Fraud detection for facial recognition systems

• US 9,836,642 B1
• Filed: 01/04/2016
• Issued: 12/05/2017
• Est. Priority Date: 12/18/2012
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Ramaswamy, Sharadh
• Primary Examiner(s): Chawan, Sheela C
• Application Number:
  US14/987,644
• Time in Patent Office: 701 Days
• Field of Search: 382103, 382116, 382118, 382115, 382154
• CPC Class Codes: G06V 40/162 using pixel segmentation or colour matching
• 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 40/193 Preprocessing; Feature extraction
• CPC Class Codes: G06V 40/45 Detection of the body part being alive
• CPC Class Codes: H04N 23/611 where the recognised objects include parts of the human body
• CPC Class Codes: H04N 7/183 for receiving images from a single remote source

Abstract

Approaches are described which enable a computing device (e.g., mobile phone, tablet computer) to utilize one or more facial recognition techniques to control access to the device and to detect when artificial representations of a user, such as a picture or photograph, are being used in an attempt to gain access to the device. Evidence indicative of artificial representations may include lack of changes in facial skin color between multiple images captured by a camera, ability to track one or more features of the human face while the camera is rotated or moved, presence of secular reflections caused by an illumination device, absence of shadows in the image, and others.

Claims

1. A computing device, comprising:

a camera;

an illumination device;

at least one processor; and

memory including instructions that, when executed by the at least one processor, cause the computing device to;

activate a facial recognition mode to control access to the computing device;

activate the illumination device to emit light to an area within a field of view of the camera;

capture an image of the area using the camera, the image containing first image data corresponding to at least a portion of a human face;

analyze the first image data to match the first image data corresponding to the at least the portion of the human face to a stored model corresponding to a face of an authorized user of the computing device;

detect evidence in the first image data indicative of an artificial two dimensional representation of the human face presented to the camera, wherein the evidence is based at least in part on the light generated by the illumination device, the evidence including at least one or more specular reflections caused by the light emitted by the illumination device, a pattern of the specular reflections having a characteristic of a flat surface; and

deny the access to the computing device in response to detecting that the evidence in the image indicates the artificial two dimensional representation presented to the camera.

2. The computing device of claim 1, wherein the memory further includes instructions that, when executed by the at least one processor, cause the computing device to:

generate, using the camera, second image data corresponding to at least a portion of a human face;

analyze the second image data to match the second image data corresponding to the at least the portion of the human face to a stored model corresponding to a face of an authorized user of the computing device; and

determine that an actual human face is presented to the camera based at least in part on evidence of one or more shadows in the second image data caused by the light emitted by the illumination device.

3. The computing device of claim 2, wherein the memory further includes instructions that, when executed by the at least one processor, cause the computing device to:

allow access to at least one function of a computing device based at least on the evidence of one or more shadows in the second image data.

4. The computing device of claim 1, wherein the memory further includes instructions that, when executed by the at least one processor, cause the computing device to:

generate second image data representing the human face using the camera;

generate third image data representing the human face using the camera; and

detect one or more changes in facial skin color in the human face based at least in part on comparing the second image data and the third image data.

5. The computing device of claim 4, wherein the memory further includes instructions that, when executed by the at least one processor, cause the computing device to:

determine that an actual human face is presented to the camera based at least in part on the one or more changes in facial skin color; and

allow access to at least one function of the computing device.

6. A computer implemented method, comprising:

under the control of one or more computer systems configured with executable instructions,

causing light to be emitted by an illumination component;

generating first image data, using a camera, the first image data representing at least a portion of a human face;

analyzing the first image data to identify the portion of the human face;

detecting one or more specular reflections in the first image data associated with the light; and

preventing access to at least one function of a computing device based at least in part on the detecting one or more specular reflections, a pattern of the specular reflections having a characteristic of a flat surface.

7. The computer implemented method of claim 6, further comprising:

generating, using the camera, second image data representing the human face;

generating, using the camera, third image data representing the human face; and

detecting one or more changes in facial skin color in the human face based at least in part on comparing the second image data and the third image data.

8. The computer implemented method of claim 7, further comprising:

allowing access to at least one function of a computing device based at least in part on the detecting one or more changes in facial skin color.

9. The computer implemented method of claim 6, further comprising:

causing the light to be emitted in an area within a field of view of the camera;

generating, using the camera, second image data of the human face; and

detecting a presence of one or more shadows in the second image data.

10. The computer implemented method of claim 9, further comprising:

determining that an actual human face is presented to the camera based at least in part on the one or more shadows in the second image data; and

allowing access to at least one function of a computing device based at least in part on the one or more shadows in the second image data.

11. The computer implemented method of claim 6, further comprising:

generating second image data using the camera;

generating third image data using the camera, the third image data generated with the camera having been at least one of;

rotated or moved after generating the second image data; and

tracking a threshold number of features between the second image data and the third image data using a homography.

12. The computer implemented method of claim 6, further comprising:

causing the light to be emitted in an area within a field of view of the camera; and

detecting, in the first image data, the one or more specular reflections, wherein the one or more specular reflections indicate a flat surface on which a two dimensional representation of the human face is rendered.

13. The computer implemented method of claim 6, further comprising:

matching the first image data representing the human face to a stored model of a face of an authorized user of the computing device.

14. A computing device, comprising:

a camera;

an illumination component;

at least one processor; and

memory including instructions that, when executed by the at least one processor, cause the computing device to;

cause light to be emitted by the illumination component;

generate first image data, using the camera, the first image data representing at least a portion of a human face;

analyze the first image data to identify the portion of the human face;

detect one or more specular reflections in the first image data associated with the light; and

prevent access to at least one function of the computing device based at least in part on the one or more specular reflections, a pattern of the specular reflections having a characteristic of a flat surface.

15. The computing device of claim 14, wherein the memory further includes instructions that, when executed by the at least one processor, cause the computing device to:

generate second image data representing the human face;

generate third image data representing the human face; and

detect one or more changes in facial skin color in the human face based at least in part on comparing the second image data and the third image data.

16. The computing device of claim 15, wherein the memory further includes instructions that, when executed by the at least one processor, cause the computing device to:

allow access to at least one function of the computing device based at least in part on the detecting one or more changes in facial skin color.

17. The computing device of claim 14, wherein the memory further includes instructions that, when executed by the at least one processor, cause the computing device to:

cause the light to be emitted in an area within a field of view of the camera;

generating second image data of the human face; and

detect a presence of one or more shadows in the second image data.

18. The computing device of claim 17, wherein the memory further includes instructions that, when executed by the at least one processor, cause the computing device to:

determine that an actual human face is presented to the camera based at least in part on the one or more shadows in the second image data; and

allow access to at least one function of a computing device based at least in part on the one or more shadows in the second image data.

19. The computing device of claim 14, wherein the memory further includes instructions that, when executed by the at least one processor, cause the computing device to:

generate second image data and third image data using the camera, the third image data captured with the camera having been at least one of;

rotated or moved after generating the second image data; and

track a threshold number of features between the second image data and the third image data using a homography.

20. The computing device of claim 14, wherein the memory further includes instructions that, when executed by the at least one processor, cause the computing device to:

cause the light to be emitted in an area within a field of view of the camera; and

detect, in the first image data, the one or more specular reflections, wherein the one or more specular reflections indicate a flat surface on which a two dimensional representation of the human face is rendered.