Authentication based on heartbeat detection and facial recognition in video data

• US 10,268,910 B1
• Filed: 03/13/2018
• Issued: 04/23/2019
• Est. Priority Date: 11/17/2015
• Status: Active Grant

• Current Assignee: United Services Automobile Association
• Original Assignee: United Services Automobile Association
• Inventors: Medina, III, Reynaldo, Chavez, Carlos, Sridaran, Nandhini, Colwell, Russell, Liu, Yuesheng, Briscoe, Storme Shelton
• Primary Examiner(s): Tucker, Wesley J
• Application Number:
  US15/920,136
• Time in Patent Office: 406 Days
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06V 40/15 Biometric patterns based on physiological signals, e.g. heartbeat, blood flow
• CPC Class Codes: G06V 40/172 Classification, e.g. identification
• CPC Class Codes: G06V 40/45 Detection of the body part being alive
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques

Abstract

Techniques are described for authenticating a user to access information through an application executing on a computing device. Multiple authentication methods may be used in combination to authenticate the user with greater confidence than authentication provided by a single method, and to verify that the user is a live person and not an image or video. Facial recognition may be used in conjunction with heartbeat detection, via video analysis, to verify the user'"'"'s identity and confirm that the user is live. Facial recognition may include capturing an image of the user'"'"'s face and comparing certain points on the face with previously gathered information regarding the user. Heartbeat detection may include capturing a video segment of the user, stabilizing the captured video data, applying motion magnification techniques to the stabilized video data, and analyzing the stabilized, magnified video data to determine a presence of the user'"'"'s heartbeat.

Claims

1. A computer-implemented method performed by at least one processor, the method comprising:

accessing, by the at least one processor, video data of a user;

accessing, by the at least one processor, at least one image of the user;

comparing, by the at least one processor, the at least one image to an aggregate image of the user that is developed based on previously captured images of the user;

analyzing, by the at least one processor, the video data to attempt to detect a pulse of the user;

determining, by the at least one processor, a score based on a plurality of weights including a first weight that corresponds to detecting the pulse of the user and a second weight that corresponds to determining that the at least one image corresponds to the aggregate image, wherein the first weight is different than the second weight, and wherein the second weight is based at least partly on a number of the previously captured images used to develop the aggregate image of the user; and

providing, by the at least one processor, access to secure information based at least partly on determining that the score is at least a threshold score.

2. The method of claim 1, wherein the at least one image includes at least a portion of at least one frame of the video data.

3. The method of claim 1, wherein analyzing the video data to detect the pulse further comprises performing motion stabilization on the video data.

4. The method of claim 1, wherein analyzing the video data to detect the pulse further comprises magnifying the video data.

5. The method of claim 1, wherein the threshold score is based at least partly on an account balance of the user.

6. The method of claim 1, wherein the threshold score is based at least partly on a risk level of the secure information to be accessed.

7. The method of claim 1, wherein the plurality of weights further include one or more weights corresponding to one or more of:

determining that an eye color of the user in the at least one image corresponds to a previously determined eye color of the user;

determining that a personal identification number (PIN) entered by the user corresponds to a valid PIN for the user;

determining that an answer provided by the user in response to a knowledge-based authentication (KBA) question corresponds to a stored answer associated with the user;

determining that an identifier of a user device of the user corresponds to a stored device identifier associated with the user;

detecting that the user has entered, through the user device, a one-time passcode previously communicated to the user; and

determining that recorded audio data of the user speaking at least partly corresponds to the stored voice print associated with the user.

8. A system comprising:

at least one processor; and

memory communicatively coupled to the at least one processor, the memory storing instructions which, when executed, cause the at least one processor to perform operations comprising;

accessing video data of a user;

accessing at least one image of the user;

comparing the at least one image to an aggregate image of the user that is developed based on previously captured images of the user;

analyzing the video data to attempt to detect a pulse of the user;

determining a score based on a plurality of weights including a first weight that corresponds to detecting the pulse of the user and a second weight that corresponds to determining that the at least one image corresponds to the aggregate image, wherein the first weight is different than the second weight, and wherein the second weight is based at least partly on a number of the previously captured images used to develop the aggregate image of the user; and

providing access to secure information based at least partly on determining that the score is at least a threshold score.

9. The system of claim 8, wherein the at least one image includes at least a portion of at least one frame of the video data.

10. The system of claim 8, wherein analyzing the video data to detect the pulse further comprises performing motion stabilization on the video data.

11. The system of claim 8, wherein analyzing the video data to detect the pulse further comprises magnifying the video data.

12. The system of claim 8, wherein the threshold score is based at least partly on an account balance of the user.

13. The system of claim 8, wherein the threshold score is based at least partly on a risk level of the secure information to be accessed.

14. The system of claim 8, wherein the plurality of weights further include one or more weights corresponding to one or more of:

determining that an eye color of the user in the at least one image corresponds to a previously determined eye color of the user;

determining that a personal identification number (PIN) entered by the user corresponds to a valid PIN for the user;

determining that an answer provided by the user in response to a knowledge-based authentication (KBA) question corresponds to a stored answer associated with the user;

determining that an identifier of a user device of the user corresponds to a stored device identifier associated with the user;

detecting that the user has entered, through the user device, a one-time passcode previously communicated to the user; and

determining that recorded audio data of the user speaking at least partly corresponds to the stored voice print associated with the user.

15. One or more non-transitory computer-readable storage media storing instructions which, when executed, cause at least one processor to perform operations comprising:

accessing video data of a user;

accessing at least one image of the user;

comparing the at least one image to an aggregate image of the user that is developed based on previously captured images of the user;

analyzing the video data to attempt to detect a pulse of the user;

determining a score based on a plurality of weights including a first weight that corresponds to detecting the pulse of the user and a second weight that corresponds to determining that the at least one image corresponds to the aggregate image, wherein the first weight is different than the second weight, and wherein the second weight is based at least partly on a number of the previously captured images used to develop the aggregate image of the user; and

providing access to secure information based at least partly on determining that the score is at least a threshold score.

16. The one or more non-transitory computer-readable storage media of claim 15, wherein the at least one image includes at least a portion of at least one frame of the video data.

17. The one or more non-transitory computer-readable storage media of claim 15, wherein analyzing the video data to detect the pulse further comprises performing motion stabilization on the video data.

18. The one or more non-transitory computer-readable storage media of claim 15, wherein analyzing the video data to detect the pulse further comprises magnifying the video data.

19. The one or more non-transitory computer-readable storage media of claim 15, wherein the threshold score is based at least in part on an account balance of the user.

20. The one or more non-transitory computer-readable storage media of claim 15, wherein the threshold score is based at least partly on a risk level of the secure information to be accessed.