Sign-in system, method, apparatus and server based on facial recognition

Patent metadata

Field	Value
Publication number	US-10664580-B2
Application number	US-201816100630-A
Country	US
Kind code	B2
Filing date	Aug 10, 2018
Priority date	Mar 4, 2016
Publication date	May 26, 2020
Grant date	May 26, 2020

Abstract

A sign-in method and server based on facial recognition are provided. The method includes: receiving a face image of a sign-in user from a sign-in terminal. According to the face image of the sign-in user, whether a target registration user matching the sign-in user exists in a pre-stored registration set is detected. The registration set includes a face image of at least one registration user. Further, the target registration user is confirmed as signed in successfully if the target registration user exists in the registration set.

First claim

What is claimed is:

1. A sign-in method based on facial recognition, comprising:
creating a current conference event carrying conference information based on input of an administrative user;
creating a registration set corresponding to the current conference event, the registration set comprising a face image of at least one registration user, each registration user having a corresponding identification, comprising:
entering, based on the input of the administrative user, a face image of a first participant of the current conference event to the registration set as the face image of a first registration user;
adding a face image of a second registration user to the registration set by searching previous conference registration data based on an identification of a second participant of the current conference event input by the administrator user; and
receiving a face image of a third registration user from a user terminal and storing the face image of the third registration user to the registration set;

receiving a face image of a sign-in user from a sign-in terminal at a conference corresponding to the current conference event;
detecting, according to the face image of the sign-in user, whether a target registration user matching the sign-in user exists in the registration set; and
confirming that the target registration user signs in successfully if the target registration user exists in the registration set.

2. The method according to claim 1 , wherein the detecting, according to the face image of the sign-in user, whether a target registration user matching the sign-in user exists in a pre-stored registration set comprises:
extracting an image feature from the face image of the sign-in user by using a facial recognition algorithm having a recognition accuracy degree greater than a first threshold;
separately calculating a similarity degree between the face image of the sign-in user and the face image of each registration user according to the image feature of the face image of the sign-in user and an image feature of the face image of each registration user;
detecting whether a maximum similarity degree is greater than a second threshold; and
confirming that a registration user corresponding to the maximum similarity degree is the target registration user matching the sign-in user, if the maximum similarity degree is greater than the second threshold.

3. The method according to claim 2 , wherein the extracting an image feature from the face image of the sign-in user by using a facial recognition algorithm having a recognition accuracy degree greater than a first threshold comprises:
performing feature point positioning on the face image of the sign-in user, to obtain a feature point positioning result;
dividing the face image of the sign-in user into multiple face regions according to the feature point positioning result;
extracting a feature from each face region by using a deep network corresponding to the face region to obtain multiple features corresponding to the multiple face regions; and
recombining the multiple features extracted from the face regions, to obtain the image feature of the face image of the sign-in user.

4. The method according to claim 1 , wherein the face image of the third registration user is submitted by the third registration user to a registration page displayed on the user terminal.

5. The method according to claim 4 , before the storing the face image of the third registration user into the registration set, further comprising:
determining whether the face image of the third registration user meets a registration condition, wherein the registration condition comprises at least one of: a light condition of the face image of the third registration user meeting a first predetermined condition or a face pose of the face image of the third registration user meeting a second predetermined condition;
storing the face image of the third registration user into the registration set if the face image of the third registration user meets the registration condition; and
sending a registration failure response to the user terminal if the face image of the third registration user does not meet the registration condition, the registration failure response being used for triggering the user terminal to present prompt information indicating that the face image of the third registration user is invalid.

6. The method according to claim 4 , further comprising:
sending registration prompt information carrying a registration link to the user terminal corresponding to the third registration user, the registration prompt information guiding a user to log on to the registration page corresponding to the registration link.

7. The method according to claim 1 , further comprising:
providing a data import option to the administrative user for creating the current conference event; and
upon detecting that the data import option is selected, editing the previous conference registration data to generate the current conference even and to establish at least part of the registration set corresponding to the current conference event.

8. The method according to claim 1 , wherein creating the registration set comprises:
providing a batch data processing option for creating the registration set; and
upon detecting that the batch data processing option, automatically adding a batch of multiple face images entered by the administrative user to the registration set, each image corresponding to a participant and has a file name indicating an identification of the participant.

9. A server, comprising:
one or more processors; and
a memory, wherein
the one or more processors are configured to perform:
creating a current conference event carrying conference information based on input of an administrative user;
creating a registration set corresponding to the current conference event, the registration set comprising a face image of at least one registration user, each registration user having a corresponding identification, comprising:
entering, based on the input of the administrative user, a face image of a first participant of the current conference event to the registration set as the face image of a first registration user;
adding a face image of a second registration user to the registration set by searching previous conference registration data based on an identification of a second participant of the current conference event input by the administrator user; and
receiving a face image of a third registration user from a user terminal and storing the face image of the third registration user to the registration set;

receiving a face image of a sign-in user from a sign-in terminal at a conference corresponding to the current conference event;
detecting, according to the face image of the sign-in user, whether a target registration user matching the sign-in user exists in the registration set; and
confirming that the target registration user signs in successfully if the target registration user exists in the registration set.

10. The server according to claim 9 , wherein the one or more processors are further configured to perform:
extracting an image feature from the face image of the sign-in user by using a facial recognition algorithm having a recognition accuracy degree greater than a first threshold;
separately calculating a similarity degree between the face image of the sign-in user and the face image of each registration user according to the image feature of the face image of the sign-in user and an image feature of the face image of each registration user;
detecting whether a maximum similarity degree is greater than a second threshold; and
confirming that a regis

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Huang Feiyue
• Wu Yongjian
• Tan Guofu
• Li Jilin
• Chen Zhibo
• Liang Xiaoqing
• Tao Zhiwei
• Zhou Kejing
• Mei Ke

Classifications

• G06Q10/10 Primary

  Office automation; Time management · CPC title

• G06F21/32 Primary

  using biometric data, e.g. fingerprints, iris scans or voiceprints · CPC title

• G06K9/6274

  Physics · mapped topic

• G06K9/00288

  Physics · mapped topic

• G06K9/00228

  Physics · mapped topic