KR20140117828A - A device and method for electroencephalography signal correction - Google Patents

Abstract

The present invention relates to an electroencephalography signal correction device. Particularly, the electroencephalography signal correction device according to one embodiment of the present invention includes: an electroencephalography signal measurement module which measures an electroencephalography signal of a user; a user movement recognition module which recognizes the movement of a user; a classification determination module which determinates a noise through the classification of the movement of the user according to the recognized user movement of the user movement recognition module, and determinates a noise; and an electroencephalography signal correction module which corrects the measured electroencephalography signal depending on the determination result of the classification determination module.

Description

TECHNICAL FIELD [0001] The present invention relates to a device for correcting an EEG signal,

The present invention relates to an apparatus and method for correcting an EEG signal, and more particularly, to an apparatus and method for correcting an EEG signal by user's movement through analysis of an EEG signal.

Recently, as smart devices such as smart phones and smart TVs have been developed, digital convergence aiming at natural fusion between devices and humans has evolved into intelligent convergence where human emotions, intelligence, and imagination are combined with systems . Accordingly, there is a demand for a new concept multi-dimensional human interface technique for acquiring and processing the intention information of a user. In order to acquire the intention information of such a user, it has become possible to grasp the potential and unconscious intentions of the user by analyzing biometric information such as brain waves.

However, when an EEG signal is measured by an existing method, an unnecessary noise is generated in order to grasp a user's intention according to an internal factor such as a movement of a user's head, which may adversely affect analysis of an EEG signal. Therefore, there is a need for research to correct the noise generated in EEG signal measurement.

Disclosure of Invention Technical Problem [8] The present invention has been devised to solve the problems described above, and its object is to detect and correct noise generated by a user's movement in EEG signal measurement, have.

The technical objects to be achieved by the present invention are not limited to the above-mentioned technical problems, and other technical subjects which are not mentioned can be clearly understood by those skilled in the art from the description of the present invention .

According to another aspect of the present invention, there is provided an apparatus for correcting an EEG signal, comprising: an EEG signal measurement module for measuring a user's EEG signal; a user movement recognition module for recognizing a user movement; A classification discrimination module for classifying the user's movement and discriminating noise, and an EEG signal correction module for correcting the measured EEG signal according to the discrimination result of the classification discrimination module. The EEG signal measurement module includes an EEG signal acquisition module for acquiring a user's EEG signal and an EEG signal feature vector extraction module for extracting a feature vector from the EEG signal acquired by the EEG acquisition module. The user motion recognition module includes a user motion information acquisition module for acquiring user motion information, and a user motion feature vector extraction module for extracting user motion feature vectors from the obtained user motion information. The user motion feature vector extraction module extracts a user motion feature vector when the user motion is equal to or greater than a predetermined value. The user motion information acquisition module includes a plurality of piezoelectric elements, a gyro sensor, and a camera. The user motion feature vector extraction module includes a piezoelectric element feature vector extraction module for extracting a piezoelectric device feature vector according to a change in voltage of a plurality of piezoelectric elements, a gyro sensor feature vector extraction unit for extracting a gyro sensor feature vector according to the angular velocity information of the gyro sensor, Module and a camera feature vector extraction module for extracting a camera feature vector according to movement information of a center point of the user's face photographed by the camera. The classification discrimination module includes a support vector machine classifier for classifying feature vectors and a noise discrimination module for discriminating noises according to classification results of a support vector machine classifier. The noise discrimination module assigns a weight to each of the piezoelectric element feature vector, the gyro sensor feature vector, the camera feature vector, and the EEG signal feature vector to discriminate the noise.

The EEG signal correcting method according to an embodiment of the present invention may include a step of measuring an EEG signal measuring a user's EEG signal, a user's motion recognition step of recognizing a user's movement, A classification discrimination step of discriminating noise, and an EEG signal correction step of correcting the measured EEG signal according to the discrimination result in the classification discrimination step. The step of measuring the EEG signal includes an EEG signal acquisition step of acquiring a user's EEG signal and an EEG signal feature vector extraction step of extracting a feature vector from the EEG signal acquired in the EEG acquisition step. The user motion recognition step includes a user motion information acquisition step of acquiring user motion information, and a user motion feature vector extraction step of extracting a user motion feature vector from the acquired user motion information. The user motion feature vector extraction step includes extracting a user motion feature vector when the user motion is equal to or greater than a predetermined value. The user motion feature vector extraction step includes a piezoelectric element feature vector extraction step for extracting a piezoelectric element feature vector according to a voltage change of a plurality of piezoelectric elements, a piezoelectric element feature vector extraction step for extracting a gyro sensor feature vector according to the angular velocity information of the gyro sensor And a piezoelectric element feature vector extraction step of extracting a camera feature vector according to movement information of a center point of the user's face photographed by the camera. The classification discrimination step includes a feature vector classification step of classifying the feature vectors, and a noise discrimination step of discriminating noises according to classification results in the feature vector classification step. The noise discrimination step includes weighting each of the piezoelectric element feature vector, the gyro sensor feature vector, the camera feature vector, and the EEG signal feature vector to discriminate the noise.

According to the present invention, when measuring and analyzing an EEG signal, it is possible to detect and correct noise generated according to a user's motion when measuring an EEG signal, thereby improving the reliability of EEG signal analysis results.

FIG. 1 schematically shows a configuration of an EEG signal correcting apparatus according to an embodiment of the present invention.
FIG. 2 is a schematic diagram illustrating a configuration of an EEG signal measurement module according to an embodiment of the present invention.
FIG. 3 is a block diagram illustrating a configuration of a user motion recognition module according to an embodiment of the present invention.
FIG. 4 is a schematic diagram illustrating a configuration of a user motion information acquisition module according to an embodiment of the present invention.
FIG. 5 is a schematic diagram illustrating a structure of a user motion feature vector extraction module according to an embodiment of the present invention.
FIG. 6 is a diagram schematically illustrating a configuration of a classification discrimination module according to an embodiment of the present invention.
7 is a flowchart illustrating a process of correcting brain waves in an EEG apparatus according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art to which the present invention pertains. The present invention may be embodied in many different forms and is not limited to the embodiments described herein.

In order to clearly illustrate the present invention, parts not related to the description are omitted, and the same or similar components are denoted by the same reference numerals throughout the specification.

Throughout the specification, when a part is referred to as being "connected" to another part, it includes not only "directly connected" but also "electrically connected" with another part in between . Also, when an element is referred to as "comprising ", it means that it can include other elements as well, without departing from the other elements unless specifically stated otherwise.

If any part is referred to as being "on" another part, it may be directly on the other part or may be accompanied by another part therebetween. In contrast, when a section is referred to as being "directly above" another section, no other section is involved.

The terms first, second and third, etc. are used to describe various portions, components, regions, layers and / or sections, but are not limited thereto. These terms are only used to distinguish any moiety, element, region, layer or section from another moiety, moiety, region, layer or section. Thus, a first portion, component, region, layer or section described below may be referred to as a second portion, component, region, layer or section without departing from the scope of the present invention.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms as used herein include plural forms as long as the phrases do not expressly express the opposite meaning thereto. Means that a particular feature, region, integer, step, operation, element and / or component is specified and that the presence or absence of other features, regions, integers, steps, operations, elements, and / It does not exclude addition.

Terms indicating relative space such as "below "," above ", and the like may be used to more easily describe the relationship to other portions of a portion shown in the figures. These terms are intended to include other meanings or acts of the apparatus in use, as well as intended meanings in the drawings. For example, when inverting a device in the figures, certain parts that are described as being "below" other parts are described as being "above " other parts. Thus, an exemplary term "below" includes both up and down directions. The device can be rotated by 90 degrees or rotated at different angles, and terms indicating relative space are interpreted accordingly.

Unless otherwise defined, all terms including technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Commonly used predefined terms are further interpreted as having a meaning consistent with the relevant technical literature and the present disclosure, and are not to be construed as ideal or very formal meanings unless defined otherwise.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein.

FIG. 1 schematically shows a configuration of an EEG signal correcting apparatus 100 according to an embodiment of the present invention. Referring to FIG.

Referring to FIG. 1, an apparatus for correcting an EEG signal 100 according to an exemplary embodiment of the present invention includes an EEG signal measurement module 200, a user motion recognition module 300, a classification discrimination module 400, 500).

The EEG signal measurement module 200 measures a user's brain wave signal. Specifically, the EEG signal measurement module 200 is configured to acquire a user's EEG signal and extract an EEG signal feature vector from the obtained EEG signal. Details will be described later with reference to Fig.

The user motion recognition module 300 is configured to acquire user motion information and extract a user motion feature vector from the acquired user motion information. Details will be described later with reference to Fig.

The classification discrimination module 400 is configured to discriminate a noise according to a user's movement recognized by the user's motion recognition module 300. Details will be described later with reference to Fig.

The EEG signal correction module 500 is configured to correct the measured EEG signal according to the discrimination result of the classification discrimination module 400.

FIG. 2 is a diagram schematically illustrating a configuration of an EEG signal measurement module 200 according to an embodiment of the present invention.

Referring to FIG. 2, the EEG signal measurement module 200 includes an EEG signal acquisition module 210 and an EEG signal feature vector extraction module 220.

The EEG signal acquisition module 210 acquires a user's EEG signal. Specifically, the EEG signal acquisition module 210 acquires a user's EEG signal using an EEG signal measuring device. The EEG signal measuring module measures EEG waves from a plurality of channels, and the EEG signal acquisition module 210 can acquire an EEG signal based on the EEG measured from the EEG measurement device. The number of the plurality of channels may be 14, but is not limited thereto. The EEG signal measuring device uses commercially available equipment and can extract 128 EEG signals per second, but the present invention is not limited thereto.

The EEG signal feature vector extraction module 220 extracts an EEG signal feature vector from the EEG signal acquired by the EEG acquisition module 210. Specifically, the EEG signal feature vector extracting module 220 calculates an average EEG amplitude of the plurality of EEG signals obtained by the EEG signal acquisition module 210, a standard deviation of the amplitude of the EEG signal, , The standard deviation of the first derivatives of the front and back inputs to the amplitude of the brain wave signal, the second moment of the amplitude of the brain wave signal, and the skewness of the amplitude of the brain wave signal, .

The above-described average and standard deviation use the average of 16 EEG signals, so that EEG signal feature vectors can obtain one feature vector at 8 seconds per second, that is, 0.125 seconds. As described above, since the first-order change amount and the second-order change amount are calculated for the EEG signal values of the fourteen channels and used as the values of the feature vectors, one feature vector is (first change amount and second change amount) X 14 channels) with a total of 28 dimensions.

That is, the EEG signal feature vector X1 is as follows.

X1 = (x11, x12, x13, x14, x15, x16)

 (Where x11 is the average of the amplitude of the EEG signal, x12 is the standard deviation of the amplitude of the EEG signal, x13 is the average of the first derivative values of the front and back inputs to the amplitude of the EEG signal, and x14 is the amplitude of the EEG signal. X15 is the second moment of the amplitude of the EEG signal, and x16 is the skewness of the amplitude of the EEG signal.

FIG. 3 schematically shows a configuration of a user motion recognition module 300 according to an embodiment of the present invention.

Referring to FIG. 3, the user motion recognition module 300 recognizes the movement of the user. Specifically, the user motion recognition module 300 includes a user motion information acquisition module 310 and a user motion feature vector extraction module 320.

The user motion information acquisition module 310 acquires user motion information. Details will be described later with reference to Fig.

The user motion feature extraction module 320 extracts a user motion feature vector from the acquired user motion information. Specifically, the user motion feature vector extracting module 320 extracts a user motion feature vector from the user motion information when the degree of user motion included in the user motion information acquired by the user motion information acquiring module 310 is equal to or greater than a predetermined value, . Noise is likely to occur in the brain waves due to user's movement when the degree of user's motion is greater than a predetermined value.

FIG. 4 is a block diagram illustrating a configuration of a user motion information acquisition module 310 according to an embodiment of the present invention.

Referring to FIG. 4, the user motion information acquisition module 310 includes a plurality of piezoelectric elements 312, a gyro sensor 314, and a camera 316.

The plurality of piezoelectric elements 312 are attached to each node of the EEG instrument and the scalp of the user so that each of the nodes of the EEG instrument and the user's scalp are connected one by one. The voltage of the piezoelectric element 312 changes according to the contact between the user's scalp and the piezoelectric element 312 and the piezoelectric element feature vector can be extracted in accordance with the change in the voltage of the piezoelectric element 312. [

Specifically, the average of the voltage change of the piezoelectric element 312, the standard deviation of the voltage change of the piezoelectric element 310, the average of the first-order differential values of the input voltage of the piezoelectric element 312, By using the standard deviation of the first derivative values of the voltage, the average of the second derivative values of the input voltage of the piezoelectric element 312, and the standard deviation of the second derivative values of the input voltage of the piezoelectric element 312, .

In order to synchronize with the EEG signal feature vector X1, the above-described average and standard deviation are extracted eight times per second. To do so, the average and standard deviation of the average and standard deviation may be measured by measuring the voltage change every 0.125 second.

That is, the piezoelectric element feature vector X2 is as follows.

X2 = (x21, x22, x23, x24, x25, x26)

 X23: an average of the first-order differential values of the input voltages of the piezoelectric elements 312, x24 (the average of the first-order differential values of the input voltages of the piezoelectric elements 312) X25 is an average of the second derivative values of the input voltage of the piezoelectric element 312, x26 is a second derivative of the input voltage of the piezoelectric element 312 Lt; / RTI >

The gyro sensor 314 acquires the angular velocity information according to the user's motion, and the gyro sensor characteristic vector can be extracted according to the obtained angular velocity information.

Specifically, the average of the rotation angle obtained by the gyro sensor 314, the standard deviation of the rotation angle, the average of the first differential values of the rotation angle, the standard deviation of the first differential values of the rotation angle, The gyro sensor feature vector is extracted using the standard deviation of the mean values of the values and the second derivative values of the rotation angle.

For the purpose of synchronizing the EEG signal characteristic vector X1 and the piezoelectric element characteristic vector X2, the above-described average and standard deviation are extracted eight times per second. To do so, the voltage variation is measured every 0.125 second, And standard deviation.

That is, the gyro sensor characteristic vector X3 is as follows.

X3 = (x31, x32, x33, x34, x35, x36)

 X32 is the standard deviation of the rotation angle, x33 is the average of the first derivative values of the rotation angle, x34 is the standard deviation of the first derivative values of the rotation angle, and x35 is the rotation angle. , And x36: the standard deviation of the second derivative values of the rotation angle)

The camera 316 can continuously photograph the face of the user and extract the camera feature vector according to the movement information of the center point of the captured user face. However, it is not necessary to adopt the movement information of the center point of the photographed user's face.

Specifically, the average of the first-order differential values of the positional change of the center point of the user's face obtained by the camera 316, the standard deviation of the first-order differential values of the positional change of the center point, , And the camera feature vector is extracted using the standard deviation of the second-order differential values of the positional change of the center point.

In order to synchronize the EEG signal feature vector X1, the piezoelectric element feature vector X2, and the gyro sensor feature vector X3, the above-described average and standard deviation are extracted eight at a time, And the average and standard deviation of the voltage may be used.

That is, the camera feature vector X4 is as follows.

X4 = (x41, x42, x43, x44)

 (Where x41 is the average of the first derivative values of the positional change of the center point of the user face, x42 is the standard deviation of the first derivative values of the positional change of the center point, x43 is the average of the second derivative values of the positional change of the center point, x44: the standard deviation of the second derivative values of the positional change of the center point)

FIG. 5 schematically shows a structure of a user motion feature vector extraction module 320 according to an embodiment of the present invention.

Referring to FIG. 5, the user motion feature vector extraction module 320 includes a piezoelectric device feature vector extraction module 322, a gyro sensor feature extraction module 324, and a camera feature vector extraction module 326.

The piezoelectric element feature vector extraction module 322 extracts the piezoelectric element feature vector according to the change in the voltage of the piezoelectric element 312. The gyro sensor characteristic vector extracting module 324 extracts the gyro sensor characteristic vector according to the angular velocity information according to the user's motion. The camera feature vector extraction module 326 extracts the camera feature vector according to the movement information of the center point of the user's face.

FIG. 6 schematically shows a configuration of a classification discrimination module 400 according to an embodiment of the present invention.

Referring to FIG. 6, the classification discrimination module 400 includes a SVM (Support Vector Machine) classifier 410 and a noise discrimination module 420.

The SVM classifier 410 is configured to classify feature vectors obtained by the EEG signal feature extraction module 220 and the user motion feature extraction module 320 according to a known technique.

Specifically, the SVM classifier is a function that finds a hyperplane that can ideally divide two sets of assumptions, assuming that there are two sets of feature vectors. The maximum margin ) Are classified into feature vectors through the process of formulization.

More specifically, the SVM classifier is a method of finding a hyperplane that is farthest from the data, among the hyperplanes separating the data for a given data. Many learning algorithms, including neural networks, have a common goal of finding a hyperplane that separates points 1 and -1 when given such learning data. A feature that differentiates SVM classifiers from other algorithms is that they only divide points , But finds the maximum-margin hyperplane among the many candidate planes that can separate the points. In this case, the margin refers to the minimum value of the distance from the hyperplane to each point. To classify the points into two classes while maximizing the margin, the distance between the minimum distance from the points belonging to class 1 and the points belonging to class- And the hyperplane is called the maximum-margin hyperplane. In conclusion, the SVM classifier is an algorithm that finds out of a large number of hyperplanes that classify the points belonging to two classes, maintaining at most two classes of points and distances.

The noise discrimination module 420 is configured to discriminate whether noise is present according to the classification result of the SVM classifier 410. More specifically, a plurality of SVM classifiers are used to assign different weighting coefficients to the values calculated by the respective classifiers according to the reliability of the apparatus, thereby generating a final variable Y for judging whether noise is present or absent.

That is, the classification result Y1 obtained from the piezoelectric element feature vector, the classification result Y2 obtained from the gyro sensor characteristic vector, the classification result Y3 obtained from the camera characteristic vector, and the EEG signal feature vector classification result Y4 The final variable Y is calculated on the basis of the threshold value p and it is finally determined whether or not noise is present.

Specifically, Y = 1 (noise) is discriminated when? Y1 +? Y2 +? Y3 +? Y4>?, And in the other cases, Y = 0 (no noise). Here,?,?,?, And? Are the weighting values of? +? +? +? = 1 and?,?,?, And?> 0, , The final variable Y can be calculated by weighting the classification result Y2 obtained from the gyro sensor feature vector, the classification result Y3 obtained from the camera feature vector, and the EEG signal feature vector classification result Y4 .

For the EEG signal discriminated as noise, the EEG signal correction module 500 corrects the EEG signal. The EEG signal correction module 500 may correct the EEG signal through Gaussian filtering, but is not limited thereto.

The reason for using the Gaussian function as a filter is that, in low pass filtering, various types of weighting functions are utilized to remove noise and preserve the original signal shape to the maximum. However, The weighted Gaussian function is used as a weighted value, and the noise reduction performance is very excellent. In addition, since the intensity of the filter can be adjusted by the average and standard deviation of the parameters of the Gaussian function, it can be very useful in terms of user convenience.

7 is a flowchart illustrating a process of correcting brain waves in the EEG 100 according to an embodiment of the present invention.

Referring to FIG. 7, the brain wave correcting apparatus 100 acquires an EEG signal (S10), acquires user motion information (S20), and if the user motion included in the user motion information is greater than or equal to a predetermined value (S30) A user motion feature vector is extracted in step S40, an EEG signal feature vector is extracted in step S50, a noise is determined on the basis of the extracted feature vector in step S60, a noise is determined in step S70, S80).

Specifically, the EEG signal acquisition module 210 acquires a user's brain wave signal (S10), and the user motion information acquisition module 310 acquires motion information of the user (S20) If it is determined that the user's motion is greater than or equal to a predetermined value (S30), that is, if noise is thought to have occurred in the brain waves, the user motion feature extraction module 320 extracts a user motion feature vector (S40) The EEG signal extracting module 220 extracts the EEG signal feature vector in step S50 and discriminates the noise in the classification discrimination module 400 in step S60. (S80).

Although the present invention has been described in connection with the specific embodiments of the present invention, it is to be understood that the present invention is not limited thereto. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. In addition, the materials of each component described in the specification can be easily selected and substituted for various materials known to those skilled in the art. Those skilled in the art will also appreciate that some of the components described herein can be omitted without degrading performance or adding components to improve performance. In addition, those skilled in the art may change the order of the method steps described herein depending on the process environment or equipment. Therefore, the scope of the present invention should be determined by the appended claims and equivalents thereof, not by the embodiments described.

100: EEG signal correction device
200: EEG signal measurement module
210: EEG signal acquisition module
220: EEG signal feature vector extraction module
300: user motion recognition module
310: User motion information acquisition module
312: piezoelectric element
314: Gyro sensor
316: Camera
320: user motion feature vector extraction module
322: Piezoelectric device characteristic vector extraction module
324: Gyro sensor feature vector extraction module
326: camera feature vector extraction module
400: Classification discrimination module
410: SVM classifier
420: noise discrimination module

Claims (15)

An apparatus for correcting an EEG signal,
An EEG signal measuring module for measuring a user's EEG signal;
A user motion recognition module for recognizing user movement;
A classification discrimination module for classifying the user's movement according to the user's movement recognized by the user's movement recognition module to discriminate noise; And
According to the discrimination result of the classification discrimination module, an EEG signal correction module
And an EEG signal correcting device. The method according to claim 1,
The EEG signal measuring module includes:
An EEG signal acquisition module for acquiring a brain wave signal of a user; And
An EEG signal feature extraction module for extracting a feature vector from the EEG signal acquired by the EEG acquisition module,
And an EEG signal correcting device. The method according to claim 1,
Wherein the user motion recognition module comprises:
A user motion information acquisition module for acquiring user motion information; And
A user motion feature extraction module for extracting a user motion feature vector from the obtained user motion information,
And an EEG signal correcting device. The method of claim 3,
Wherein the user motion feature vector extraction module comprises:
And extracts the user motion feature vector when the user motion is equal to or greater than a predetermined value. 5. The method of claim 4,
The user motion information acquisition module includes:
A plurality of piezoelectric elements, a gyro sensor, and a camera. 6. The method of claim 5,
Wherein the user motion feature vector extraction module comprises:
A piezoelectric element feature vector extracting module for extracting a piezoelectric element feature vector according to a voltage change of the plurality of piezoelectric elements;
A gyro sensor feature vector extraction module for extracting a gyro sensor feature vector according to the angular velocity information of the gyro sensor; And
A camera feature vector extraction module for extracting a camera feature vector according to movement information of a center point of the user's face photographed by the camera;
And an EEG signal correcting device. The method according to claim 6,
Wherein the classification discrimination module comprises:
A support vector machine classifier for classifying feature vectors; And
A noise discrimination module for discriminating noise according to a classification result of the support vector machine classifier;
And an EEG signal correcting device. 8. The method of claim 7,
The noise discrimination module comprises:
And a weight is assigned to each of the piezoelectric element feature vector, the gyro sensor characteristic vector, the camera feature vector, and the EEG signal feature vector to discriminate noise. A method for correcting an EEG signal,
An EEG signal measuring step of measuring a user's EEG signal;
A user movement recognition step of recognizing a user movement;
A classification discrimination step of classifying the user's movement according to the user's movement recognized in the user's movement recognition step to discriminate noise; And
According to the discrimination result in the class discriminating step, an EEG signal correction step for correcting the measured EEG signal
And an EEG signal correction method. 10. The method of claim 9,
The EEG signal measuring step may include:
An EEG signal acquisition step of acquiring a brain wave signal of a user; And
Extracting an EEG signal feature vector extracting a feature vector from the EEG signal acquired in the EEG signal acquisition step
And an EEG signal correction method. 10. The method of claim 9,
Wherein the user motion recognition step comprises:
A user motion information acquiring step of acquiring user motion information; And
A user motion feature vector extraction step of extracting a user motion feature vector from the acquired user motion information
And an EEG signal correction method. 12. The method of claim 11,
Wherein the user motion feature vector extraction step comprises:
And extracting the user motion feature vector if the user motion is greater than or equal to a predetermined value. 13. The method of claim 12,
Wherein the user motion feature vector extraction step comprises:
A piezoelectric element characteristic vector extracting step of extracting a piezoelectric element characteristic vector according to a voltage change of the plurality of piezoelectric elements;
A piezoelectric element feature vector extracting step of extracting a gyro sensor feature vector according to the angular velocity information of the gyro sensor; And
Extracting a feature vector of a camera according to movement information of a center point of a user's face photographed by the camera,
And an EEG signal correction method. 14. The method of claim 13,
Wherein the classifying step comprises:
A feature vector classifying step of classifying the feature vector; And
A noise discrimination step of discriminating noise according to a result of classification in the feature vector classification step
And an EEG signal correction method. 15. The method of claim 14,
Wherein the noise discrimination step comprises:
And assigning a weight to each of the piezoelectric element feature vector, the gyro sensor characteristic vector, the camera feature vector, and the EEG signal feature vector to discriminate noise.

Images