EEG attention signal domain adaptation method and focusing method
By extracting domain influence factors and environmental information from EEG attention signals and dynamically adjusting weights to generate target attention signals, the problem of signal heterogeneity caused by individual differences and environmental factors is solved, and real-time, personalized EEG attention signal standardization and dynamic focusing are achieved.
Patent Information
- Application Number
- CN202510974394.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-15
- Publication Date
- 2025-12-16
AI Technical Summary
Existing EEG attention signal processing methods cannot effectively address signal heterogeneity caused by individual differences and environmental factors, making it difficult to achieve real-time, personalized, and standardized processing.
By acquiring the real-time attention signal of the target user, preprocessing it, extracting the baseline attention signal, and using domain influence factors such as historical mean, variance, alpha wave to beta wave energy ratio, and real-time feedback factor, combined with signal-to-noise ratio and noise information, the weights are dynamically adjusted to generate the target attention signal.
It achieves real-time, automated EEG attention signal standardization processing, adapts to dynamic scenes, and supports dynamic focusing and cross-device applications.
Smart Images
Figure CN121129263A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to an EEG attention signal domain adaptation method and focusing method. Background Technology
[0002] EEG attention signals vary due to individual differences (e.g., scalp thickness, EEG activity intensity), and at the same attention level, the mean attention signal value may differ by 10–20% between different users. Since dynamic focusing systems require standardized EEG attention signal values as input to ensure consistency in refractive error mapping, standardizing EEG attention signals is a pressing issue that needs to be addressed.
[0003] First, most existing EEG attention signals are calibrated offline. During calibration, users need to manually input the baseline value, which is complicated and not suitable for real-time applications.
[0004] Secondly, environmental factors (such as noise) and differences in signals across devices further increase the difficulty of standardization, and existing standardization methods do not take this factor into account.
[0005] In addition, some patents have introduced domain adaptation technology. For example, patent CN119442038A discloses a cross-subject emotion recognition method based on EEG multi-branch graph convolution. It uses domain adaptation technology to solve the problem of EEG signal distribution variation across subjects, but does not consider the differences in individual user characteristics, so it cannot achieve personalized processing of EEG attention signals.
[0006] Therefore, a personalized domain adaptation method needs to be developed to improve the standardization effect of EEG attention signals. Summary of the Invention
[0007] The technical problem to be solved by the present invention is to provide an EEG attention signal domain adaptation method and a focusing method, which can standardize the EEG attention signal.
[0008] To address the aforementioned technical problems, this invention provides an EEG attention signal domain adaptation method, comprising: acquiring a real-time attention signal of a target user; preprocessing the real-time attention signal to extract a baseline attention signal; extracting a domain influence factor, wherein the domain influence factor includes the mean, variance, and maximum variance of historical attention signals; and adjusting the baseline attention signal according to the domain influence factor to generate a target attention signal.
[0009] As an improvement to the above scheme, the step of adjusting the baseline attention signal according to the domain influence factor to generate the target attention signal includes: according to the formula A=a+(M0-M)×W1+(S / S) maxThe baseline attention signal is adjusted by )×W2 to generate the target attention signal; where A is the target attention signal, a is the baseline attention signal, M0 is the target mean, M is the mean of the historical attention signals, S is the variance of the historical attention signals, and S max W1 represents the maximum variance of the historical attention signal, W2 represents the first dynamic weight value, and W3 represents the second dynamic weight value.
[0010] As an improvement to the above scheme, the domain influence factor also includes the energy ratio of α waves to β waves in the benchmark attention signal.
[0011] As an improvement to the above scheme, the step of adjusting the baseline attention signal according to the domain influence factor to generate the target attention signal includes: according to the formula A=a+(M0-M)×W1+(S / S) max )×W2+(RR O The baseline attention signal is adjusted by )×W3 to generate the target attention signal; where A is the target attention signal, a is the baseline attention signal, M0 is the target mean, M is the mean of the historical attention signals, S is the variance of the historical attention signals, and S max R is the maximum variance of the historical attention signal, R is the energy ratio of α wave to β wave in the baseline attention signal, R0 is the target energy ratio of α wave to β wave, W1 is the first dynamic weight value, W2 is the second dynamic weight value, and W3 is the third dynamic weight value.
[0012] As an improvement to the above scheme, the domain influence factor also includes the energy ratio of α waves to β waves in the benchmark attention signal and the real-time feedback factor of the target user.
[0013] As an improvement to the above scheme, the step of adjusting the baseline attention signal according to the domain influence factor to generate the target attention signal includes: according to the formula A=a+(M0-M)×W1+(S / S) max )×W2+(RR O )×W3+(FF O The baseline attention signal is adjusted by )×W4 to generate the target attention signal; where A is the target attention signal, a is the baseline attention signal, M0 is the target mean, M is the mean of the historical attention signals, S is the variance of the historical attention signals, and S max R is the maximum variance of the historical attention signal, R is the energy ratio of α waves to β waves in the baseline attention signal, R0 is the target energy ratio of α waves to β waves, and F is the real-time feedback factor. O The target feedback factor is defined as follows: W1 is the first dynamic weight value, W2 is the second dynamic weight value, W3 is the third dynamic weight value, and W4 is the fourth dynamic weight value.
[0014] As an improvement to the above scheme, the EEG attention signal domain adaptation method further includes: acquiring the signal-to-noise ratio information of the current environment, and adjusting the first dynamic weight value, the second dynamic weight value and the third dynamic weight value according to the signal-to-noise ratio information; and / or acquiring the noise information of the current environment, and adjusting the fourth dynamic weight value according to the noise information.
[0015] As an improvement to the above scheme, the step of adjusting the first dynamic weight value, the second dynamic weight value, and the third dynamic weight value according to the signal-to-noise ratio information includes: increasing the first dynamic weight value and the second dynamic weight value when the signal-to-noise ratio information is greater than a preset signal-to-noise ratio, otherwise increasing the third dynamic weight value; and / or adjusting the fourth dynamic weight value according to the noise information includes: increasing the fourth dynamic weight value when the noise information is greater than a preset noise level.
[0016] As an improvement to the above scheme, the step of preprocessing the real-time attention signal to extract the reference attention signal includes: performing artifact detection processing on the real-time attention signal to generate an initial attention signal; and extracting the reference attention signal from the initial attention signal.
[0017] Accordingly, the present invention also provides a focusing method based on EEG attention signals, including: the steps of the above-described EEG attention signal domain adaptation method; further including: mapping the domain-adapted target attention signal into a diopter control signal; and adjusting the diopter of the EEG glasses according to the diopter control signal.
[0018] Implementing this invention has the following beneficial effects:
[0019] This invention can introduce a domain influence factor based on individual differences, thereby achieving real-time and automated standardization processing of EEG attention signals;
[0020] Furthermore, the present invention can dynamically adjust the weight of the domain influence based on environmental information such as signal-to-noise ratio and / or noise, and combine user individual characteristics, frequency domain characteristics and user feedback to standardize the EEG attention signal to adapt to dynamic scenarios;
[0021] In addition, this invention applies a personalized domain adaptation method to the focusing method, which can standardize EEG attention signals and support dynamic focusing, attention monitoring and cross-device applications. Attached Figure Description
[0022] Figure 1 This is a flowchart of the first embodiment of the EEG attention signal domain adaptation method of the present invention;
[0023] Figure 2 This is a flowchart of the second embodiment of the EEG attention signal domain adaptation method of the present invention;
[0024] Figure 3 This is a flowchart of the third embodiment of the EEG attention signal domain adaptation method of the present invention;
[0025] Figure 4 This is a flowchart of an embodiment of the focusing method based on EEG attention signals of the present invention. Detailed Implementation
[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings.
[0027] See Figure 1 , Figure 1 A flowchart of a first embodiment of the EEG attention signal domain adaptation method of the present invention is shown, which includes:
[0028] S101, acquire the target user's real-time attention signal;
[0029] S102, preprocess the real-time attention signal to extract the baseline attention signal;
[0030] Accordingly, the steps of preprocessing the real-time attention signal to extract the baseline attention signal include:
[0031] (1) Perform artifact detection processing on the real-time attention signal to generate the initial attention signal;
[0032] In practical applications, the artifact_removal() function can be used to remove artifacts from real-time attention signals, aiming to eliminate various interference signals (such as eye movements, muscle activity, ECG, etc.) and output clean EEG signals.
[0033] (2) Extract the baseline attention signal from the initial attention signal.
[0034] In practical applications, the extract_attention() function can be used to extract the reference attention signal from the initial attention signal.
[0035] S103, Extract the domain impact factor;
[0036] In this embodiment, the domain influence factor includes the mean, variance, and maximum variance of the historical attention signal;
[0037] In practical applications, the user's historical attention signals over the past 7 days can be stored in a buffer, and the mean and variance of the historical attention signals can be calculated using a sliding window.
[0038] Accordingly, the buffer capacity is preferably 50 historical attention signals, but this is not a limitation and can be set according to the actual situation.
[0039] S104, Adjust the baseline attention signal based on the mean, variance and maximum variance of the historical attention signal to generate the target attention signal;
[0040] Accordingly, the baseline attention signal can be adjusted to generate the target attention signal according to the following formula:
[0041] A=a+(M0-M)×W1+(S / S) max )×W2
[0042] in:
[0043] A represents the target attention signal;
[0044] 'a' represents the baseline attention signal;
[0045] M0 is the target mean, which ranges from 40 to 60. In this embodiment, M0 = 50, but it is not a limitation and can be set according to the actual situation.
[0046] M is the mean of historical attention signals;
[0047] S is the variance of the historical attention signal;
[0048] S max The maximum variance of historical attention signals;
[0049] W1 is the first dynamic weight value;
[0050] W2 is the second dynamic weight value.
[0051] Furthermore, this embodiment can also obtain the signal-to-noise ratio information of the current environment, and adjust the first dynamic weight value and the second dynamic weight value according to the signal-to-noise ratio information.
[0052] Specifically, when the signal-to-noise ratio (SNR) is greater than a preset SNR, the first dynamic weight value and the second dynamic weight value are increased. Preferably, the adjustment range (e.g., increment) of the first dynamic weight value and the second dynamic weight value is [0.05, 0.15], but this is not a limitation and can be set according to the actual situation.
[0053] For example, with a preset signal-to-noise ratio of 10, the initial value of the first dynamic weight is 0.1, and the adjusted value of the first dynamic weight is 0.12; the initial value of the second dynamic weight is 0.05, and the adjusted value of the second dynamic weight is 0.07; when the signal-to-noise ratio is 5 (5 < 10), the first dynamic weight is the initial value (i.e., 0.1), and the second dynamic weight is the initial value (i.e., 0.05); when the signal-to-noise ratio is 11 (11 > 10), the first dynamic weight is the adjusted value (i.e., 0.12), and the second dynamic weight is the adjusted value (i.e., 0.07).
[0054] Therefore, by monitoring the signal-to-noise ratio information in real time, the present invention can realize the dynamic adjustment of the first dynamic weight value and the second dynamic weight value, thereby further optimizing the target attention signal and providing high flexibility.
[0055] See Figure 2 , Figure 2 A flowchart of a second embodiment of the EEG attention signal domain adaptation method of the present invention is shown, which includes:
[0056] S201, acquire the target user's real-time attention signal;
[0057] S202, preprocess the real-time attention signal to extract the baseline attention signal;
[0058] S203, Extract the domain impact factor;
[0059] and Figure 1 Unlike the first embodiment shown, in this embodiment, the domain influence factor also includes the energy ratio of α waves to β waves in the baseline attention signal.
[0060] It should be noted that alpha waves have a frequency of 8–12 Hz and can reflect a relaxed state, which is used for mental health assessment (e.g., anxiety scores); while beta waves have a frequency of 12–30 Hz and can reflect a focused state, which is used for attention monitoring and games (e.g., reaction speed optimization).
[0061] In practical applications, the energy ratio of alpha waves to beta waves in the attention signal can be referenced using the compute_alpha_beta_ratio() function.
[0062] S204, adjust the baseline attention signal to generate the target attention signal based on the mean, variance, maximum variance and energy ratio of alpha and beta waves of the historical attention signal;
[0063] Accordingly, the baseline attention signal can be adjusted to generate the target attention signal according to the following formula:
[0064] A=a+(M0-M)×W1+(S / S) max )×W2+(RRO )×W3
[0065] in:
[0066] A represents the target attention signal;
[0067] 'a' represents the baseline attention signal;
[0068] M0 is the target mean;
[0069] M is the mean of historical attention signals;
[0070] S is the variance of the historical attention signal;
[0071] S max The maximum variance of historical attention signals;
[0072] R is the energy ratio of α waves to β waves in the baseline attention signal;
[0073] R0 is the target energy ratio of α wave to β wave. In this embodiment, R0 = 1.0, but it is not a limitation and can be set according to the actual situation.
[0074] W1 is the first dynamic weight value;
[0075] W2 is the second dynamic weight value;
[0076] W3 is the third dynamic weight value.
[0077] Furthermore, this embodiment can also acquire the signal-to-noise ratio (SNR) information of the current environment, and adjust the first dynamic weight value, the second dynamic weight value, and the third dynamic weight value based on the SNR information. Specifically:
[0078] (1) When the signal-to-noise ratio is greater than the preset signal-to-noise ratio, increase the first dynamic weight value and the second dynamic weight value;
[0079] (2) When the signal-to-noise ratio information is less than or equal to the preset signal-to-noise ratio, increase the third dynamic weight value;
[0080] Preferably, the adjustment range (e.g., increment) of the first dynamic weight value, the second dynamic weight value, and the third dynamic weight value is [0.05, 0.15], but this is not a limitation and can be set according to the actual situation.
[0081] For example, with a preset signal-to-noise ratio (SNR) of 10, the initial value of the first dynamic weight is 0.1, and the adjusted value of the first dynamic weight is 0.12; the initial value of the second dynamic weight is 0.05, and the adjusted value of the second dynamic weight is 0.07; the initial value of the third dynamic weight is 0.02, and the adjusted value of the third dynamic weight is 0.04; when the SNR is 5 (5 < 10), the first dynamic weight is the initial value (i.e., 0.1), the second dynamic weight is the initial value (i.e., 0.05), and the third dynamic weight is the adjusted value (i.e., 0.04); when the SNR is 11 (11 > 10), the first dynamic weight is the adjusted value (i.e., 0.12), the second dynamic weight is the adjusted value (i.e., 0.07), and the third dynamic weight is the initial value (i.e., 0.02).
[0082] Therefore, this invention can handle individual differences and environmental interference, and achieve real-time, automated EEG attention signal standardization processing.
[0083] See Figure 3 , Figure 3 The flowchart of a third embodiment of the EEG attention signal domain adaptation method of the present invention is shown, which includes:
[0084] S301, acquires the target user's real-time attention signal;
[0085] S302, preprocesses the real-time attention signal to extract the baseline attention signal;
[0086] S303, Extract the domain impact factor;
[0087] and Figure 2 The difference between the second embodiment shown is that in this embodiment, the domain influence factor also includes the real-time feedback factor of the target user.
[0088] Correspondingly, the feedback factor is an attention rating of 1 to 5, with 3 being a neutral rating. Target users can upload real-time feedback factors based on their experience.
[0089] S304, adjusts the baseline attention signal to generate the target attention signal based on the mean, variance, maximum variance, energy ratio of alpha and beta waves, and real-time feedback factor of the historical attention signal;
[0090] Accordingly, the baseline attention signal can be adjusted to generate the target attention signal according to the following formula:
[0091] A=a+(M0-M)×W1+(S / S) max )×W2+(RR O )×W3+(FF O )×W4
[0092] in:
[0093] A represents the target attention signal;
[0094] 'a' represents the baseline attention signal;
[0095] M0 is the target mean;
[0096] M is the mean of historical attention signals;
[0097] S is the variance of the historical attention signal;
[0098] S max The maximum variance of historical attention signals;
[0099] R is the energy ratio of α waves to β waves in the baseline attention signal;
[0100] R0 is the target energy ratio of α wave to β wave;
[0101] F is the real-time feedback factor;
[0102] F O The target feedback factor is F0 = 3 in this embodiment, but it is not a limitation and can be set according to the actual situation.
[0103] W1 is the first dynamic weight value;
[0104] W2 is the second dynamic weight value;
[0105] W3 is the third dynamic weight value;
[0106] W4 is the fourth dynamic weight value.
[0107] Furthermore, this embodiment can also acquire the signal-to-noise ratio (SNR) information and noise information of the current environment, adjust the first dynamic weight value, the second dynamic weight value, and the third dynamic weight value based on the SNR information, and adjust the fourth dynamic weight value based on the noise information. Specifically:
[0108] (1) When the signal-to-noise ratio is greater than the preset signal-to-noise ratio, increase the first dynamic weight value and the second dynamic weight value;
[0109] (2) When the signal-to-noise ratio information is less than or equal to the preset signal-to-noise ratio, increase the third dynamic weight value;
[0110] (3) When the noise information is greater than the preset noise, increase the fourth dynamic weight value.
[0111] Preferably, the adjustment range (e.g., increment) of the first dynamic weight value, the second dynamic weight value, the third dynamic weight value and the fourth dynamic weight value is [0.05, 0.15], but this is not a limitation and can be set according to the actual situation.
[0112] For example, the preset signal-to-noise ratio is 10, the preset noise level is 60dB, the initial value of the first dynamic weight is 0.1, and the adjusted value of the first dynamic weight is 0.12; the initial value of the second dynamic weight is 0.05, and the adjusted value of the second dynamic weight is 0.07; the initial value of the third dynamic weight is 0.02, and the adjusted value of the third dynamic weight is 0.04; the initial value of the fourth dynamic weight is 0.01, and the adjusted value of the fourth dynamic weight is 0.03. Wherein:
[0113] When the signal-to-noise ratio is 5 (5 < 10) and the noise is 55dB (55dB < 60dB), the first dynamic weight is the initial value (i.e., 0.1), the second dynamic weight is the initial value (i.e., 0.05), the third dynamic weight is the adjusted value (i.e., 0.04), and the fourth dynamic weight is the initial value (i.e., 0.01).
[0114] When the signal-to-noise ratio is 11 (11 > 10) and the noise level is 55dB (55dB < 60dB), the first dynamic weight is the adjusted value (i.e., 0.12), the second dynamic weight is the adjusted value (i.e., 0.07), the third dynamic weight is the initial value (i.e., 0.02), and the fourth dynamic weight is the initial value (i.e., 0.01).
[0115] When the signal-to-noise ratio is 5 (5 < 10) and the noise level is 75 dB (75 dB > 60 dB), the first dynamic weight is the initial value (i.e., 0.1), the second dynamic weight is the initial value (i.e., 0.05), the third dynamic weight is the adjusted value (i.e., 0.04), and the fourth dynamic weight is the adjusted value (i.e., 0.03).
[0116] When the signal-to-noise ratio is 11 (11 > 10) and the noise level is 75dB (75dB > 60dB), the first dynamic weight is adjusted (i.e., 0.12), the second dynamic weight is adjusted (i.e., 0.07), the third dynamic weight is the initial value (i.e., 0.02), and the fourth dynamic weight is adjusted (i.e., 0.03).
[0117] Therefore, this invention can dynamically adjust the weights based on environmental information such as signal-to-noise ratio and noise, and combine user individual characteristics, frequency domain characteristics and user feedback to standardize the EEG attention signal, adapting to dynamic scenarios and distributed BCI systems.
[0118] See Figure 4 , Figure 4 The flowchart of an embodiment of the focusing method based on EEG attention signals of the present invention is shown, which includes:
[0119] S401, acquires the target user's real-time attention signal;
[0120] S402, preprocesses the real-time attention signal to extract the baseline attention signal;
[0121] S403, Extract the domain impact factor;
[0122] S404, adjust the baseline attention signal according to the domain influence factor to generate the target attention signal;
[0123] S405 maps the domain-adapted target attention signal into a diopter control signal;
[0124] S406 adjusts the diopter of the EEG glasses according to the diopter control signal.
[0125] Therefore, applying personalized domain adaptation methods to focusing methods can standardize EEG attention signals, supporting dynamic focusing, attention monitoring, and cross-device applications.
[0126] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. A method for adapting the EEG attention signal domain, characterized in that, include: Acquire the target user's real-time attention signal; The real-time attention signal is preprocessed to extract the baseline attention signal; Extract the domain influence factor, which includes the mean, variance, and maximum variance of the historical attention signal; The baseline attention signal is adjusted based on the domain influence factor to generate the target attention signal.
2. The EEG attention signal domain adaptation method as described in claim 1, characterized in that, The step of adjusting the baseline attention signal according to the domain influence factor to generate the target attention signal includes: According to the formula A=a+(M0-M)×W1+(S / S) max The baseline attention signal is adjusted by ×W2 to generate the target attention signal; Where A is the target attention signal, a is the baseline attention signal, M0 is the target mean, M is the mean of the historical attention signals, and S is the variance of the historical attention signals. max W1 represents the maximum variance of the historical attention signal, W2 represents the first dynamic weight value, and W3 represents the second dynamic weight value.
3. The EEG attention signal domain adaptation method as described in claim 1, characterized in that, The domain influence factor also includes the energy ratio of α waves to β waves in the baseline attention signal.
4. The EEG attention signal domain adaptation method as described in claim 3, characterized in that, The step of adjusting the baseline attention signal according to the domain influence factor to generate the target attention signal includes: According to the formula A=a+(M0-M)×W1+(S / S) max )×W2+(RR O )×W3 adjusts the baseline attention signal to generate the target attention signal; Where A is the target attention signal, a is the baseline attention signal, M0 is the target mean, M is the mean of the historical attention signals, and S is the variance of the historical attention signals. max R is the maximum variance of the historical attention signal, R is the energy ratio of α wave to β wave in the baseline attention signal, R0 is the target energy ratio of α wave to β wave, W1 is the first dynamic weight value, W2 is the second dynamic weight value, and W3 is the third dynamic weight value.
5. The EEG attention signal domain adaptation method as described in claim 1, characterized in that, The domain influence factor also includes the energy ratio of α waves to β waves in the benchmark attention signal and the real-time feedback factor of the target user.
6. The EEG attention signal domain adaptation method as described in claim 5, characterized in that, The step of adjusting the baseline attention signal according to the domain influence factor to generate the target attention signal includes: According to the formula A=a+(M0-M)×W1+(S / S) max )×W2+(RR O )×W3+(FF O The reference attention signal is adjusted by )×W4 to generate the target attention signal; Where A is the target attention signal, a is the baseline attention signal, M0 is the target mean, M is the mean of the historical attention signals, and S is the variance of the historical attention signals. max R is the maximum variance of the historical attention signal, R is the energy ratio of α waves to β waves in the baseline attention signal, R0 is the target energy ratio of α waves to β waves, and F is the real-time feedback factor. O The target feedback factor is defined as follows: W1 is the first dynamic weight value, W2 is the second dynamic weight value, W3 is the third dynamic weight value, and W4 is the fourth dynamic weight value.
7. The EEG attention signal domain adaptation method as described in claim 6, characterized in that, Also includes: Obtain the signal-to-noise ratio (SNR) information of the current environment, and adjust the first dynamic weight value, the second dynamic weight value, and the third dynamic weight value based on the SNR information; and / or Obtain the noise information of the current environment, and adjust the fourth dynamic weight value according to the noise information.
8. The EEG attention signal domain adaptation method as described in claim 7, characterized in that, The step of adjusting the first dynamic weight value, the second dynamic weight value, and the third dynamic weight value according to the signal-to-noise ratio information includes: increasing the first dynamic weight value and the second dynamic weight value when the signal-to-noise ratio information is greater than a preset signal-to-noise ratio; otherwise, increasing the third dynamic weight value; and / or The step of adjusting the fourth dynamic weight value based on the noise information includes: increasing the fourth dynamic weight value when the noise information is greater than a preset noise level.
9. The EEG attention signal domain adaptation method as described in claim 1, characterized in that, The step of preprocessing the real-time attention signal to extract the baseline attention signal includes: The real-time attention signal is subjected to artifact detection processing to generate an initial attention signal; Extract the reference attention signal from the initial attention signal.
10. A focusing method based on EEG attention signals, characterized in that, Includes the steps of the EEG attention signal domain adaptation method according to any one of claims 1 to 9; Also includes: The domain-adapted target attention signal is mapped into a diopter control signal; The diopter of the EEG glasses is adjusted according to the diopter control signal.
Citation Information
Patent Citations
Cross-subject emotion recognition method, device and system based on EEG multi-branch graph convolution and storage medium
CN119442038A