Detection of spoofing attacks for video-based authentication

• US 9,594,958 B2
• Filed: 11/24/2014
• Issued: 03/14/2017
• Est. Priority Date: 11/24/2014
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Ben-Shalom, Omer, Duenias, Oren, Levi, Shachaf, Naor, Ayelet
• Primary Examiner(s): Kassa, Yosef
• Application Number:
  US14/551,839
• Publication Number:
  US 20160148066A1
• Time in Patent Office: 841 Days
• Field of Search: 382/115, 382/170, 382/171, 382/194, 382/224, 382/229, 382/118, 382/128, 382/154, 382/209, 382/218, 382/278, 340/5.8, 340/5.86, 340/5.81, 340/5.82
• US Class Current: 1/1
• CPC Class Codes: G06V 20/40 in video content extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44
• CPC Class Codes: G06V 40/16 Human faces, e.g. facial parts, sketches or expressions
• CPC Class Codes: G06V 40/166 using acquisition arrangements
• CPC Class Codes: G06V 40/40 Spoof detection, e.g. liveness detection
• CPC Class Codes: H04N 2005/91392 using means for preventing making copies of projected video images
• CPC Class Codes: H04N 23/745 Detection of flicker frequency or suppression of flicker wherein the flicker is caused by illumination, e.g. due to fluorescent tube illumination or pulsed LED illumination

Abstract

Methods, apparatus, systems and articles of manufacture detect spoofing attacks for video-based authentication are disclosed. Disclosed example method to perform video-based authentication include determining whether a sequence of input images provided to perform video-based authentication of a subject exhibits a first region having fluctuating pixel values. Such example methods also include determining that the sequence of input images is associated with a spoofing attack in response to determining that the sequence of input images exhibits the first region having fluctuating pixel values.

Claims

1. A method to perform video-based authentication, the method comprising:

determining, by executing an instruction with a processor, whether a sequence of input images provided to perform video-based authentication of a subject exhibits a first region having pixel values that fluctuate across successive ones of at least some of the images in the sequence; and

determining, by executing an instruction with the processor, that the sequence of input images is associated with a spoofing attack in response to determining that the sequence of input images exhibits the first region having pixel values that fluctuate.

2. The method as defined in claim 1, wherein the sequence of input images is a second sequence of input images, and further including randomly sampling a first sequence of input images to obtain the second sequence of input images.

3. The method as defined in claim 2, wherein randomly sampling the first sequence of input images includes:

capturing the first sequence of input images at a first rate higher than a second rate;

grouping the first sequence of input images into successive groups of input images containing respective numbers of images based on a relationship between the first rate and the second rate; and

randomly selecting one input image from respective ones of the successive groups of input images to form the second sequence of input images.

4. The method as defined in claim 1, wherein the determining of whether the sequence of input images exhibits the first region having pixels values that fluctuate includes:

determining differences between successive pairs of images in the sequence of input images to determine a sequence of difference images; and

processing the sequence of difference images to determine whether the sequence of input images exhibits the first region having pixel values that fluctuate.

5. The method as defined in claim 4, wherein the processing of the sequence of difference images includes:

processing successive pairs of difference images in the sequence of difference images to identify fluctuating pixels, the fluctuating pixels being pixels that fluctuate between at least two values across three successive images in the sequence of input images;

determining a number of pixels included in a first group of neighboring fluctuating pixels; and

if the number of pixels satisfies a threshold, determining that the sequence of input images exhibits the first region having pixel values that fluctuate, the first region corresponding to the first group of neighboring fluctuating pixels.

6. The method as defined in claim 1, further including triggering operation of a further access control procedure to authenticate the subject based on the sequence of input images in response to determining that the sequence of input images does not exhibit any region having fluctuating pixel values.

7. The method as defined in claim 1, further including determining that the sequence of input images depicts a scene including content generated by a video display in response to determining that the sequence of input images exhibits the first region having pixel values that fluctuate.

8. A tangible machine readable storage medium comprising machine readable instructions which, when executed, cause a machine to at least:

determine whether a sequence of input images provided to perform video-based authentication of a subject exhibits a first region having pixel values that fluctuate across successive ones of at least some of the images in the sequence; and

determine that the sequence of input images is associated with a spoofing attack in response to determining that the sequence of input images exhibits the first region having pixel values that fluctuate.

9. The storage medium as defined in claim 8, wherein the sequence of input images is a second sequence of input images, and the machine readable instructions, when executed, further cause the machine to randomly sample a first sequence of input images to obtain the second sequence of input images.

10. The storage medium as defined in claim 9, wherein to randomly sample the first sequence of input images, the machine readable instructions, when executed, further cause the machine to:

capture the first sequence of input images at a first rate higher than a second rate;

group the first sequence of input images into successive groups of input images containing respective numbers of images based on a relationship between the first rate and the second rate; and

randomly select one input image from respective ones of the successive groups of input images to form the second sequence of input images.

11. The storage medium as defined in claim 8, wherein to determine whether the sequence of input images exhibits the first region having pixels values that fluctuate, the machine readable instructions, when executed, further cause the machine to:

determine differences between successive pairs of images in the sequence of input images to determine a sequence of difference images; and

process the sequence of difference images to determine whether the sequence of input images exhibits the first region having pixel values that fluctuate.

12. The storage medium as defined in claim 11, wherein to process the sequence of difference images, the machine readable instructions, when executed, further cause the machine to:

process successive pairs of difference images in the sequence of difference images to identify fluctuating pixels, the fluctuating pixels being pixels that fluctuate between at least two values across three successive images in the sequence of input images;

determine a number of pixels included in a first group of neighboring fluctuating pixels; and

if the number of pixels satisfies a threshold, determine that the sequence of input images exhibits the first region having pixel values that fluctuate, the first region corresponding to the first group of neighboring fluctuating pixels.

13. The storage medium as defined in claim 8, wherein the machine readable instructions, when executed, further cause the machine to trigger operation of a further access control procedure to authenticate the subject based on the sequence of input images in response to determining that the sequence of input images does not exhibit any region having fluctuating pixel values.

14. The storage medium as defined in claim 8, wherein the machine readable instructions, when executed, further cause the machine to determine that the sequence of input images depicts a scene including content generated by a video display in response to determining that the sequence of input images exhibits the first region having fluctuating pixel values.

15. An apparatus to perform video-based authentication, the apparatus comprising:

a fluctuating pixel detector to determine whether a sequence of input images provided to perform video-based authentication of a subject exhibits a first region having pixel values that fluctuate across successive ones of at least some of the images in the sequence; and

a video sequence validator to determine that the sequence of input images is associated with a spoofing attack in response to determining that the sequence of input images exhibits the first region having pixel values that fluctuate.

16. The apparatus as defined in claim 15, wherein the sequence of input images is a second sequence of input images, and further including an image capturer to randomly sample a first sequence of input images to obtain the second sequence of input images.

17. The apparatus as defined in claim 16, wherein the image capturer is to randomly sample the first sequence of input images by:

capturing the first sequence of input images at a first rate higher than a second rate;

grouping the first sequence of input images into successive groups of input images containing respective numbers of images based on a relationship between the first rate and the second rate; and

randomly selecting one input image from respective ones of the successive groups of input images to form the second sequence of input images.

18. The apparatus as defined in claim 15, wherein the fluctuating pixel detector is further to:

determine differences between successive pairs of images in the sequence of input images to determine a sequence of difference images; and

process the sequence of difference images to determine whether the sequence of input images exhibits the first region having pixel values that fluctuate.

19. The apparatus as defined in claim 18, wherein the fluctuating pixel detector is to process the sequence of difference images by:

processing successive pairs of difference images in the sequence of difference images to identify fluctuating pixels, the fluctuating pixels being pixels that fluctuate between at least two values across three successive images in the sequence of input images;

determining a number of pixels included in a first group of neighboring fluctuating pixels; and

if the number of pixels satisfies a threshold, determining that the sequence of input images exhibits the first region having pixel values that fluctuate, the first region corresponding to the first group of neighboring fluctuating pixels.

20. The apparatus as defined in claim 15, wherein the video sequence validator is further to trigger operation of a further access control procedure to authenticate the subject based on the sequence of input images in response to determining that the sequence of input images does not exhibit any region having fluctuating pixel values.

21. The apparatus as defined in claim 15, wherein the video sequence validator is further to determine that the sequence of input images depicts a scene including content generated by a video display in response to determining that the sequence of input images exhibits the first region having pixel values that fluctuate.

22. A system to perform video-based authentication, the system comprising:

means for determining whether a sequence of input images provided to perform video-based authentication of a subject exhibits a first region having pixel values that fluctuate across successive ones of at least some of the images in the sequence; and

means for determining that the sequence of input images is associated with a spoofing attack in response to determining that the sequence of input images exhibits the first region having pixel values that fluctuate.

23. The system as defined in claim 22, wherein the sequence of input images is a second sequence of input images, and further including means for randomly sampling a first sequence of input images to obtain the second sequence of input images.

24. The system as defined in claim 22, wherein the means for determining whether the sequence of input images exhibits the first region having pixels values that fluctuate includes:

means for determining differences between successive pairs of images in the sequence of input images to determine a sequence of difference images; and

means for processing the sequence of difference images to determine whether the sequence of input images exhibits the first region having pixel values that fluctuate.

25. The system as defined in claim 22, further including means for triggering operation of a further access control procedure to authenticate the subject based on the sequence of input images in response to determining that the sequence of input images does not exhibit any region having fluctuating pixel values.