal-to-noise ratio for a region in which the pattern was previously found, across the plurality of images; and

making the region the potential spoofing region when the signal-to-noise ratio is beyond a threshold.

17. At least one machine readable medium including instructions to prevent face-based authentication spoofing, the instructions, when executed by a machine, cause the machine to perform operations comprising:

controlling a visible light emitter to project a pattern into a field of view for a camera during an authentication attempt using a face;

obtaining an image from the camera for the authentication attempt;

identifying a potential spoofing region of the image by finding the pattern within predetermined tolerances in the image; and

preventing an authentication attempt based on a face found in the potential spoofing region.

18. The at least one machine readable medium of claim 17, wherein the pattern is moving.

19. The at least one machine readable medium of claim 17, wherein the pattern includes a foreground and a background.

20. The at least one machine readable medium of claim 19, wherein the background is selected to illuminate the face for the authentication attempt.

21. The at least one machine readable medium of claim 17, comprising:

receiving a user submitted pattern; and

testing the user submitted pattern to determine whether it has features to distinguish it from organic features.

22. The at least one machine readable medium of claim 21, comprising including the user submitted pattern in a library of patterns when it does have features to distinguish it from organic features, and rejecting the user submitted pattern otherwise.

23. The at least one machine readable medium of claim 21, comprising:

adding features that are distinguishable from organic features to the user submitted pattern to create a modified user submitted pattern; and

putting the modified user submitted pattern to the library of patterns.

24. The at least one machine readable medium of claim 17, wherein identifying a potential spoofing region of the image by finding the pattern within predetermined tolerances in the image includes:

finding the pattern across a plurality of images, including the image;

measuring a signal-to-noise ratio for a region in which the pattern was previously found, across the plurality of images; and

making the region the potential spoofing region when the signal-to-noise ratio is beyond a threshold.