Emotion recognition data set construction method and system based on brain wave mapping value and application

Abstract

The invention relates to the technical field of brain waves, and provides a brain wave mapping value-based emotion recognition data set construction method and system and application. The method comprises the following steps: S1, acquiring brain wave mapping values through a plurality of sampling points; s2, according to the acquired brain wave mapping values, separating out alpha waves, beta waves and gamma waves of the brain waves, and according to the values of the alpha waves, the beta waves and the gamma waves in each sampling interval, judging whether the sampling interval is a first emotional state, a second emotional state or a third emotional state according to a set criterion; and S3, collecting the brain wave mapping values according to different emotional states, and removing abnormal data from the collected data of each emotional state to obtain a brain wave emotion recognition data set. According to the method, a solution is provided for removing abnormal values in different emotion brain waves, and the problems that brain wave emotion recognition data is difficult to construct and multiple abnormal values exist are solved.

Description

Technical Field

[0001] The present invention relates to the field of brain wave technology, and in particular to a method, system and application for constructing an emotion recognition data set based on brain wave mapping values. Background Art

[0002] Human emotions are closely related to brain activity, and brain electrical activity can be monitored through EEG. EEG reflects the physiological state of the brain, and different emotional states can cause different patterns of brain electrical activity. Emotion recognition is very important in many application scenarios, such as medical health monitoring, human-computer interaction, education and training, and safety assessment. There are a large number of outliers in the existing EEG datasets, which leads to deviations in the constructed machine learning and deep learning models, affecting the generalization ability of the model and the interpretability of the results.

[0003] In response to the above problems, the present invention proposes a method for constructing a brain wave emotion recognition dataset to automatically remove brain wave outliers and construct a high-quality dataset. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method, system and application for constructing an emotion recognition dataset based on EEG mapping values, which automatically removes outliers in EEG data and constructs a high-quality dataset.

[0005] The present invention adopts the following technical solution:

[0006] In one aspect, the present invention provides a method for constructing an emotion recognition dataset based on brain wave mapping values, comprising:

[0007] S1, collecting brain wave mapping values ​​through multiple sampling points;

[0008] S2, separating the alpha wave, beta wave, and gamma wave of the brain wave according to the collected brain wave mapping values, and judging the sampling interval as the first emotional state, the second emotional state, or the third emotional state according to the values ​​of the alpha wave, beta wave, and gamma wave in each sampling interval according to the set criteria;

[0009] S3. Collect brain wave mapping values ​​according to different emotional states, remove abnormal data from the collected data of each emotional state, and obtain a brain wave emotion recognition data set.

[0010] According to the above aspects and any possible implementation, an implementation is further provided, in step S2, multiple sampling points are arranged in time sequence, and each m1 point is a sampling interval, and there are n sampling intervals in total;

[0011] The criteria for judging whether the sampling interval is the second emotional state are as follows: if there are m2 or more first peak points of the β wave in the sampling interval, the sampling interval is judged to be the second emotional state; the first peak point is the sampling point where the β wave mapping value is greater than the mean of the β wave mapping values ​​of all sampling points in the interval by more than three standard deviations;

[0012] The criteria for judging whether the sampling interval is the third emotional state are as follows: if the γ wave has m3 or more second peak points in the sampling interval, the sampling interval is judged to be the third emotional state; the second peak point is the sampling point where the γ wave mapping value is greater than the mean of the γ wave mapping values ​​of all sampling points in the interval by more than three standard deviations;

[0013] The criterion for judging whether the sampling interval is the first emotional state is: if the α wave, β wave, and γ wave are relatively stable in the sampling interval, and all have less than m4 peak points, and m4≤min(m2, m3), then the sampling interval is judged to be the first emotional state;

[0014] n, m1, m2, m3, and m4 are all natural numbers, and the proportional relationship between m2 and m1, and the proportional relationship between m3 and m1 are determined through experiments.

[0015] According to the above aspects and any possible implementation, an implementation is further provided, in step S2, the sampling time interval is 1 second, the total number of sampling points is 360, every 10 sampling points is a sampling interval, and there are 36 sampling intervals in total;

[0016] The criterion for judging whether the sampling interval is the second emotional state is: if the beta wave has 3 or more peak points in the sampling interval, the sampling interval is judged to be the second emotional state;

[0017] The criterion for judging whether the sampling interval is the third emotional state is: if the gamma wave has 4 or more peak points in the sampling interval, the sampling interval is judged to be the third emotional state;

[0018] The criterion for judging whether the sampling interval is the first emotional state is: if the α wave, β wave, and γ wave are relatively stable within the sampling interval and have less than 3 peak points, then the sampling interval is judged to be the first emotional state.

[0019] According to the aspects described above and any possible implementation method, an implementation method is further provided. In step S1, the original brain waves of the tester are collected using a brain wave sensor, and the original brain waves are locally amplified and represented by brain wave mapping values ​​to avoid the phenomenon that the original brain wave factor value is too small to cause the characteristics to be unclear.

[0020] According to the aspects described above and any possible implementation method, an implementation method is further provided. In step S3, a box plot rule is used to automatically remove abnormal mapping values ​​in the first emotional state data, and a sliding window local extreme value filtering method is used to automatically remove abnormal mapping values ​​in the second emotional state and the third emotional state data.

[0021] According to the above aspects and any possible implementation, an implementation is further provided, and a specific method for removing abnormal mapping values ​​is as follows:

[0022] Arrange all the EEG mapping values ​​of the first emotional state from small to large to obtain the first quartile and the third quartile of each wave of α wave, β wave and γ wave. Subtract the first quartile from the third quartile to obtain the IQR (Interquartile Range). Take the first quartile minus 1.5 times the IQR as the lower bound, take the third quartile plus 1.5 times the IQR as the upper bound. The data outside the upper and lower bounds are abnormal EEG waves in the first emotional state data.

[0023] A sliding window is set for the beta wave in the second emotional state, the maximum value in each window is calculated, and three data points near the maximum value are retained, and the remaining data points are outliers of this sliding window;

[0024] A sliding window is set for the gamma wave in the third emotional state, the maximum value in each window is calculated, and three data points near the maximum value are retained. The remaining data points are outliers of this sliding window.

[0025] As described above and any possible implementation method, an implementation method is further provided. In step S2, the locally amplified EEG mapping value is used to draw an EEG, identify α waves, β waves, and γ waves, and determine the emotional state corresponding to the EEG sampling interval.

[0026] On the other hand, the present invention also provides a system for constructing an emotion recognition dataset based on brain wave mapping values, the system being used to implement the above method, the system comprising:

[0027] The brain wave acquisition and processing unit is used to acquire brain wave mapping values ​​and separate the alpha wave, beta wave and gamma wave of the brain wave;

[0028] An emotional state judgment unit, which judges whether the sampling interval is the first emotional state, the second emotional state or the third emotional state according to the values ​​of the alpha wave, the beta wave and the gamma wave in each sampling interval and the set criteria;

[0029] The abnormal data removal unit collects brain wave mapping values ​​according to different emotional states, and removes abnormal data from the collected data of each emotional state.

[0030] According to the aspects and any possible implementations described above, an implementation is further provided, wherein the first emotional state corresponds to a human calm state, the second emotional state corresponds to a human concentration state, and the third emotional state corresponds to a human happiness state.

[0031] On the other hand, the present invention also provides an application of the above-mentioned method for constructing an emotion recognition dataset based on brain wave mapping values, characterized in that the application includes:

[0032] The collected brain wave mapping values ​​are processed using the method to obtain a brain wave emotion recognition data set with abnormal data removed;

[0033] Establishing a machine learning or deep learning model for brain wave emotion recognition, and training the model using the brain wave emotion recognition dataset with abnormal data removed to obtain a trained model;

[0034] Use the trained model to perform emotion recognition on the brain waves to be identified.

[0035] The beneficial effects of the present invention are as follows: the method of the present invention provides a solution for removing outliers in brain waves with different emotions, thereby solving the problems of difficulty in constructing brain wave emotion recognition data and a large number of outliers. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 Shown is a waveform of the entire EEG mapping graph collected from a test subject.

[0037] Figure 2 Shown from Figure 1 The waveform diagram of the first emotional state separated from the waveform diagram of .

[0038] Figure 3 Shown from Figure 1 The waveform diagram of the second emotional state separated from the waveform diagram of .

[0039] Figure 4 Shown from Figure 1 The waveform diagram of the third emotional state separated from the waveform diagram of .

[0040] Figure 5 Shown Figure 2 Waveform of the first emotional state after removing outliers.

[0041] Figure 6 Shown Figure 3 Waveform of the second emotional state after removing outliers (sliding window parameter is 10 and step size is 3).

[0042] Figure 7 Shown Figure 4Waveform of the third emotional state after removing outliers (sliding window parameter is 10 and step size is 3).

[0043] Figure 8 The figure is a flow chart of a method for constructing an emotion recognition dataset based on brain wave mapping values ​​according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The specific embodiments of the present invention will be described in detail below in conjunction with specific drawings. It should be noted that the technical features or combinations of technical features described in the following embodiments should not be considered isolated, and they can be combined with each other to achieve better technical effects.

[0045] Brainwaves are the macroscopic manifestation of the electrical activity of brain neurons. According to different frequency ranges, brainwaves can be divided into several main types:

[0046] Delta waves: Frequency range: 0.5-4 Hz

[0047] Characteristics: It appears in the deep sleep stage and is the slowest brain wave.

[0048] Theta waves: Frequency range: 4-8 Hz

[0049] Characteristics: Commonly found in the brains of children and adolescents, it is associated with memory, creativity, emotional regulation, and states of deep relaxation.

[0050] Alpha waves: Frequency range: 8-13 Hz

[0051] Characteristics: Associated with relaxation, meditation, and a light sleep state, usually occurring when resting with eyes closed.

[0052] Beta waves: Frequency range: 13-30 Hz

[0053] Characteristics: Associated with wakefulness, alertness, logical thinking, and daily cognitive activities.

[0054] Gamma waves: Frequency range: above 30 Hz

[0055] Characteristics: Related to cognitive functions, attention, consciousness and synchronized activity of neural networks.

[0056] Sigma waves: Frequency range: 12-16 Hz

[0057] Characteristics: Associated with the light sleep stage and certain types of sleep disorders.

[0058] Mu waves: Frequency range: 7-13 Hz

[0059] Characteristics: Related to motor control and sensory inhibition, usually seen during fine motor movements.

[0060] Specific waveform changes are associated with specific human emotional states. The present invention focuses on the correlation between alpha waves, beta waves, and gamma waves and human emotional states. Various waves tend to be flat and the phased waveforms with little change are associated with calm emotions. There are many peaks in the beta wave, the number of peaks increases, and the waveforms with rapid fluctuations are associated with focused emotions. There are many peaks in the gamma wave, the number of peaks is the largest, and the waveforms with extremely rapid fluctuations are associated with happy emotions. According to the distribution characteristics of different waves, the present invention splits the data into brain wave data of three emotions.

[0061] As shown in the figure, an embodiment of the present invention provides a method for constructing an emotion recognition dataset based on brain wave mapping values, including:

[0062] S1, collecting brain wave mapping values ​​through multiple sampling points;

[0063] S2, separating the alpha wave, beta wave, and gamma wave of the brain wave according to the collected brain wave mapping values, and judging the sampling interval as the first emotional state, the second emotional state, or the third emotional state according to the values ​​of the alpha wave, beta wave, and gamma wave in each sampling interval according to the set criteria;

[0064] S3. Collect brain wave mapping values ​​according to different emotional states, remove abnormal data from the collected data of each emotional state, and obtain a brain wave emotion recognition data set.

[0065] The specific process is as follows Figure 8 shown.

[0066] In a specific embodiment, in step S2, multiple sampling points are arranged in time sequence, and each m1 point is a sampling interval, and there are n sampling intervals in total;

[0067] The criteria for judging whether the sampling interval is the second emotional state are as follows: if there are m2 or more first peak points of the β wave in the sampling interval, the sampling interval is judged to be the second emotional state; the first peak point is the sampling point where the β wave mapping value is greater than the mean of the β wave mapping values ​​of all sampling points in the interval by more than three standard deviations;

[0068] The criteria for judging whether the sampling interval is the third emotional state are as follows: if the γ wave has m3 or more second peak points in the sampling interval, the sampling interval is judged to be the third emotional state; the second peak point is the sampling point where the γ wave mapping value is greater than the mean of the γ wave mapping values ​​of all sampling points in the interval by more than three standard deviations;

[0069] The criterion for judging whether the sampling interval is the first emotional state is: if the α wave, β wave, and γ wave are relatively stable in the sampling interval, and all have less than m4 peak points, and m4≤min(m2, m3), then the sampling interval is judged to be the first emotional state;

[0070] n, m1, m2, m3, and m4 are all natural numbers, and the proportional relationship between m2 and m1, and the proportional relationship between m3 and m1 are determined through experiments.

[0071] In a specific embodiment, in step S2, the sampling time interval is 1 second, the total number of sampling points is 360, every 10 sampling points is a sampling interval, and there are 36 sampling intervals in total;

[0072] The criterion for judging whether the sampling interval is the second emotional state is: if the beta wave has 3 or more first peak points in the sampling interval, the sampling interval is judged to be the second emotional state;

[0073] The criterion for judging whether the sampling interval is the third emotional state is: if the gamma wave has 4 or more second peak points in the sampling interval, the sampling interval is judged to be the third emotional state;

[0074] The criterion for judging whether the sampling interval is the first emotional state is: if the α wave, β wave, and γ wave are relatively stable within the sampling interval and have less than 3 peak points, then the sampling interval is judged to be the first emotional state.

[0075] In a specific embodiment, the first emotional state is associated with a human calm state, the second emotional state is associated with a human concentration state, and the third emotional state is associated with a human happiness state.

[0076] In a specific embodiment, in step S1, the original brain waves of the tester are collected using a brain wave sensor, and the original brain waves are locally amplified and represented by brain wave mapping values ​​to avoid the phenomenon that the original brain waves have unclear features due to too small values.

[0077] In a specific embodiment, in step S3, a box plot rule is used to automatically remove abnormal mapping values ​​in the first emotional state data, and a sliding window local extreme value filtering method is used to automatically remove abnormal mapping values ​​in the second emotional state and the third emotional state data.

[0078] In a specific embodiment, the specific method for removing abnormal mapping values ​​is:

[0079] Arrange all the EEG mapping values ​​of the first emotional state from small to large to obtain the first quartile and the third quartile of each wave of α wave, β wave and γ wave. Subtract the first quartile from the third quartile to obtain the IQR (Interquartile Range). Take the first quartile minus 1.5 times the IQR as the lower bound, take the third quartile plus 1.5 times the IQR as the upper bound. The data outside the upper and lower bounds are abnormal EEG waves in the first emotional state data.

[0080] Calculate IQR (Interquartile Range):

[0081] IQR=Q3-Q1 (1)

[0082] Among them, Q3 is the third quantile and Q1 is the first quantile.

[0083] Calculate the lower bound N and upper bound M:

[0084] N=Q1-1.5IQR (2)

[0085] M=Q3-1.5IQR (3)

[0086] A sliding window is set for the β wave in the second emotional state, the maximum value in each window is calculated, and two data points on the left and right of the maximum value are retained. The remaining data points are the outliers of this sliding window.

[0087] A sliding window is set for the gamma wave in the third emotional state, the maximum value in each window is calculated, and two data points on the left and right of the maximum value are retained. The remaining data points are the outliers of this sliding window.

[0088] In a specific embodiment, the locally amplified brain wave mapping values ​​are used to draw an electroencephalogram, identify alpha waves, beta waves, and gamma waves, and determine the emotional state corresponding to the electroencephalogram sampling interval.

[0089] An embodiment of the present invention provides a system for constructing an emotion recognition data set based on brain wave mapping values, the system being used to implement the above method, and the system comprising:

[0090] The brain wave acquisition and processing unit is used to acquire brain wave mapping values ​​and separate the alpha wave, beta wave and gamma wave of the brain wave;

[0091] An emotional state judgment unit, which judges whether the sampling interval is the first emotional state, the second emotional state or the third emotional state according to the values ​​of the alpha wave, the beta wave and the gamma wave in each sampling interval and the set criteria;

[0092] The abnormal data removal unit collects brain wave mapping values ​​according to different emotional states, and removes abnormal data from the collected data of each emotional state.

[0093] In a specific embodiment, a brain wave sensor is used to collect brain waves of multiple testers, and a local amplification module is installed in the sensor to directly obtain brain wave mapping values.

[0094] When collecting brain waves, in order to efficiently collect brain waves of various emotions, the testers watched video clips that could stimulate the first emotional state (calm state), the second emotional state (focused state), and the third emotional state (happy state) during the collection process.

[0095] An embodiment of the present invention provides an application of the above-mentioned method for constructing an emotion recognition dataset based on brain wave mapping values, the application comprising:

[0096] The collected brain wave mapping values ​​are processed using the method to obtain a brain wave emotion recognition data set with abnormal data removed;

[0097] Establishing a machine learning or deep learning model for brain wave emotion recognition, and training the model using the brain wave emotion recognition dataset with abnormal data removed to obtain a trained model;

[0098] Use the trained model to perform emotion recognition on the brain waves to be identified.

[0099] Example

[0100] Figure 1 A waveform diagram of the entire EEG mapping value collected for a test subject.

[0101] The data sampling frequency is once per second. The entire data segment consists of 360 sampling points. The entire data segment is split into 10 sampling points as a sampling interval. Figure 2 , Figure 3 , Figure 4 They are the waveform diagrams of the three emotions after separation: the first emotional state (calm), the second emotional state (focus) and the third emotional state (happy).

[0102] The "box plot rule" is used to automatically remove abnormal mapping values ​​in the first emotional state (calm) data. Figure 5 It is the waveform diagram of the first emotional state after removing outliers.

[0103] The “sliding window local extreme value filtering method” is used to automatically remove abnormal mapping values ​​in the second emotional state (focus) and the third emotional state (happiness) data. The sliding window parameter is 10 and the step size is 3. Figure 6 , Figure 7They are the waveform diagrams of the second emotional state and the third emotional state after removing outliers.

[0104] After the above abnormal data removal process, an emotion recognition data set containing 54 data items, 120 sampling points in each data item, and a total of 6480 sampling points is obtained.

[0105] In order to verify the objectivity and effectiveness of the method of the present invention, the applicant conducted multiple groups of comparative tests. In the process of obtaining brain waves, artificial interference (such as adding electrical noise and bioelectric noise) was added at the same time, and the brain waves doped with interference were collected. The brain wave data processed by the method of the present invention obviously removed the abnormal data caused by electrical noise and bioelectric noise. At the same time, the mutual interference between various emotional states was also effectively removed. This shows that the present invention is highly efficient in removing abnormal data.

[0106] The above description shows and describes several preferred embodiments of the present application, but as mentioned above, it should be understood that the present application is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the application concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not deviate from the spirit and scope of the present application, and should be within the scope of protection of the claims attached to the present application.

Claims

1. A method for constructing an emotion recognition dataset based on brain wave mapping values, characterized in that: The method comprises: S1, collecting brain wave mapping values ​​through multiple sampling points; S2, separating the alpha wave, beta wave, and gamma wave of the brain wave according to the collected brain wave mapping values, and judging the sampling interval as the first emotional state, the second emotional state, or the third emotional state according to the values ​​of the alpha wave, beta wave, and gamma wave in each sampling interval according to the set criteria; S3. Collect brain wave mapping values ​​according to different emotional states, remove abnormal data from the collected data of each emotional state, and obtain a brain wave emotion recognition data set.

2. The method for constructing an emotion recognition dataset based on brain wave mapping values ​​according to claim 1, characterized in that: In step S2, multiple sampling points are arranged in time sequence, with each m1 point being a sampling interval, and there are n sampling intervals in total; The criteria for judging whether the sampling interval is the second emotional state are as follows: if there are m2 or more first peak points of the β wave in the sampling interval, the sampling interval is judged to be the second emotional state; the first peak point is the sampling point where the β wave mapping value is greater than the mean of the β wave mapping values ​​of all sampling points in the interval by more than three standard deviations; The criteria for judging whether the sampling interval is the third emotional state are as follows: if the γ wave has m3 or more second peak points in the sampling interval, the sampling interval is judged to be the third emotional state; the second peak point is the sampling point where the γ wave mapping value is greater than the mean of the γ wave mapping values ​​of all sampling points in the interval by more than three standard deviations; The criterion for judging whether the sampling interval is the first emotional state is: if the α wave, β wave, and γ wave are relatively stable in the sampling interval, and all have less than m4 peak points, and m4≤min(m2, m3), then the sampling interval is judged to be the first emotional state; n, m1, m2, m3, and m4 are all natural numbers.

3. The method for constructing an emotion recognition data set based on brain wave mapping values ​​according to claim 2, characterized in that: In step S2, the sampling time interval is 1 second, the total number of sampling points is 360, every 10 sampling points is a sampling interval, and there are 36 sampling intervals in total; The criterion for judging whether the sampling interval is the second emotional state is: if the beta wave has 3 or more first peak points in the sampling interval, the sampling interval is judged to be the second emotional state; The criterion for judging whether the sampling interval is the third emotional state is: if the gamma wave has 4 or more second peak points in the sampling interval, the sampling interval is judged to be the third emotional state; The criterion for judging whether the sampling interval is the first emotional state is: if the α wave, β wave, and γ wave are relatively stable within the sampling interval and have less than 3 peak points, then the sampling interval is judged to be the first emotional state.

4. The method for constructing an emotion recognition data set based on brain wave mapping values ​​according to claim 1, characterized in that: In step S1, the original brain waves of the tester are collected by using a brain wave sensor, and the original brain waves are locally amplified and represented by brain wave mapping values.

5. The method for constructing an emotion recognition data set based on brain wave mapping values ​​according to claim 2 or 3, characterized in that: In step S3, the box plot rule is used to automatically remove abnormal mapping values ​​in the first emotional state data, and the sliding window local extreme value filtering method is used to automatically remove abnormal mapping values ​​in the second emotional state and the third emotional state data.

6. The method for constructing an emotion recognition data set based on brain wave mapping values ​​according to claim 5, characterized in that: The specific method for removing abnormal mapping values ​​is: Arrange all the EEG mapping values ​​of the first emotional state from small to large to obtain the first quartile and the third quartile of each wave of α wave, β wave and γ wave, subtract the first quartile from the third quartile to obtain the IQR, take the first quartile minus 1.5 times the IQR as the lower bound, take the third quartile plus 1.5 times the IQR as the upper bound, and the data outside the upper and lower bounds are abnormal EEG waves in the first emotional state data; A sliding window is set for the β wave in the second emotional state, the maximum value in each window is calculated, and two data points on the left and right of the maximum value are retained. The remaining data points are the outliers of this sliding window. A sliding window is set for the gamma wave in the third emotional state, the maximum value in each window is calculated, and two data points on the left and right of the maximum value are retained. The remaining data points are the outliers of this sliding window.

7. The method for constructing an emotion recognition dataset based on brain wave mapping values ​​according to claim 1, characterized in that: In step S2, the locally amplified EEG mapping value is used to draw an EEG, identify α waves, β waves, and γ waves, and determine the emotional state corresponding to the EEG sampling interval.

8. A system for constructing an emotion recognition dataset based on brain wave mapping values, characterized in that: The system is used to implement the method according to any one of claims 1 to 7, and the system includes: The brain wave acquisition and processing unit is used to acquire brain wave mapping values ​​and separate the alpha wave, beta wave and gamma wave of the brain wave; The emotional state judgment unit judges, according to the values ​​of the alpha wave, beta wave and gamma wave in each sampling interval, that the sampling interval is the first emotional state, the second emotional state or the third emotional state according to the set criteria; The abnormal data removal unit collects brain wave mapping values ​​according to different emotional states, and removes abnormal data from the collected data of each emotional state.

9. The system for constructing an emotion recognition data set based on brain wave mapping values ​​according to claim 8, characterized in that: The first emotional state corresponds to a human calm state, the second emotional state corresponds to a human concentration state, and the third emotional state corresponds to a human happiness state.

10. An application of the method for constructing an emotion recognition dataset based on brain wave mapping values ​​as described in any one of claims 1 to 7, characterized in that: The applications include: The collected brain wave mapping values ​​are processed using the method to obtain a brain wave emotion recognition data set with abnormal data removed; Establishing a machine learning or deep learning model for brain wave emotion recognition, and training the model using the brain wave emotion recognition dataset with abnormal data removed to obtain a trained model; Use the trained model to perform emotion recognition on the brain waves to be identified.

Images