Patent US10719692

Vein matching for difficult biometric authentication cases
[+] Show Details

US10719692

1. A method, comprising: (7) (10)

obtaining one or more first images of a face of a user using a camera located on a device, the device comprising a computer processor and a memory;

encoding the first image to generate at least one first feature vector, wherein the first feature vector represents one or more facial features of the user in the encoded first image;

generating a first matching score by comparing the at least one first feature vector to a plurality of reference templates stored in the memory of the device;

comparing the first matching score to a first unlock threshold for the device;

in response to the first matching score being above the first unlock threshold, comparing the first matching score to a second unlock threshold for the device, the second unlock threshold being above the first unlock threshold;

automatically obtaining one or more second images of the face of the user using the camera without prompting the user, wherein the camera is activated to obtain the second images, by the computer processor, in response to the first matching score being above the first unlock threshold and below the second unlock threshold;

encoding at least one of the second images to generate at least one second feature vector, wherein the second feature vector represents at least one vein feature of the user in the encoded second image;

generating a second matching score by comparing the at least one second feature vector to at least one vein reference template for the at least one region of the face of the user, the at least one vein reference template being stored in the memory of the device;

comparing the second matching score to a third unlock threshold for the device; and

authenticating the user to use the device in response to the second matching score being above the third unlock threshold.

2. The method of claim 1, wherein encoding the at least one vein feature, generating the second matching score, and authenticating the user in response to the second matching score being above the third unlock threshold are performed by the device in response to input provided by an authorized user of the device during an enrollment process on the device. (0)

3. The method of claim 1, wherein the at least one image used to generate the second matching score comprises an image captured at a different wavelength from the at least one image used to generate the first matching score. (0)

4. The method of claim 1, wherein the third unlock threshold is a threshold for matching of the at least one vein feature to at least one vein reference template. (0)

5. The method of claim 1, wherein the at least one region with the at least one vein feature is one of a forehead, an eye, and/or a cheek of the user. (0)

6. The method of claim 1, wherein the reference templates comprise feature vectors generated for an authorized user of the device. (0)

7. The method of claim 1, wherein the at least one vein reference template comprises feature vectors generated for an authorized user of the device and/or a pattern of veins unique to the authorized user. (0)

8. The method of claim 1, further comprising prompting the user to use an additional authentication protocol for authentication in response to the second matching score being below the third unlock threshold. (0)

9. A device, comprising: (4) (5)

a computer processor;

a memory;

a camera;

at least one illuminator providing infrared illumination;

circuitry coupled to the camera and the illuminator, wherein the circuitry is configured to: obtain a plurality of first images of a face of a user using the camera;

encode at least one image selected from the first images to generate at least one first feature vector, wherein the first feature vector represents one or more facial features of the user in the encoded first image;

generate a first matching score by comparing the at least one first feature vector to a plurality of reference templates stored in the memory of the device;

compare the first matching score to a first unlock threshold for the device and a second unlock threshold for the device, the first unlock threshold being below the second unlock threshold;

wherein: in response to the first matching score being above the second unlock threshold, the user is authenticated to use the device;

in response to the first matching score being below the second unlock threshold and above the first unlock threshold, the circuitry is configured to automatically, without prompting the user, proceed to: encode at least one additional image selected from the first images to generate at least one second feature vector, wherein the second feature vector represents at least one vein feature of the user in the encoded first image;

generate a second matching score by comparing the at least one second feature vector to at least one vein reference template for the at least one region of the face of the user, the at least one vein reference template being stored in the memory of the device;

compare the second matching score to a third unlock threshold; and

authenticate the user to use the device in response to the second matching score being above the third unlock threshold.

10. The device of claim 9, wherein the camera comprises an infrared sensor. (0)

11. The device of claim 9, wherein the camera comprises an infrared sensor configured to detect sub epidermal blood vessels. (0)

12. The device of claim 9, wherein the at least one illuminator comprises a flood infrared illuminator and a speckle pattern infrared illuminator. (0)

13. The device of claim 9, wherein at least one illuminator provides at least some illumination that penetrates into a sub epidermal layer of the face of the user. (0)

14. A method, comprising: (5) (2)

determining whether a vein matching operation is to be operated on one or more images of a face of a user captured using a camera located on a device, the device comprising a computer processor and a memory;

wherein determining whether the vein matching operation is to be operated comprises: generating a matching score by comparing a first feature vector for at least one feature in the face of the user in at least one captured image to at least one feature vector from a plurality of reference templates stored in the memory of the device for an authorized user of the device; and

automatically activating without prompting the user, by the computer processor, the vein matching operation in response to determining that the matching score is between a first unlock threshold and a second unlock threshold; wherein the vein matching operation comprises:

comparing a feature vector for at least one vein in the face of the user in at least one additional captured image to at least one feature vector from a plurality of reference templates of vein features stored in the memory of the device for the authorized user of the device, wherein the at least one additional captured image is captured in response to activation of the vein matching operation;

determining whether the face of the user in the at least one additional captured image is the authorized user for the device based on the comparison of the feature vector for the at least one vein and the at least one feature vector from the plurality of reference templates; and

authenticating the user to use the device in response to determining the face of the user in the at least one additional captured image is the authorized user for the device.

15. The method of claim 14, further comprising generating the feature vector for the at least one vein in the face of the user by encoding the additional captured image to generate the feature vector for the at least one vein. (0)

16. The method of claim 14, wherein determining whether the face of the user in the additional captured image is the authorized user for the device comprises: (1) (2)

18. The method of claim 14, further comprising authenticating the user to use the device when the matching score is above the second unlock threshold, the second unlock threshold being higher than the first unlock threshold. (0)

19. The method of claim 14, further comprising generating the first feature vector by encoding the at least one captured image. (0)

20. The method of claim 14, further comprising generating the feature vector for the vein in the face of the user by encoding the at least one additional captured image. (0)

View Abstract and Specification Size

PDF with Images and Document Face >

Full Text Publication >

Patent US10719692

Vein matching for difficult biometric authentication cases

Patent Matrix® Search