Liveness test method and apparatus

Patent metadata

Field	Value
Publication number	US-10121059-B2
Application number	US-201715656350-A
Country	US
Kind code	B2
Filing date	Jul 21, 2017
Priority date	Aug 23, 2016
Publication date	Nov 6, 2018
Grant date	Nov 6, 2018

Abstract

A liveness test method and apparatus is disclosed. A processor implemented liveness test method includes extracting an interest region of an object from a portion of the object in an input image, performing a liveness test on the object using a neural network model-based liveness test model, the liveness test model using image information of the interest region as provided first input image information to the liveness test model and determining liveness based at least on extracted texture information from the information of the interest region by the liveness test model, and indicating a result of the liveness test.

First claim

What is claimed is:

1. A processor implemented liveness test method, the liveness test method comprising:
extracting an interest region of an object from a portion of the object in an input image;
performing a liveness test on the object using a neural network model-based liveness test model for each of one or more images of the object respectively input to the liveness test model,
wherein image information of the interest region is provided, as first input image information, to a first input layer of the liveness test model,
wherein image information of the input image or image information of the portion of the object is provided, as second input image information, to a second input layer of the liveness test model, and
wherein the performing of the liveness test further includes determining, by the liveness test model, liveness of the object based at least on extracted texture information from the first input image information by the liveness test model and information extracted from the second input image information by the liveness test model; and
indicating a result of the liveness test.

2. The method of claim 1 , wherein the extracting of the interest region comprises:
extracting a select portion including a pupil region as the interest region from the portion of the object in the input image.

3. The method of claim 2 ,
wherein the liveness test model uses the image information of the portion of the object in the input image, within a facial region of the object and including the select portion, as the second input image information provided to the liveness test model that is also provided the first input image information, and
wherein the determining of the liveness is based at least on the extracted texture information from the select portion and spatial information extracted based on the second input image information by the liveness test model.

4. The method of claim 1 , wherein the extracting of the interest region comprises:
extracting respective select portions, from the portion of the object in the input image, including at least one of an eye, a nose, and lips as the interest region.

5. The method of claim 4 ,
wherein the liveness test model uses the image information of the portion of the object in the input image, within a facial region of the object and including one or more of the respective select portions, as the second input image information provided to the liveness test model that is also provided the first input image information, and
wherein the determining of the liveness is based at least on extracted texture information from the one or more of the respective select portions by the liveness test model and spatial information extracted based on the second input image information by the liveness test model.

6. The method of claim 1 , wherein the extracting of the interest region comprises:
extracting respective one or more select portions, from a fingerprint region as the portion of the object in the input image, as respective interest regions.

7. The method of claim 6 ,
wherein the liveness test model uses the image information of the portion of the object in the input image, as the second input image information provided to the liveness test model that is also provided the first input image information, and
wherein the determining of the liveness is based at least on extracted texture information from the respective one or more select portions by the liveness test model and spatial information extracted based on the second input image information by the liveness test model.

8. The method of claim 1 , wherein the extracting of the interest region comprises:
extracting a select portion, from a vein region as the portion of the object in the input image, as the interest region.

9. The method of claim 8 ,
wherein the liveness test model uses the image information of the object in the portion of the input image, as the second input image information provided to the liveness test model that is also provided the first input image information, and
wherein the determining of the liveness is based at least on the extracted texture information from the select portion and spatial information extracted based on the second input image information by the liveness test model.

10. The method of claim 1 , wherein the extracting of the interest region comprises:
detecting feature points in the input image to determine the portion of the object in the input image; and
determining the interest region based on the detected feature points.

11. The method of claim 1 , wherein a location of the interest region is determined based on a determined type of biometric information for which the liveness test is determined to be performed.

12. The method of claim 1 , further comprising selectively controlling access to functions of a computing apparatus based on results of the performed liveness test.

13. The method of claim 1 ,
wherein the liveness test model uses image information of the input image or image information of the portion of the object in the input image as provided second input image information to the liveness test model that is also provided the first input image information, and
wherein the determining of the liveness is based at least on the extracted texture information from the first input image information and spatial information extracted based on the second input image information by the liveness test model.

14. The method of claim 13 , further comprising:
adjusting a pixel size of at least the image information of the portion of the object in the input image, after the extracting of the interest region, to generate the second input image information having a pixel size same as a pixel size of the image information of the interest region.

15. A non-transitory computer-readable storage medium storing instructions, that when executed by a processor, cause the processor to perform the method of claim 1 .

16. A processor implemented liveness test method, the liveness test method comprising:
extracting an interest region of an object from a portion of the object in an input image;
performing a liveness test on the object using a neural network model-based liveness test model, the liveness test model using image information of the interest region as provided first input image information to the liveness test model and determining liveness based at least on extracted texture information from the image information of the interest region by the liveness test model; and
indicating a result of the liveness test,
wherein the liveness test model uses image information of the input image or image information of the portion of the object in the input image as provided second input image information to the liveness test model that is also provided the first input image information,
wherein the determining of the liveness is based at least on the extracted texture information from the first input image information and spatial information extracted based on the second input image information by the liveness test model, and
wherein an effective resolution of the first input image information is higher than an effective resolution of the second input image information.

17. The method of claim 16 , further comprising:
adjusting a pixel size of at least the image information of the portion of the object in the input image, after the extracting of the interest region, to generate the second input image information.

18. The method of claim 17 , wherein a pixel size of the first input image information is equal to a

Assignees

• Samsung Electronics Co Ltd

Inventors

• Yoo Byungin
• Kwak Youngjun
• Kim Jungbae
• Son Jinwoo
• Lee Changkyo
• Choi Chang Kyu
• Han Jaejoon

Classifications

• G06V40/40 Primary

  Spoof detection, e.g. liveness detection · CPC title

• G06V40/172

  Classification, e.g. identification · CPC title

• G06K9/00906

  Physics · mapped topic

• G06K9/00268

  Physics · mapped topic

• G06K9/00114

  Physics · mapped topic