Granted Patent B2

US 12,204,624 · App. 17/844,870 · Granted Jan 21, 2025

Inventors: Seong-Whan Lee (Seoul, KR); Young-Eun Lee (Seongnam-si, KR); Seo-Hyun Lee (Seoul, KR)

Assignee: Korea University Research and Business Foundation

G06F21/32A61B5/117A61B5/374A61B5/378A61B5/38A61B5/7267G06F3/015

Abstract

Claims (71)

1. A user authentication method implemented through communication connection between a brainwave measurement device for measuring a brainwave signal according to an imagined speech and a user authentication apparatus, the user authentication method comprising:

extracting and storing characteristic information of a first brainwave signal generated by an imagined speech of a first user based on preset imagined speech induction information provided to the first user by the user authentication apparatus; and

determining whether a second user matches the first user by extracting characteristic information of a second brainwave signal generated by an imagined speech of the second user, based on preset imagined speech induction information provided to the second user, comparing the characteristic information of the first brainwave signal with the characteristic information of the second brainwave signal stored according to the extracting and storing, and analyzing a result of the comparing by the user authentication apparatus,

wherein the determining comprises evaluating similarity between the characteristic information of the first brainwave signal and the characteristic information of the second brainwave signal based on a time-frequency graph of each brainwave signal, and

wherein each brainwave signal is acquired by using only a single brainwave signal measurement channel.

2. The user authentication method of claim 1 , wherein the preset imagined speech induction information includes at least one of auditory information and visual information on a language composed of at least one character.

3. The user authentication method of claim 1 , wherein the first brainwave signal includes one brainwave signal measurement channel and is a signal of a specific pattern generated in at least one of a Broca's brain area and a Wernicke's brain area of the first user measured by the brainwave measurement device disposed in a region around an ear of the first user, and

wherein the second brainwave signal is a signal of a specific pattern generated in at least one of a Broca's brain area and a Wernicke's brain area of the second user measured by the brainwave measurement device disposed in a region around an ear of the second user.

4. The user authentication method of claim 1 , wherein the extracting and storing includes:

providing, by the user authentication apparatus, the preset imagined speech induction information to the first user,

measuring, by the brainwave measurement device, the first brainwave signal according to the imagined speech of the first user, based on the preset imagined speech induction information, and

constructing a user classifier by receiving the first brainwave signal from the brainwave measurement device and extracting and storing the characteristic information of the first brainwave signal by the user authentication apparatus.

5. The user authentication method of claim 4 , wherein the providing includes:

providing a guide voice for the preset imaged speech induction information or visually displaying a guide phrase for the preset imaged speech induction information by the user authentication apparatus such that the first user performs an imaged speech according to the preset imaged speech induction information.

6. The user authentication method of claim 4 , wherein the measuring includes:

collecting and storing the first brainwave signal for a specific time, according to the imagined speech of the first user, based on the preset imagined speech induction information, and

transmitting the first brainwave signal to the user authentication apparatus by the brainwave measurement device.

7. The user authentication method of claim 4 , wherein the constructing includes:

extracting characteristic information including time-frequency characteristics of the first brainwave signal transmitted from the brainwave measurement device by the user authentication apparatus by utilizing a deep learning model designed to extract and classify time-frequency characteristics of brainwave signals, and

constructing a user classifier based on the extracted characteristic information.

8. The user authentication method of claim 1 , wherein the determining includes:

providing, by the user authentication apparatus, the preset imagined speech induction information to the second user,

measuring, by the brainwave measurement device, the second brainwave signal according to an imagined speech of the second user, based on the preset imagined speech induction information, and

determining whether the second user matches the first user by receiving the second brainwave signal from the brainwave measurement device,

extracting the characteristic information of the second brainwave signal,

comparing the characteristic information of the second brainwave signal with the characteristic information of the first brainwave signal, and

analyzing a result of the comparing by the user authentication apparatus.

9. The user authentication method of claim 8 , wherein the providing includes:

providing a guide voice for the preset imaged speech induction information or visually display a guide phrase for the preset imaged speech induction information by the user authentication apparatus such that the second user performs an imaged speech according to the preset imaged speech induction information.

10. The user authentication method of claim 8 , wherein the measuring includes:

collecting and storing the second brainwave signal for a specific time, according to the imagined speech of the second user, based on the preset imagined speech induction information, and

transmitting the second brainwave signal to the user authentication apparatus by the brainwave measurement device.

11. The user authentication method of claim 8 , wherein the determining includes:

extracting the characteristic information including time-frequency characteristics of the second brainwave signal transmitted from the brainwave measurement device by utilizing a deep learning model designed to extract and classify the time-frequency characteristics of first brainwave signals, and

performing user classification and user authentication based on the extracted characteristic information by the user authentication apparatus.

12. The user authentication method of claim 1 , wherein the determining includes:

determining that the second user matches the first user when the similarity is greater than or equal to a preset value, and

determining that the second user does not match the first user when the similarity is less than the preset value.

13. A user authentication apparatus using a brainwave measurement device for measuring brainwave signals according to an imagined speech, the user authentication apparatus comprising:

the brainwave measurement device including one brainwave signal measurement channel and disposed in a region around an ear of a user to measure the brainwave signals according to an imagined speech of the user;

a communication module comprising network devices including hardware and software configured to receive measured brainwave signals from the brainwave measurement device;

a memory storing a user authentication program using the brainwave signals; and

a processor configured to:

execute the user authentication program using the brain wave signals stored in the memory,

provide preset imagined speech induction information to a first user,

receive a first brainwave signal according to an imagined speech of the first user performed based on the preset imagined speech induction information from the brainwave measurement device through the communication module,

extract and store characteristic information of the first brainwave signal, and

perform user classification, and

wherein the processor is configured to:

provide preset imagined speech induction information to a second user,

receive a second brainwave signal according to an imagined speech of the second user performed based on the preset imagined speech induction information from the brainwave measurement device through the communication module,

extract characteristic information of the second brainwave signal,

compare the characteristic information of the first brainwave signal with the characteristic information of the second brainwave signal,

analyze a result of the comparison, and

determine whether the second user matches the first user by evaluating similarity between the characteristic information of the first and second brainwave signals based on a time-frequency graph of each brainwave signal,

wherein each brainwave signal is acquired by using only a single brainwave signal measurement channel.

14. The user authentication apparatus of claim 13 , wherein the preset imagined speech induction information includes at least one of auditory information and visual information on a language composed of at least one character.

15. The user authentication apparatus of claim 13 , wherein the first brainwave signal is a signal of a specific pattern generated in at least one of a Broca's brain area and a Wernicke's brain area of the first user measured by the brainwave measurement device, and

wherein the second brainwave signal is a signal of a specific pattern generated in at least one of a Broca's brain area and a Wernicke's brain area of the second user measured by the brainwave measurement device.

16. The user authentication apparatus of claim 13 , wherein the processor executes the user authentication program using the brain wave signals,

wherein the user authentication apparatus provides a guide voice for the preset imaged speech induction information or visually displays a guide phrase for the preset imaged speech induction information such that the first user performs an imaged speech according to the preset imaged speech induction information, and

wherein the processor extracts characteristic information including time-frequency characteristics of the first brainwave signal received from the brainwave measurement device through the communication module by utilizing a deep learning model designed to extract and classify time-frequency characteristics of brainwave signals and further perform user classification based on the extracted characteristic information.

17. The user authentication apparatus of claim 13 ,

wherein the processor executes the user authentication program using the brain wave signal,

wherein the user authentication apparatus provides a guide voice for the preset imaged speech induction information or visually displays a guide phrase for the preset imaged speech induction information such that the second user performs an imaged speech according to the preset imaged speech induction information, and

wherein the processor extracts characteristic information including time-frequency characteristics of the first brainwave signal received from the brainwave measurement device through the communication module by utilizing a deep learning model designed to extract and classify time-frequency characteristics of brainwave signals and performs user classification and user authentication based on the extracted characteristic information.

18. The user authentication apparatus of claim 13 , wherein the processor is configured to:

determine that the second user matches the first user when the similarity is greater than or equal to a preset value, and

determine that the second user does not match the first user when the similarity is less than the preset value.

19. A non-transitory computer-readable recording medium comprising:

a computer program for performing the user authentication method using the brainwave signal according to the imagined speech, according to claim 1 .

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (13)