Granted Patent B2

US 12,536,261 · App. 17/985,322 · Granted Jan 27, 2026

Inventors: Tingqiu Yuan (Shenzhen, CN); Libin Zhang (Beijing, CN); Chang Liu (Shenzhen, CN); Huimin Zhang (Shenzhen, CN)

Assignee: HUAWEI TECHNOLOGIES CO., LTD.

G06F21/32

Abstract

Claims (79)

1 . A device enabling method comprising:

collecting at least two types of biometric feature information of a user;

performing an identity authentication on the user based on the at least two types of biometric feature information; and

when the identity authentication succeeds, enabling at least one of a wearable device or a smart device, wherein the wearable device is communicatively connected to the smart device, wherein:

the performance of the identity authentication on the user based on the at least two types of biometric feature information comprises:

performing biometric authentication based on the at least two types of biometric feature information and a biometric authentication model: and

determining an identity authentication result of the user based on a biometric authentication result of the biometric authentication;

each type of the at least two types of biometric feature information corresponds to one of biometric authentication models, respectively, and each of the corresponding biometric authentication models is used to perform biometric authentication based on the corresponding biometric feature information:

the performance of the biometric authentication based on the at least two types of biometric feature information and the biometric authentication model comprises:

performing biometric authentication based on each type of biometric feature information and the corresponding biometric authentication models, to respectively obtain biometric authentication results corresponding to the biometric feature information; and

the determining of the identity authentication result of the user based on the biometric authentication result comprises:

determining the identity authentication result of the user based on the biometric authentication results respectively corresponding to the at least two types of biometric feature information; and

the determining of the identity authentication result of the user based on the biometric authentication results corresponding to the at least two types of biometric feature information comprises:

performing weighted calculation on the biometric authentication results corresponding to the at least two types of biometric feature information, and determining the identity authentication result of the user based on a result obtained through the weighted calculation wherein a sum of weights corresponding to the at least two types of biometric feature information is 1, and a weight corresponding to each type of biometric feature information is adjustable.

2 . The method according to claim 1 , wherein

before the performing of biometric authentication based on the at least two types of biometric feature information and the biometric authentication model, the method further comprises:

performing training based on sample data to generate the biometric authentication model.

3 . The method according to claim 2 , wherein

each type of the at least two types of biometric feature information corresponds to one of biometric authentication models, respectively;

the sample data comprises at least two types of sample feature information and sample authentication results corresponding to the at least two types of sample feature information, and each type of sample feature information is one respective type of biometric feature information; and

the performing of training based on the sample data to generate the biometric authentication model comprises:

inputting each type of sample feature information into an initial authentication model to perform feature authentication, to obtain a feature authentication result corresponding to the sample feature information;

adjusting a model parameter of the initial authentication model based on the feature authentication result corresponding to the sample feature information and the sample authentication result corresponding to the sample feature information until a difference between the feature authentication result obtained through feature authentication based on the sample feature information and the sample authentication result corresponding to the sample feature information meets a preset condition; and

determining an authentication model that meets the preset condition as the biometric authentication model corresponding to the biometric feature information.

4 . The method according to claim 2 , wherein

the at least two types of biometric feature information correspond to one comprehensive biometric authentication model;

the sample data comprises at least two types of sample feature information and a sample authentication result corresponding to the at least two types of sample feature information, and each type of sample feature information is one type of biometric feature information; and

the performing of training based on the sample data to generate the biometric authentication model comprises:

inputting the at least two types of sample feature information into an initial authentication model to perform feature authentication, to obtain a feature authentication result corresponding to the at least two types of sample feature information;

adjusting a model parameter of the initial authentication model based on the feature authentication result corresponding to the at least two types of sample feature information and the sample authentication result corresponding to the at least two types of sample feature information until a difference between the feature authentication result obtained through feature authentication based on the at least two types of sample feature information and the sample authentication result corresponding to the at least two types of sample feature information meets a preset condition; and

determining the authentication model that meets the preset condition as the biometric authentication model corresponding to the at least two types of biometric feature information.

5 . The method according to claim 1 , wherein

the at least two types of biometric feature information comprise at least two types of ear canal feature information, auricle feature information, vibration feature information, voiceprint feature information, facial feature information, fingerprint feature information, and iris feature information; and the vibration feature information is vibration information of a part on which the wearable device is worn.

6 . A device enabling apparatus comprising a processor and a memory; the memory stores a program; and the processor is configured to execute the program stored in the memory, so that the device enabling apparatus is at least configured to

collect at least two types of biometric feature information of a user;

perform identity authentication on the user based on the at least two types of biometric feature information; and

when the identity authentication succeeds, enable at least one of a wearable device or a smart device, wherein the wearable device is communicatively connected to the smart device,

the performance of the identity authentication on the user based on the at least two types of biometric feature information comprises:

performance of biometric authentication based on the at least two types of biometric feature information and a biometric authentication model: and

determination of an identity authentication result of the user based on a biometric authentication result of the biometric authentication each type of the at least two types of biometric feature information corresponds to one of biometric authentication models, respectively, and each of the corresponding biometric authentication models is used to perform the biometric authentication based on the corresponding biometric feature information:

the performance of the biometric authentication based on the at least two types of biometric feature information and the biometric authentication model comprises:

performance of biometric authentication based on each type of biometric feature information and the corresponding biometric authentication models, to respectively obtain biometric authentication results corresponding to the biometric feature information; and

the determination of the identity authentication result of the user based on the biometric authentication result comprises:

performance of a weighted calculation on the biometric authentication results corresponding to the at least two types of biometric feature information, and determining the identity authentication result of the user based on a result obtained through the weighted calculation, wherein a sum of weights corresponding to the at least two types of biometric feature information is 1, and a weight corresponding to each type of biometric feature information is adjustable.

7 . The apparatus according to claim 6 , wherein the device enabling apparatus is further configured to:

before the performance of the biometric authentication based on the at least two types of biometric feature information and the biometric authentication model,

perform training based on sample data to generate the biometric authentication model.

8 . The apparatus according to claim 7 , wherein

each type of the at least two types of biometric feature information corresponds to one of biometric authentication model, respectively;

the sample data comprises at least two types of sample feature information and sample authentication results corresponding to the at least two types of sample feature information, and each type of sample feature information is one respective type of biometric feature information; and

the performance of training based on the sample data to generate the biometric authentication model comprises:

input of each type of sample feature information into an initial authentication model to perform feature authentication, to obtain a feature authentication result corresponding to the sample feature information;

adjustment of a model parameter of the initial authentication model based on the feature authentication result corresponding to the sample feature information and the sample authentication result corresponding to the sample feature information until a difference between the feature authentication result obtained through feature authentication based on the sample feature information and the sample authentication result corresponding to the sample feature information meets a preset condition; and

determination of an authentication model that meets the preset condition as the biometric authentication model corresponding to the biometric feature information.

9 . The apparatus according to claim 7 , wherein

the at least two types of biometric feature information correspond to one comprehensive biometric authentication model;

the sample data comprises at least two types of sample feature information and a sample authentication result corresponding to the at least two types of sample feature information, and each type of sample feature information is one type of biometric feature information; and

the performance of training based on the sample data to generate the biometric authentication model comprises:

input of the at least two types of sample feature information into an initial authentication model to perform feature authentication, to obtain a feature authentication result corresponding to the at least two types of sample feature information;

adjustment of a model parameter of the initial authentication model based on the feature authentication result corresponding to the at least two types of sample feature information and the sample authentication result corresponding to the at least two types of sample feature information until a difference between the feature authentication result obtained through feature authentication based on the at least two types of sample feature information and the sample authentication result corresponding to the at least two types of sample feature information meets a preset condition; and

determination of the authentication model that meets the preset condition as the biometric authentication model corresponding to the at least two types of biometric feature information.

10 . The apparatus according to claim 6 , wherein

the at least two types of biometric feature information comprise at least two types of ear canal feature information, auricle feature information, vibration feature information, voiceprint feature information, facial feature information, fingerprint feature information, and iris feature information; and the vibration feature information is vibration information of a part on which the wearable device is worn.

11 . A device enabling system comprising:

at least one of a wearable device and a smart device, and the at least one of the wearable device or the smart device comprises a device enabling apparatus, wherein the device enabling apparatus comprises a processor and a memory; the memory stores a program; and the processor is configured to execute the program stored in the memory, so that the device enabling apparatus is at least configured to:

collect at least two types of biometric feature information of a user;

perform identity authentication on the user based on the at least two types of biometric feature information; and

when the identity authentication succeeds, enable at least one of the wearable device or the smart device, wherein the wearable device is communicatively connected to the smart device,

wherein;

the performance of the identity authentication on the user based on the at least two types of biometric feature information comprises:

performance of biometric authentication based on the at least two types of biometric feature information and a biometric authentication model; and

determination of an identity authentication result of the user based on a biometric authentication result of the biometric authentication;

each type of the at least two types of biometric feature information corresponds to one of biometric authentication models, respectively, and each of the corresponding biometric authentication models is used to perform biometric authentication based on the corresponding biometric feature information:

the performance of the biometric authentication based on the at least two types of biometric feature information and the biometric authentication model comprises:

performance of biometric authentication based on each type of biometric feature information and the corresponding biometric authentication models, to respectively obtain biometric authentication results correspondi o the biometric feature information; and

the determination of the identity authentication result of the user based on the biometric authentication result comprises:

determination of the identity authentication result of the user based on the biometric authentication results respectively corresponding to the at least two types of biometric feature information;

the determination of the identity authentication result of the user based on the biometric authentication results corresponding to the at least two types of biometric feature information comprises: performance of a weighted calculation on the biometric authentication results corresponding to the at least two types of biometric feature information, and determining the identity authentication result of the user based on a result obtained through the weighted calculation, wherein a sum of weights corresponding to the at least two types of biometric feature information is 1, and a weight corresponding to each type of biometric feature information is adjustable.

12 . The device enabling system according to claim 11 , wherein the wearable device comprises either of a smart headset or smart glasses.

Assignments (1)

Priority Claims (1)

Continuity (2)

References Cited (29)