Face anti-counterfeiting detection methods and systems, electronic devices, programs and media

Patent metadata

Field	Value
Publication number	US-11482040-B2
Application number	US-202117203435-A
Country	US
Kind code	B2
Filing date	Mar 16, 2021
Priority date	Mar 16, 2017
Publication date	Oct 25, 2022
Grant date	Oct 25, 2022

Abstract

A face anti-counterfeiting detection method includes: obtaining an image or video to be detected containing a face; extracting a feature of the image or video to be detected, and detecting whether the extracted feature contains counterfeited face clue information; and determining whether the face passes the face anti-counterfeiting detection according to a detection result.

First claim

What is claimed is:

1. A face anti-counterfeiting detection method, comprising:
obtaining an image or video to be detected containing a face;
extracting a feature of the image or video to be detected, and detecting whether the extracted feature contains counterfeited face clue information, wherein the counterfeited face clue information comprises counterfeited clue information of an imaging medium, and the counterfeited clue information of the imaging medium comprises edge information, reflection information, and material information of the imaging medium; and
determining whether the face passes the face anti-counterfeiting detection according to a detection result.

2. The method according to claim 1 , wherein the extracted feature comprises one or more of the following: a local binary pattern feature, a histogram of sparse coding feature, a panorama feature, a face map feature, and a face detail map feature; or
wherein the counterfeited face clue information has human eye observability under a visible light condition.

3. The method according to claim 1 , wherein the extracting a feature of the image or video to be detected, and detecting whether the extracted feature contains counterfeited face clue information comprise:
inputting the image or video to be detected to a neural network, and outputting, by the neural network, a detection result for indicating whether the image or video to be detected contains at least one piece of counterfeited face clue information, wherein the neural network is pre-trained based on a training image set containing the counterfeited face clue information.

4. The method according to claim 3 , wherein the training image set comprises a plurality of facial images serving as positive samples for training and a plurality of images serving as negative samples for training;
the training image set containing the counterfeited face clue information is obtained by the following operations:
obtaining the plurality of facial images serving as positive samples for training; and
performing image processing for simulating the counterfeited face clue information on at least a part of the obtained at least one facial image to generate at least one image serving as a negative sample for training.

5. The method according to claim 1 , wherein the obtaining an image or video to be detected containing a face comprises:
obtaining, by a visible light camera of a terminal device, the image or video to be detected containing a face.

6. The method according to claim 3 , wherein the neural network comprises a first neural network located in the terminal device;
the determining whether the face passes the face anti-counterfeiting detection according to a detection result comprises: determining, by the terminal device, whether the face passes the face anti-counterfeiting detection according to a detection result output by the first neural network.

7. The method according to claim 1 , wherein the obtaining an image or video to be detected containing a face comprises:
receiving, by a server, the image or video to be detected containing a face sent by the terminal device.

8. The method according to claim 3 , wherein the neural network comprises a second neural network located in the server, wherein the determining whether the image or video to be detected passes the face anti-counterfeiting detection according to a detection result comprises: determining, by the server, whether the face passes the face anti-counterfeiting detection according to a detection result output by the second neural network, and returning to the terminal device a determination result about whether the face passes the face anti-counterfeiting detection.

9. The method according to claim 8 , wherein the neural network further comprises a first neural network located in the terminal device, and a size of the first neural network is less than a size of the second neural network;
the method further comprises:
inputting a video containing a face obtained by the terminal device to the first neural network, and outputting, by the first neural network, a detection result for indicating whether the video containing a face contains at least one piece of counterfeited face clue information; and
in response to the detection result indicating that the video containing a face does not contain the counterfeited face clue information, selecting a partial video or image from the video containing a face as the image or video to be detected to be sent to the server.

10. The method according to claim 9 , wherein the selecting a partial video or image from the video containing a face as the image or video to be detected to be sent to the server comprises:
obtaining a status of a network currently used by the terminal device; and at least one of the following operations:
if the status of the network currently used by the terminal device satisfies a first preset condition, selecting a partial video from the video obtained by the terminal device as the video to be detected to be sent to the server; or
if the status of the network currently used by the terminal device does not satisfy the first preset condition, but the status of the network currently used by the terminal device satisfies a second preset condition, selecting at least one image that satisfies a preset standard from the video obtained by the terminal device as the image to be detected to be sent to the server.

11. The method according to claim 10 , wherein when selecting a partial video from the video obtained by the terminal device as the video to be detected to be sent to the server, inputting the video to be detected to the second neural network, and outputting, by the second neural network, a detection result for indicating whether the video to be detected contains at least one piece of counterfeited face clue information, comprising:
selecting, by the server, at least one image from the video to be detected as the image to be detected to be input to the second neural network, and outputting, by the second neural network, a detection result for indicating whether the image to be detected contains at least one piece of counterfeited face clue information.

12. The method according to claim 9 , wherein in response to the detection result indicates that the video containing a face contains at least one piece of counterfeited face clue information, the determining whether the face passes the face anti-counterfeiting detection according to a detection result comprises: determining, by the terminal device, that the face fails to pass the face anti-counterfeiting detection according to the detection result output by the first neural network.

13. The method according to claim 9 , further comprising: returning, by the server, the detection result output by the second neural network to the terminal device;
the determining whether the face passes the face anti-counterfeiting detection according to a detection result comprises: determining, by the terminal device, whether the face passes the face anti-counterfeiting detection according to the detection result output by the second neural network.

14. The method according to claim 9 , wherein the determining whether the face passes the face anti-counterfeiting detection according to a detection result comprises: determining, by the server, whether the face passes the face anti-counterfeiting detection according to the detection result output by the second neural network, and sending to the terminal device a determination result about whether the face passes the face anti-counterfeiting detection.

15. The method according to claim 3 , further compr

Assignees

• Beijing Sensetime Tech Development Co Ltd

Inventors

• Wu Liwei
• Bao Tianpeng
• Yu Meng
• Che Yinghui
• Zhao Chenxu

Classifications

• G06V40/45

  Detection of the body part being alive · CPC title

• G06V30/19173

  Classification techniques · CPC title

• G06V10/82 Primary

  using neural networks · CPC title

• G06V40/168 Primary

  Feature extraction; Face representation · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title