Facial recognition

• US 8,411,909 B1
• Filed: 06/26/2012
• Issued: 04/02/2013
• Est. Priority Date: 06/26/2012
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Zhao, Yong, Schneiderman, Henry Will, Sipe, Michael Andrew
• Primary Examiner(s): MARIAM, DANIEL G
• Application Number:
  US13/533,723
• Time in Patent Office: 280 Days
• Field of Search: 382115-118, 382/218
• US Class Current: 382/116
• CPC Class Codes: G06V 40/165 using facial parts and geometric relationships
• CPC Class Codes: G06V 40/40 Spoof detection, e.g. liveness detection

Abstract

An example method includes initiating an emission, by a computing device, of at least one light beam, receiving, from an image capture device of the computing device, an image of a face of a user, and identifying at least one representation of an eye of the user in the image. The method further includes determining, by the computing device, whether at least one reflection of light from the light beam is associated with a cornea of the representation of the eye of the user in the image. The method further includes determining, based at least in part on the first determination, whether to deny authentication of the user with respect to accessing one or more functionalities controlled by the computing device, wherein the authentication is denied independent of performing facial recognition based at least in part on the image.

Claims

1. A method comprising:

initiating an emission, by a computing device, of at least one light beam;

receiving, from an image capture device coupled to the computing device, an image of a face;

identifying at least one representation of an eye in the image of the face;

determining, by the computing device, whether at least one reflection of light from the light beam is associated with the representation of the eye in the image of the face; and

determining, based at least in part on whether the at least one reflection of light from the light beam is associated with the representation of the eye in the image of the face, whether to deny authentication of a user with respect to accessing one or more functionalities controlled by the computing device.

2. The method of claim 1, wherein the at least one reflection of light includes a glint reflected by a cornea of the representation of the eye in the image of the face.

3. The method of claim 1, wherein determining whether to deny authentication of the user comprises determining whether to deny authentication of the user prior to performing facial recognition based at least in part on the image of the face.

4. The method of claim 3,

responsive to determining that the at least one reflection of light from the light beam is not associated with the representation of the eye of the user in the image of the face:

denying authentication of the user with respect to accessing one or more functionalities controlled by the computing device; and

determining not to perform facial recognition on the image of the face.

5. The method of claim 1, further comprising:

performing, by the computing device, facial recognition based at least in part on the image of the face,

wherein determining whether to deny authentication of the user comprises determining whether to deny authentication of the user substantially concurrently with performing facial recognition based at least in part on the image of the face.

6. The method of claim 1, further comprising:

performing, by the computing device, facial recognition based at least in part on the image of the face, wherein performing the facial recognition based at least in part on the image of the face produces one or more results, and

wherein determining whether to deny authentication of the user comprises determining whether to deny authentication of the user after performing the facial recognition based at least in part on the image of the face and irrespective of the one or more results of performing the facial recognition.

7. The method of claim 1, further comprising:

receiving, from the image capture device, a reference facial image prior to initiating the emission; and

identifying at least one reference representation of an eye in the reference facial image.

8. The method of claim 7, further comprising:

comparing the at least one reflection of light to a reference reflection of light in the reference facial image, wherein the reference reflection of light is associated with the at least one reference representation of the eye in the reference facial image.

9. The method of claim 1, further comprising:

terminating the emission; and

receiving, from the image capture device, a reference facial image after terminating the emission.

10. The method of claim 1, wherein determining whether to deny authentication of the user further comprises:

when an intensity of the at least one reflection of light is less than a threshold intensity, denying authentication of the user with respect to accessing one or more functionalities controlled by the computing device.

11. The method of claim 10, further comprising:

notifying the user that access to the computing device is denied.

12. The method of claim 1, wherein determining whether to deny authentication of the user comprises:

determining an expected frequency range having an upper bound and a lower bound; and

when a frequency associated with the at least one reflection of light is greater than the upper bound or less than the lower bound of the expected frequency range, denying authentication of the user with respect to accessing one or more functionalities controlled by the computing device.

13. The method of claim 1, further comprising:

initiating a second emission, by the computing device, of at least a second light beam having a second intensity, wherein the at least one light beam of the emission has a first intensity, and wherein the first intensity and the second intensity differ by an emitted intensity coefficient.

14. The method of claim 13, wherein the image of the face comprises at least a first image of the face, the method further comprising:

receiving, from the image capture device, a second image of the face, the second image associated with the second emission; and

comparing a first reflective intensity associated with the first image of the face to a second reflective intensity associated with the second image of the face.

15. The method of claim 14, wherein determining whether to deny authentication of the user comprises:

calculating a reflective intensity coefficient based on a difference between the first reflective intensity and the second reflective intensity;

obtaining an intensity ratio at least in part by comparing the reflective intensity coefficient to the emitted intensity coefficient; and

when the intensity ratio is greater than a threshold ratio, denying authentication of the user with respect to accessing one or more functionalities controlled by the computing device.

16. The method of claim 1, further comprising:

initiating a second emission, by the computing device, of at least a second light beam having a second frequency, wherein the at least one light beam of the emission has a first frequency, and wherein the first frequency and the second frequency differ by an emitted frequency coefficient.

17. The method of claim 16, wherein the image of the face comprises at least a first image of the face, the method further comprising:

receiving, from the image capture device, a second image of the face, the second image associated with the second emission; and

comparing a first reflected frequency associated with the first image of the face and a second reflected frequency associated with the second image of the face.

18. The method of claim 17, wherein determining whether to deny authentication of the user comprises:

calculating a reflected frequency coefficient based on a difference between the first reflected frequency and the second reflected frequency;

obtaining a frequency ratio at least in part by comparing the reflected frequency coefficient to the emitted frequency coefficient; and

when the frequency ratio is greater than a threshold ratio, denying authentication of the user with respect to accessing one or more functionalities controlled by the computing device.

19. The method of claim 1, further comprising:

upon detecting that the at least one reflection of light does not exist in the image of the face;

determining not to compare the image of the face to the enrollment image; and

preventing a device communicatively coupled to the computing device from transitioning from a locked state to an unlocked state.

20. The method of claim 19, wherein the device communicatively coupled to the computing device is a door lock.

21. The method of claim 1, wherein determining whether to deny authentication of the user comprises:

determining an expected wavelength range having an upper bound and a lower bound; and

when a wavelength associated with the at least one reflection of light is greater than the upper bound or less than the lower bound, denying authentication of the user with respect to accessing one or more functionalities controlled by the computing device.

22. A computer-readable storage device encoded with instructions that, when executed, cause at least one programmable processor of a computing device to perform operations including:

initiating an emission of at least one light beam;

receiving, from an image capture device coupled to the computing device, an image of a face of a user;

identifying at least one representation of an eye of the user in the image of the face;

detecting whether at least one reflection of light associated with the light beam exists in the image of the face, wherein the at least one reflection of light is associated with the at least one representation of the eye; and

determining, based at least in part on the detecting, whether to deny authentication of the user with respect to accessing one or more functionalities controlled by the computing device.

23. A device comprising:

at least one processor;

at least one image capture device operable to capture an image of a face;

a light source operable to initiate an emission of at least one light beam

a glint detection module operable to identify at least one representation of an eye of the user in the image of the face, and determine whether at least one reflection of light from the light beam is associated with the representation of the eye of the user in the image of the face; and

a facial recognition module operable to determine whether to deny authentication of the user with respect to accessing one or more functionalities controlled by the computing device.

24. The device of claim 23, wherein the device is selected from a group consisting of:

a mobile phone, a tablet computer, a laptop computer, a netbook, a personal digital assistant, a biometric door lock, a desktop computer, a set-top box, a television, a vehicle ignition system, and a presence verification device.

25. The device of claim 23, wherein the glint detection module is further operable to:

receive, from the image capture device, a reference facial image prior to the light source initiating the emission; and

identify at least one reference representation of an eye in the reference facial image.

26. The device of claim 25, wherein the glint detection module is further operable to:

compare the at least one reflection of light to a reference reflection of light in the reference facial image, wherein the reference reflection of light is associated with the at least one reference representation of the eye in the reference facial image.

27. The device of claim 23, wherein the glint detection module is further operable to:

when an intensity of the at least one reflection of light is less than a threshold intensity, deny authentication of the user with respect to accessing one or more functionalities controlled by the computing device.

28. The device of claim 23, wherein the light source is further operable to initiate a second emission of at least a second light beam having a second intensity, wherein the at least one light beam of the emission has a first intensity, and wherein the first intensity and the second intensity differ by an emitted intensity coefficient.

29. The device of claim 28, wherein the captured image of the face comprises at least a first image of the face, and wherein the glint detection module is further operable to:

receive, from the image capture device, a second image of the face, the second image of the face associated with the second emission; and

compare a first reflective intensity associated with the first image of the face to a second reflective intensity associated with the second image of the face.

30. The device of claim 29, wherein the facial recognition module is further operable to:

calculate a reflective intensity coefficient based on a difference between the first reflective intensity and the second reflective intensity;

obtain an intensity ratio at least in part by comparing the reflective intensity coefficient to the emitted intensity coefficient; and

when the intensity ratio is greater than a threshold ratio, deny authentication of the user with respect to accessing one or more functionalities controlled by the computing device.