Attention state detection method and system based on electroencephalogram signals

Abstract

The invention discloses an attention state detection method and system based on electroencephalogram signals. The method comprises the steps that original electroencephalogram signals of M channels of each brain function area in K brain function areas in the watching process of a to-be-detected person in a non-interference detection mode and an interference detection mode are collected; performing primary processing on the original electroencephalogram signals of each channel of each brain function area in the two modes; selecting an optimal electroencephalogram signal of each brain function region in the non-interference detection mode, and selecting an optimal electroencephalogram signal of each brain function region in the interference detection mode; performing feature extraction on the optimal electroencephalogram signals of the K brain function areas in the non-interference detection mode to construct an electroencephalogram attention feature set in the non-interference detection mode; performing feature extraction on the optimal electroencephalogram signals of the K brain function areas in the interference detection mode to construct an electroencephalogram attention feature set in the interference detection mode; and inputting the electroencephalogram attention feature sets in the non-interference detection mode and the interference detection mode into an attention state detection model for detection, and outputting an attention state detection result of the to-be-detected person.

Description

Technical Field

[0001] The present invention relates to the technical field of electroencephalogram (EEG) attention state detection, and in particular to an EEG signal-based attention state detection method and system. Background Art

[0002] The inventor of the present invention has previously applied for an invention patent with application number CN202110184656.3 and the name of the attention state monitoring device and method based on brain-computer interface equipment technology. In this patent, the user continuously collects the user's EEG brain wave signal when completing the attention monitoring task, and extracts the time-frequency domain features after simple processing of the EEG brain wave signal. The extracted time-frequency domain features are input into the deep learning algorithm for analysis to obtain the attention evaluation results. This patented technology is relatively rough and not sophisticated enough to obtain attention evaluation results with high accuracy. On this basis, the present invention further optimizes and improves this attention state monitoring technology. Through refined conception and design, it can obtain the optimal EEG signal, and perform feature extraction based on the optimal EEG signal, so as to obtain attention evaluation results with high accuracy. Summary of the invention

[0003] In view of the problems and shortcomings of the prior art, the present invention provides an attention state detection method and system based on EEG signals.

[0004] The present invention solves the above technical problems through the following technical solutions:

[0005] The present invention provides an attention state detection method based on electroencephalogram signals, which is characterized in that it comprises the following steps:

[0006] S1. Obtain basic information containing the age and identity of the person to be detected, and play the video in the non-interference detection mode and the video in the interference detection mode that match the age of the person to be detected in sequence for the person to be detected to watch. The video in the non-interference detection mode contains N sample videos without interference, and the video in the interference detection mode contains N sample videos with interference. The playback time of each sample video in the non-interference detection mode is the same and the required attention level is different. The playback time of each sample video in the interference detection mode is the same and the required attention level is different. The attention level required for the nth sample video in the non-interference detection mode is the same as the attention level required for the nth sample video in the interference detection mode. The original EEG signals of M channels in each of K brain functional areas during the viewing process of the person to be detected in the non-interference detection mode and the interference detection mode are respectively collected, where N is a positive integer and N≥2, 1≤n≤N, K is a positive integer and K≥2, and M is a positive integer and M≥2;

[0007] S2, respectively, performing preliminary processing of removing noise and artifacts on the original EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode, and obtaining preliminary EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode;

[0008] S3, for the EEG signals of M channels of the kth brain functional area in the non-interference detection mode, select an optimal channel EEG signal from the M channels as the optimal EEG signal of the kth brain functional area, and based on this, select the optimal EEG signal of each brain functional area in the non-interference detection mode;

[0009] For the EEG signals of M channels of the kth brain functional area in the interference detection mode, an EEG signal of an optimal channel is selected from the M channels as the optimal EEG signal of the kth brain functional area, and based on this, the optimal EEG signals of each brain functional area in the interference detection mode are selected;

[0010] S4, extracting features from the optimal EEG signals of K brain functional areas in the non-interference detection mode, obtaining the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area, and the connection features between any two brain functional areas, and constructing the EEG attention feature set in the non-interference detection mode from the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area arranged in sequence first and the connection features between two brain functional areas arranged in sequence later;

[0011] Extract features of the optimal EEG signals of K brain functional areas in the interference detection mode, obtain the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area, and the connection features between any two brain functional areas, and construct the EEG attention feature set in the interference detection mode by the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area arranged in sequence first and the connection features between two brain functional areas arranged in sequence later;

[0012] S5. Input the EEG attention feature set in the undisturbed detection mode and the EEG attention feature set in the interfered detection mode of the person to be detected into the trained attention state detection model constructed based on the convolutional neural network model for detection, so as to output the attention state detection result of the person to be detected. The attention state detection result is divided into 0 to 10 levels, representing the degree of attention state of 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100% respectively, among which 0% is unconscious and 100% is over-concentration.

[0013] The present invention also provides an attention state detection system based on EEG signals, which is characterized in that it includes an EEG acquisition module, an EEG processing module, an optimal EEG screening module, an EEG feature extraction module and an attention state detection module;

[0014] The EEG acquisition module is used to obtain basic information containing the age and identity of the person to be detected, and sequentially play videos in the non-interference detection mode and the interference detection mode that match the age of the person to be detected for the person to be detected to watch, the video in the non-interference detection mode contains N sample videos without interference, and the video in the interference detection mode contains N sample videos with interference. The playback time of each sample video in the non-interference detection mode is the same and the required attention level is different. The playback time of each sample video in the interference detection mode is the same and the required attention level is different. The attention level required for the nth sample video in the non-interference detection mode is the same as the attention level required for the nth sample video in the interference detection mode. The original EEG signals of M channels of each brain functional area in K brain functional areas during the viewing process of the person to be detected in the non-interference detection mode and the interference detection mode are respectively collected, where N is a positive integer and N≥2, 1≤n≤N, K is a positive integer and K≥2, and M is a positive integer and M≥2;

[0015] The EEG processing module is used to perform preliminary processing of noise removal, filtering and artifact removal on the original EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode, respectively, to obtain preliminary EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode, respectively;

[0016] The optimal EEG screening module is used for selecting an optimal channel EEG signal from the M channels of the kth brain functional area in the non-interference detection mode as the optimal EEG signal of the kth brain functional area, based on which the optimal EEG signal of each brain functional area in the non-interference detection mode is selected; for the EEG signal of the M channels of the kth brain functional area in the interference detection mode, selecting an optimal channel EEG signal from the M channels as the optimal EEG signal of the kth brain functional area, based on which the optimal EEG signal of each brain functional area in the interference detection mode is selected;

[0017] The EEG feature extraction module is used to extract features from the optimal EEG signals of K brain function areas in the non-interference detection mode, obtain the attention time domain features and attention frequency domain features corresponding to each sample video of each brain function area, and the connection features between any two brain function areas, and construct the EEG attention feature set in the non-interference detection mode from the attention time domain features and attention frequency domain features corresponding to each sample video of each brain function area arranged in sequence first and the connection features between two brain function areas arranged in sequence later; extract features from the optimal EEG signals of K brain function areas in the interference detection mode, obtain the attention time domain features and attention frequency domain features corresponding to each sample video of each brain function area, and the connection features between any two brain function areas, and construct the EEG attention feature set in the interference detection mode from the attention time domain features and attention frequency domain features corresponding to each sample video of each brain function area arranged in sequence first and the connection features between two brain function areas arranged in sequence later;

[0018] The attention state detection module is used to input the EEG attention feature set in the undisturbed detection mode and the EEG attention feature set in the interfered detection mode of the person to be detected into the trained attention state detection model constructed based on the convolutional neural network model for detection, so as to output the attention state detection result of the person to be detected. The attention state detection result is divided into 0 to 10 levels, respectively representing the degree of attention state 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100%, among which 0% is unconscious and 100% is over-concentration.

[0019] The positive and progressive effects of the present invention are:

[0020] The electroencephalogram attention state detection technology designed by the present invention collects multi-brain functional area multi-channel electroencephalogram signals corresponding to two different detection modes (non-interference detection mode and interference detection mode), analyzes the multi-brain functional area multi-channel electroencephalogram signals in the non-interference detection mode, and obtains the electroencephalogram signals of the optimal channels of each brain functional area in the non-interference detection mode as the optimal electroencephalogram signals, analyzes the multi-brain functional area multi-channel electroencephalogram signals in the interference detection mode, and obtains the electroencephalogram signals of the optimal channels of each brain functional area in the interference detection mode as the optimal electroencephalogram signals, extracts features for the optimal electroencephalogram signals of each brain functional area in the non-interference detection mode, extracts features for the optimal electroencephalogram signals of each brain functional area in the interference detection mode, and obtains the optimal electroencephalogram features in the interference detection mode, and uses the optimal electroencephalogram features in the non-interference detection mode and the optimal electroencephalogram features in the interference detection mode for model learning to obtain the accurate attention state detection results of the person to be detected. The present invention can obtain the optimal electroencephalogram signal through refined conception and design, and extracts features based on the optimal electroencephalogram signal, so as to obtain the attention evaluation result with high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flow chart of a method for detecting attention status based on EEG signals according to a preferred embodiment of the present invention.

[0022] Figure 2 This is a structural block diagram of an attention state detection system based on EEG signals according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0024] like Figure 1 As shown, this embodiment provides an attention state detection method based on EEG signals, which includes the following steps:

[0025] Step 101, obtaining basic information including the age and identity of the person to be detected, and playing the video in the non-interference detection mode and the video in the interference detection mode that match the age of the person to be detected in sequence for the person to be detected to watch, the video in the non-interference detection mode contains N sample videos without interference, and the video in the interference detection mode contains N sample videos with interference. The playback time of each sample video in the non-interference detection mode is the same and the required attention level is different. The playback time of each sample video in the interference detection mode is the same and the required attention level is different. The attention level required for the nth sample video in the non-interference detection mode is the same as the attention level required for the nth sample video in the interference detection mode. The original EEG signals of M channels of each brain functional area in K brain functional areas during the viewing process of the person to be detected in the non-interference detection mode and the interference detection mode are respectively collected, and the K brain functional areas correspond to each other, N is a positive integer, an even number and N≥2, 1≤n≤N, K is a positive integer and K≥2, and M is a positive integer and M≥2.

[0026] In this step, N sample videos in the non-interference detection mode that match the age of the person to be detected are played first, and then N sample videos in the interference detection mode that match the age of the person to be detected are played.

[0027] Specifically, the original EEG signals of M channels of each of the eight brain functional areas in the viewing process of the person to be tested in the non-interference detection mode are collected. The eight brain functional areas are respectively the left prefrontal lobe Fp1, the right prefrontal lobe Fp2, the left central C3, the right central C4, the left parietal lobe P3, the right parietal lobe P4, the left occipital lobe O1 and the right occipital lobe O2. The left prefrontal lobe Fp1 corresponds to the right prefrontal lobe Fp2, the left central C3 corresponds to the right central C4, the left parietal lobe P3 corresponds to the right parietal lobe P4, and the left occipital lobe O1 corresponds to the right occipital lobe O2.

[0028] The original EEG signals of M channels of each of the eight brain functional areas were collected during the viewing of the test personnel in the interference detection mode. The eight brain functional areas were the left prefrontal lobe Fp1, right prefrontal lobe Fp2, left central C3, right central C4, left parietal lobe P3, right parietal lobe P4, left occipital lobe O1 and right occipital lobe O2.

[0029] Step 102: perform preliminary processing (denoising and filtering to remove artifacts) on the original EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode, respectively, to obtain preliminary EEG signals of each channel of each brain functional area in the non-interference detection mode and preliminary EEG signals of each channel of each brain functional area in the interference detection mode.

[0030] Step 103, for the EEG signals of M channels of the kth brain functional area in the non-interference detection mode, select the EEG signal of the optimal channel from the M channels as the optimal EEG signal of the kth brain functional area; based on this, the optimal EEG signal of each brain functional area in the non-interference detection mode can be selected; for the EEG signals of M channels of the kth brain functional area in the interference detection mode, select the EEG signal of the optimal channel from the M channels as the optimal EEG signal of the kth brain functional area, based on this, the optimal EEG signal of each brain functional area in the interference detection mode can be selected.

[0031] For the EEG signals of M channels of the kth brain functional area in the non-interference detection mode, select an EEG signal of an optimal channel from the M channels as the optimal EEG signal of the kth brain functional area. The specific implementation method is as follows: for any two corresponding brain functional areas Ea and Eb in the K brain functional areas in the non-interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the brain functional areas Ea and Eb, and when the abnormal signals are yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals, and when the abnormal signals are no, the preliminary EEG signals of each channel are used as normal EEG signals; for the normal EEG signals of each channel in the brain functional areas Ea and Eb: use the EEG signal of the i-th channel in the brain functional area Ea and the EEG signal of the j-th channel in the brain functional area Eb to calculate the correlation coefficient EX between the i-th channel and the j-th channel ij , thus obtaining 2M correlation coefficients EX for brain functional areas Ea and Eb ij , the two channels corresponding to the maximum value of the correlation coefficient are selected as the optimal channel of the brain functional area Ea and the optimal channel of the brain functional area Eb respectively, the EEG signal of the optimal channel of the brain functional area Ea is used as the optimal EEG signal of the brain functional area Ea, and the EEG signal of the optimal channel of the brain functional area Eb is used as the optimal EEG signal of the brain functional area Eb. Based on this, the optimal EEG signal of each brain functional area in the non-interference detection mode is selected, 1≤i≤M, 1≤j≤M, and the abnormal signal refers to the signal with abnormal amplitude or the missing signal.

[0032] For example, for the left frontal lobe Fp1 and the right frontal lobe Fp2 in the interference-free detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left frontal lobe Fp1 and the right frontal lobe Fp2. If yes, interpolate the abnormal signals to replace them to obtain normal EEG signals. If no, use the preliminary EEG signals of each channel as normal EEG signals. For the normal EEG signals of each channel in the left frontal lobe Fp1 and the right frontal lobe Fp2: use the EEG signal of the i-th channel in the left frontal lobe Fp1 and the EEG signal of the j-th channel in the right frontal lobe Fp2 to calculate the correlation coefficient Fp-X between the i-th channel and the j-th channel. ij, thereby obtaining 2M correlation coefficients Fp-X for the left frontal lobe Fp1 and the right frontal lobe Fp2 ij , the two channels corresponding to the maximum correlation coefficient are selected as the optimal channel of the left prefrontal lobe Fp1 and the optimal channel of the right prefrontal lobe Fp2 respectively, the EEG signal of the optimal channel of the left prefrontal lobe Fp1 is used as the optimal EEG signal of the left prefrontal lobe Fp1, and the EEG signal of the optimal channel of the right prefrontal lobe Fp2 is used as the optimal EEG signal of the right prefrontal lobe Fp2.

[0033] For the left central C3 and the right central C4 in the non-interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left central C3 and the right central C4. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals. For the normal EEG signals of each channel in the left central C3 and the right central C4: use the EEG signal of the i-th channel in the left central C3 and the EEG signal of the j-th channel in the right central C4 to calculate the correlation coefficient CX between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients CX for the left center C3 and the right center C4 ij , the two channels corresponding to the maximum correlation coefficient are selected as the optimal channel of left central C3 and the optimal channel of right central C4 respectively, the EEG signal of the optimal channel of left central C3 is used as the optimal EEG signal of left central C3, and the EEG signal of the optimal channel of right central C4 is used as the optimal EEG signal of right central C4.

[0034] For the left parietal lobe P3 and the right parietal lobe P4 in the non-interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left parietal lobe P3 and the right parietal lobe P4. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals; For the normal EEG signals of each channel in the left parietal lobe P3 and the right parietal lobe P4: use the EEG signal of the i-th channel in the left parietal lobe P3 and the EEG signal of the j-th channel in the right parietal lobe P4 to calculate the correlation coefficient PX between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients PX for the left parietal lobe P3 and the right parietal lobe P4 ij , the two channels corresponding to the largest correlation coefficients are selected as the optimal channel of the left parietal lobe P3 and the optimal channel of the right parietal lobe P4 respectively, the EEG signal of the optimal channel of the left parietal lobe P3 is used as the optimal EEG signal of the left parietal lobe P3, and the EEG signal of the optimal channel of the right parietal lobe P4 is used as the optimal EEG signal of the right parietal lobe P4.

[0035] For the left occipital lobe O1 and the right occipital lobe O2 in the non-interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left occipital lobe O1 and the right occipital lobe O2. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals; For the normal EEG signals of each channel in the left occipital lobe O1 and the right occipital lobe O2: use the EEG signal of the i-th channel in the left occipital lobe O1 and the EEG signal of the j-th channel in the right occipital lobe O2 to calculate the correlation coefficient OX between the i-th channel and the j-th channel ij , thereby obtaining 2M correlation coefficients OX for the left occipital lobe O1 and the right occipital lobe O2 ij , the two channels corresponding to the maximum correlation coefficient are selected as the optimal channel of the left occipital lobe O1 and the optimal channel of the right occipital lobe O2 respectively, the EEG signal of the optimal channel of the left occipital lobe O1 is used as the optimal EEG signal of the left occipital lobe O1, and the EEG signal of the optimal channel of the right occipital lobe O2 is used as the optimal EEG signal of the right occipital lobe O2.

[0036] For the EEG signals of the M channels of the kth brain functional area in the interference detection mode, select the EEG signal of an optimal channel from the M channels as the optimal EEG signal of the kth brain functional area, and select the optimal EEG signals of each brain functional area in the interference detection mode based on this: for any two corresponding brain functional areas Ea' and Eb' in the K brain functional areas in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the brain functional areas Ea' and Eb', and when it is yes, interpolate and replace the abnormal signals to obtain normal EEG signals, and when it is no, the preliminary EEG signals of each channel are used as normal EEG signals; for the normal EEG signals of each channel in the brain functional areas Ea' and Eb': use the EEG signal of the i-th channel in the brain functional area Ea' and the EEG signal of the j-th channel in the brain functional area Eb' to calculate the correlation coefficient E'-X between the i-th channel and the j-th channel ij , thus obtaining 2M correlation coefficients E'-X for brain functional areas Ea' and Eb' ij , the two channels corresponding to the largest correlation coefficient are selected as the optimal channel of the brain functional area Ea' and the optimal channel of the brain functional area Eb' respectively, the EEG signal of the optimal channel of the brain functional area Ea' is used as the optimal EEG signal of the brain functional area Ea', and the EEG signal of the optimal channel of the brain functional area Eb' is used as the optimal EEG signal of the brain functional area Eb'. Based on this, the optimal EEG signal of each brain functional area in the interference detection mode is selected.

[0037] Specifically, for the left frontal lobe Fp1' and the right frontal lobe Fp2' in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left frontal lobe Fp1' and the right frontal lobe Fp2', and if yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals, and if no, use the preliminary EEG signals of each channel as normal EEG signals; for the normal EEG signals of each channel in the left frontal lobe Fp1' and the right frontal lobe Fp2': use the EEG signal of the i-th channel in the left frontal lobe Fp1' and the EEG signal of the j-th channel in the right frontal lobe Fp2' to calculate the correlation coefficient Fp'-X between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients Fp'-X for the left frontal lobe Fp1' and the right frontal lobe Fp2' ij , the two channels corresponding to the maximum correlation coefficient are selected as the optimal channel of the left prefrontal lobe Fp1' and the optimal channel of the right prefrontal lobe Fp2' respectively, the EEG signal of the optimal channel of the left prefrontal lobe Fp1' is used as the optimal EEG signal of the left prefrontal lobe Fp1', and the EEG signal of the optimal channel of the right prefrontal lobe Fp2' is used as the optimal EEG signal of the right prefrontal lobe Fp2'.

[0038] For the left central C3' and right central C4' in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left central C3' and the right central C4'. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals; For the normal EEG signals of each channel in the left central C3' and the right central C4': use the EEG signal of the i-th channel in the left central C3' and the EEG signal of the j-th channel in the right central C4' to calculate the correlation coefficient C'-X between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients C'-X for the left center C3' and the right center C4' ij , the two channels corresponding to the maximum correlation coefficient are selected as the optimal channel of the left center C3' and the optimal channel of the right center C4' respectively, the EEG signal of the optimal channel of the left center C3' is used as the optimal EEG signal of the left center C3', and the EEG signal of the optimal channel of the right center C4' is used as the optimal EEG signal of the right center C4'.

[0039] For the left parietal lobe P3' and the right parietal lobe P4' in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left parietal lobe P3' and the right parietal lobe P4'. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals; For the normal EEG signals of each channel in the left parietal lobe P3' and the right parietal lobe P4': use the EEG signal of the i-th channel in the left parietal lobe P3' and the EEG signal of the j-th channel in the right parietal lobe P4' to calculate the correlation coefficient P'-X between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients P'-X for the left parietal lobe P3' and the right parietal lobe P4' ij , the two channels corresponding to the largest correlation coefficients are selected as the optimal channel of the left parietal lobe P3' and the optimal channel of the right parietal lobe P4' respectively, the EEG signal of the optimal channel of the left parietal lobe P3' is used as the optimal EEG signal of the left parietal lobe P3', and the EEG signal of the optimal channel of the right parietal lobe P4' is used as the optimal EEG signal of the right parietal lobe P4'.

[0040] For the left occipital lobe O1' and the right occipital lobe O2' in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left occipital lobe O1' and the right occipital lobe O2'. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals; For the normal EEG signals of each channel in the left occipital lobe O1' and the right occipital lobe O2': use the EEG signal of the i-th channel in the left occipital lobe O1' and the EEG signal of the j-th channel in the right occipital lobe O2' to calculate the correlation coefficient O'-X between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients O'-X for the left occipital lobe O1' and the right occipital lobe O2' ij , the two channels corresponding to the largest correlation coefficient are selected as the optimal channel of the left occipital lobe O1' and the optimal channel of the right occipital lobe O2' respectively, the EEG signal of the optimal channel of the left occipital lobe O1' is used as the optimal EEG signal of the left occipital lobe O1', and the EEG signal of the optimal channel of the right occipital lobe O2' is used as the optimal EEG signal of the right occipital lobe O2'.

[0041] Step 104, feature extraction is performed on the optimal EEG signals of the K brain functional areas in the non-interference detection mode, and the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area, as well as the connection features between any two brain functional areas are obtained. The EEG attention feature set in the non-interference detection mode is constructed by the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area arranged in sequence first, and the connection features between two brain functional areas arranged in sequence later; feature extraction is performed on the optimal EEG signals of the K brain functional areas in the interference detection mode, and the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area, as well as the connection features between any two brain functional areas are obtained. The EEG attention feature set in the interference detection mode is constructed by the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area arranged in sequence first, and the connection features between two brain functional areas arranged in sequence later.

[0042] Among them, the time domain characteristics of attention are variance characteristics, and the frequency domain characteristics of attention are the α wave power spectral density in the α band and the β wave power spectral density in the β band.

[0043] For the connection feature L between any two brain functional areas Ea and Eb in the K brain functional areas in the non-interference detection mode E :

[0044]

[0045] Where H refers to the number of sampling time points, l is an imaginary number, φ Ea (h) represents the phase of the optimal EEG signal of the brain functional area Ea at the hth sampling time point, φ Eb (h) represents the phase of the optimal EEG signal of the brain functional area Eb at the hth sampling time point;

[0046] For the connection feature L between any two brain functional areas Ea' and Eb' in the K brain functional areas in the interference detection mode E ':

[0047]

[0048] where φ E a'(h) represents the phase of the optimal EEG signal of the brain functional area Ea' at the hth sampling time point. E b'(h) represents the phase of the optimal EEG signal of the brain functional area Eb' at the hth sampling time point.

[0049] Step 105: input the EEG attention feature set in the undisturbed detection mode and the EEG attention feature set in the interfered detection mode of the person to be detected into the trained attention state detection model constructed based on the convolutional neural network model for detection, so as to output the attention state detection result of the person to be detected. The attention state detection result is divided into 0 to 10 levels, representing the attention state degree of 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100% respectively, where 0% is unconscious and 100% is over-concentration.

[0050] like Figure 2 As shown, this embodiment also provides an attention state detection system based on EEG signals, which includes an EEG acquisition module 1, an EEG processing module 2, an optimal EEG screening module 3, an EEG feature extraction module 4 and an attention state detection module 5.

[0051] The EEG acquisition module 1 is used to obtain basic information containing the age and identity of the person to be detected, and sequentially play videos in the non-interference detection mode and the interference detection mode that match the age of the person to be detected for the person to be detected to watch. The video in the non-interference detection mode contains N sample videos without interference, and the video in the interference detection mode contains N sample videos with interference. The playback time of each sample video in the non-interference detection mode is the same and the required attention level is different. The playback time of each sample video in the interference detection mode is the same and the required attention level is different. The attention level required for the nth sample video in the non-interference detection mode is the same as the attention level required for the nth sample video in the interference detection mode. The original EEG signals of M channels of each brain functional area in K brain functional areas during the viewing process of the person to be detected in the non-interference detection mode and the interference detection mode are respectively collected. The K brain functional areas correspond to each other, N is a positive integer and N≥2, 1≤n≤N, K is a positive integer, an even number and K≥2, and M is a positive integer and M≥2.

[0052] The EEG processing module 2 is used to perform preliminary processing of noise removal, filtering and artifact removal on the original EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode, respectively, to obtain preliminary EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode, respectively.

[0053] The optimal EEG screening module 3 is used for selecting an optimal channel EEG signal from the M channels of the kth brain functional area in the non-interference detection mode as the optimal EEG signal of the kth brain functional area. Specifically, for any two corresponding brain functional areas Ea and Eb among the K brain functional areas in the non-interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the brain functional areas Ea and Eb, and if yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals, and if no, use the preliminary EEG signals of each channel as normal EEG signals; for the normal EEG signals of each channel in the brain functional areas Ea and Eb: use the EEG signal of the i-th channel in the brain functional area Ea and the EEG signal of the j-th channel in the brain functional area Eb to calculate the correlation coefficient EX between the i-th channel and the j-th channel. ij , thus obtaining 2M correlation coefficients EX for brain functional areas Ea and Eb ij , the two channels corresponding to the largest correlation coefficient are selected as the optimal channel of brain functional area Ea and the optimal channel of brain functional area Eb respectively, the EEG signal of the optimal channel of brain functional area Ea is used as the optimal EEG signal of brain functional area Ea, and the EEG signal of the optimal channel of brain functional area Eb is used as the optimal EEG signal of brain functional area Eb. Based on this, the optimal EEG signal of each brain functional area in the non-interference detection mode is selected, 1≤i≤M, 1≤j≤M, and the abnormal signal refers to the signal with abnormal amplitude or the missing signal.

[0054] The optimal EEG screening module 3 is used to select the optimal EEG signal of each brain functional area in the non-interference detection mode based on this; for the EEG signals of M channels of the kth brain functional area in the interference detection mode, an optimal channel EEG signal is selected from the M channels as the optimal EEG signal of the kth brain functional area, and the optimal EEG signal of each brain functional area in the interference detection mode is selected based on this. Specifically, for any two corresponding brain function areas Ea' and Eb' among the K brain function areas in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the brain function areas Ea' and Eb', and if yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals, and if no, use the preliminary EEG signals of each channel as normal EEG signals; for the normal EEG signals of each channel in the brain function areas Ea' and Eb': use the EEG signal of the i-th channel in the brain function area Ea' and the EEG signal of the j-th channel in the brain function area Eb' to calculate the correlation coefficient E'-X between the i-th channel and the j-th channel ij , thus obtaining 2M correlation coefficients E'-X for brain functional areas Ea' and Eb' ij, the two channels corresponding to the largest correlation coefficient are selected as the optimal channel of the brain functional area Ea' and the optimal channel of the brain functional area Eb' respectively, the EEG signal of the optimal channel of the brain functional area Ea' is used as the optimal EEG signal of the brain functional area Ea', and the EEG signal of the optimal channel of the brain functional area Eb' is used as the optimal EEG signal of the brain functional area Eb'. Based on this, the optimal EEG signal of each brain functional area in the interference detection mode is selected.

[0055] The EEG feature extraction module 4 is used to extract features from the optimal EEG signals of K brain functional areas in the non-interference detection mode, obtain the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area, and the connection features between any two brain functional areas, and construct the EEG attention feature set in the non-interference detection mode from the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area arranged in sequence first and the connection features between two brain functional areas arranged in sequence later; extract features from the optimal EEG signals of K brain functional areas in the interference detection mode, obtain the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area, and the connection features between any two brain functional areas, and construct the EEG attention feature set in the interference detection mode from the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area arranged in sequence first and the connection features between two brain functional areas arranged in sequence later.

[0056] The attention state detection module 5 is used to input the EEG attention feature set in the undisturbed detection mode and the EEG attention feature set in the interfered detection mode of the person to be detected into the trained attention state detection model constructed based on the convolutional neural network model for detection, so as to output the attention state detection result of the person to be detected. The attention state detection result is divided into 0 to 10 levels, representing the attention state degree of 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100% respectively, among which 0% is unconscious and 100% is over-concentration.

[0057] The attention state detection system based on EEG signals provided in this embodiment is consistent with the working principle and beneficial effects of the above-mentioned attention state detection method based on EEG signals, so it will not be described in detail here. For specific contents, please refer to the introduction of the above-mentioned method embodiment.

[0058] This embodiment also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the aforementioned attention state detection method when executing the computer program.

[0059] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed, the attention state detection method as described above is implemented.

[0060] Although the specific embodiments of the present invention are described above, those skilled in the art should understand that these are only examples, and the protection scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.

Claims

1. A method for detecting attention state based on EEG signals, characterized in that: It includes the following steps: S1. Obtain basic information containing the age and identity of the person to be detected, and play the video in the non-interference detection mode and the video in the interference detection mode that match the age of the person to be detected in sequence for the person to be detected to watch. The video in the non-interference detection mode contains N sample videos without interference, and the video in the interference detection mode contains N sample videos with interference. The playback time of each sample video in the non-interference detection mode is the same and the required attention level is different. The playback time of each sample video in the interference detection mode is the same and the required attention level is different. The attention level required for the nth sample video in the non-interference detection mode is the same as the attention level required for the nth sample video in the interference detection mode. The original EEG signals of M channels in each of K brain functional areas during the viewing process of the person to be detected in the non-interference detection mode and the interference detection mode are respectively collected, where N is a positive integer and N≥2, 1≤n≤N, K is a positive integer and K≥2, and M is a positive integer and M≥2; S2, respectively, performing preliminary processing of removing noise and artifacts on the original EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode, and obtaining preliminary EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode; S3, for the EEG signals of M channels of the kth brain functional area in the non-interference detection mode, select an optimal channel EEG signal from the M channels as the optimal EEG signal of the kth brain functional area, and based on this, select the optimal EEG signal of each brain functional area in the non-interference detection mode; For the EEG signals of M channels of the kth brain functional area in the interference detection mode, an EEG signal of an optimal channel is selected from the M channels as the optimal EEG signal of the kth brain functional area, and based on this, the optimal EEG signals of each brain functional area in the interference detection mode are selected; S4, extracting features from the optimal EEG signals of K brain functional areas in the non-interference detection mode, obtaining the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area, and the connection features between any two brain functional areas, and constructing the EEG attention feature set in the non-interference detection mode from the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area arranged in sequence first and the connection features between two brain functional areas arranged in sequence later; Extract features of the optimal EEG signals of K brain functional areas in the interference detection mode, obtain the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area, and the connection features between any two brain functional areas, and construct the EEG attention feature set in the interference detection mode by the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area arranged in sequence first and the connection features between two brain functional areas arranged in sequence later; S5. Input the EEG attention feature set in the undisturbed detection mode and the EEG attention feature set in the interfered detection mode of the person to be detected into the trained attention state detection model constructed based on the convolutional neural network model for detection, so as to output the attention state detection result of the person to be detected. The attention state detection result is divided into 0 to 10 levels, representing the degree of attention state of 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100% respectively, among which 0% is unconscious and 100% is over-concentration.

2. The method for detecting attention state based on EEG signals as claimed in claim 1, characterized in that: In step S1, the K brain functional areas correspond to each other, and K is an even number; In step S3, for any two corresponding brain function areas Ea and Eb among the K brain function areas in the non-interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the brain function areas Ea and Eb, and if yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals, and if no, use the preliminary EEG signals of each channel as normal EEG signals; For the normal EEG signals of each channel in the brain functional areas Ea and Eb: the correlation coefficient EX between the i-th channel and the j-th channel is calculated using the EEG signal of the i-th channel in the brain functional area Ea and the EEG signal of the j-th channel in the brain functional area Eb. ij , thus obtaining 2M correlation coefficients EX for brain functional areas Ea and Eb ij , the two channels corresponding to the largest correlation coefficient are selected as the optimal channel of brain function area Ea and the optimal channel of brain function area Eb respectively, the EEG signal of the optimal channel of brain function area Ea is used as the optimal EEG signal of brain function area Ea, and the EEG signal of the optimal channel of brain function area Eb is used as the optimal EEG signal of brain function area Eb, based on which the optimal EEG signal of each brain function area in the non-interference detection mode is selected, 1≤i≤M, 1≤j≤M, abnormal signal refers to a signal with abnormal amplitude or a missing signal; For any two corresponding brain function areas Ea' and Eb' among the K brain function areas in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the brain function areas Ea' and Eb', and if yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals, and if no, use the preliminary EEG signals of each channel as normal EEG signals; For the normal EEG signals of each channel in the brain functional areas Ea' and Eb': use the EEG signal of the ith channel in the brain functional area Ea' and the EEG signal of the jth channel in the brain functional area Eb' to calculate the correlation coefficient E'-X between the ith channel and the jth channel ij , thus obtaining 2M correlation coefficients E'-X for brain functional areas Ea' and Eb' ij , the two channels corresponding to the largest correlation coefficient are selected as the optimal channel of the brain functional area Ea' and the optimal channel of the brain functional area Eb' respectively, the EEG signal of the optimal channel of the brain functional area Ea' is used as the optimal EEG signal of the brain functional area Ea', and the EEG signal of the optimal channel of the brain functional area Eb' is used as the optimal EEG signal of the brain functional area Eb'. Based on this, the optimal EEG signal of each brain functional area in the interference detection mode is selected.

3. The method for detecting attention state based on EEG signals as claimed in claim 2, characterized in that: In step S1, the original EEG signals of M channels of each of the eight brain functional areas are collected during the viewing process of the person to be tested in the non-interference detection mode, and the eight brain functional areas are the left frontal lobe Fp1, the right frontal lobe Fp2, the left central C3, the right central C4, the left parietal lobe P3, the right parietal lobe P4, the left occipital lobe O1 and the right occipital lobe O2; In step S3, for the left frontal lobe Fp1 and the right frontal lobe Fp2 in the interference-free detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left frontal lobe Fp1 and the right frontal lobe Fp2, and if yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals, and if no, use the preliminary EEG signals of each channel as normal EEG signals; for the normal EEG signals of each channel in the left frontal lobe Fp1 and the right frontal lobe Fp2: calculate the correlation coefficient Fp-X between the i-th channel and the j-th channel using the EEG signal of the i-th channel in the left frontal lobe Fp1 and the EEG signal of the j-th channel in the right frontal lobe Fp2 ij , thereby obtaining 2M correlation coefficients Fp-X for the left frontal lobe Fp1 and the right frontal lobe Fp2 ij , the two channels with the largest correlation coefficient are selected as the optimal channel of the left prefrontal lobe Fp1 and the optimal channel of the right prefrontal lobe Fp2 respectively, the EEG signal of the optimal channel of the left prefrontal lobe Fp1 is used as the optimal EEG signal of the left prefrontal lobe Fp1, and the EEG signal of the optimal channel of the right prefrontal lobe Fp2 is used as the optimal EEG signal of the right prefrontal lobe Fp2; For the left central C3 and the right central C4 in the non-interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left central C3 and the right central C4. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals. For the normal EEG signals of each channel in the left central C3 and the right central C4: use the EEG signal of the i-th channel in the left central C3 and the EEG signal of the j-th channel in the right central C4 to calculate the correlation coefficient CX between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients CX for the left center C3 and the right center C4 ij , the two channels with the largest correlation coefficient are selected as the optimal channel of left central C3 and the optimal channel of right central C4 respectively, the EEG signal of the optimal channel of left central C3 is used as the optimal EEG signal of left central C3, and the EEG signal of the optimal channel of right central C4 is used as the optimal EEG signal of right central C4; For the left parietal lobe P3 and the right parietal lobe P4 in the non-interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left parietal lobe P3 and the right parietal lobe P4. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals; For the normal EEG signals of each channel in the left parietal lobe P3 and the right parietal lobe P4: use the EEG signal of the i-th channel in the left parietal lobe P3 and the EEG signal of the j-th channel in the right parietal lobe P4 to calculate the correlation coefficient PX between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients PX for the left parietal lobe P3 and the right parietal lobe P4 ij , the two channels with the largest correlation coefficient are selected as the optimal channel of the left parietal lobe P3 and the optimal channel of the right parietal lobe P4 respectively, the EEG signal of the optimal channel of the left parietal lobe P3 is used as the optimal EEG signal of the left parietal lobe P3, and the EEG signal of the optimal channel of the right parietal lobe P4 is used as the optimal EEG signal of the right parietal lobe P4; For the left occipital lobe O1 and the right occipital lobe O2 in the non-interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left occipital lobe O1 and the right occipital lobe O2. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals; For the normal EEG signals of each channel in the left occipital lobe O1 and the right occipital lobe O2: use the EEG signal of the i-th channel in the left occipital lobe O1 and the EEG signal of the j-th channel in the right occipital lobe O2 to calculate the correlation coefficient OX between the i-th channel and the j-th channel ij , thereby obtaining 2M correlation coefficients OX for the left occipital lobe O1 and the right occipital lobe O2 ij , the two channels corresponding to the largest correlation coefficient are selected as the optimal channel of the left occipital lobe O1 and the optimal channel of the right occipital lobe O2 respectively, the EEG signal of the optimal channel of the left occipital lobe O1 is used as the optimal EEG signal of the left occipital lobe O1, and the EEG signal of the optimal channel of the right occipital lobe O2 is used as the optimal EEG signal of the right occipital lobe O2.

4. The method for detecting attention state based on EEG signals as claimed in claim 2, characterized in that: In step S1, the original EEG signals of M channels of each of the eight brain functional areas are collected during the viewing process of the person to be detected in the interference detection mode, and the eight brain functional areas are the left frontal lobe Fp1, the right frontal lobe Fp2, the left central C3, the right central C4, the left parietal lobe P3, the right parietal lobe P4, the left occipital lobe O1 and the right occipital lobe O2; In step S3, for the left frontal lobe Fp1' and the right frontal lobe Fp2' in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left frontal lobe Fp1' and the right frontal lobe Fp2', and if yes, interpolate the abnormal signals to replace the abnormal signals to obtain normal EEG signals, and if no, use the preliminary EEG signals of each channel as normal EEG signals; for the normal EEG signals of each channel in the left frontal lobe Fp1' and the right frontal lobe Fp2': use the EEG signal of the i-th channel in the left frontal lobe Fp1' and the EEG signal of the j-th channel in the right frontal lobe Fp2' to calculate the correlation coefficient Fp'-X between the i-th channel and the j-th channel ij , thereby obtaining 2M correlation coefficients Fp'-X for the left frontal lobe Fp1' and the right frontal lobe Fp2' ij , the two channels with the largest correlation coefficient are selected as the optimal channel of the left prefrontal lobe Fp1' and the optimal channel of the right prefrontal lobe Fp2' respectively, the EEG signal of the optimal channel of the left prefrontal lobe Fp1' is used as the optimal EEG signal of the left prefrontal lobe Fp1', and the EEG signal of the optimal channel of the right prefrontal lobe Fp2' is used as the optimal EEG signal of the right prefrontal lobe Fp2'; For the left central C3' and right central C4' in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left central C3' and the right central C4'. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals; For the normal EEG signals of each channel in the left central C3' and the right central C4': use the EEG signal of the i-th channel in the left central C3' and the EEG signal of the j-th channel in the right central C4' to calculate the correlation coefficient C'-X between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients C'-X for the left center C3' and the right center C4' ij , the two channels with the largest correlation coefficient are selected as the optimal channel of the left center C3' and the optimal channel of the right center C4' respectively, the EEG signal of the optimal channel of the left center C3' is used as the optimal EEG signal of the left center C3', and the EEG signal of the optimal channel of the right center C4' is used as the optimal EEG signal of the right center C4'; For the left parietal lobe P3' and the right parietal lobe P4' in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left parietal lobe P3' and the right parietal lobe P4'. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals; For the normal EEG signals of each channel in the left parietal lobe P3' and the right parietal lobe P4': use the EEG signal of the i-th channel in the left parietal lobe P3' and the EEG signal of the j-th channel in the right parietal lobe P4' to calculate the correlation coefficient P'-X between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients P'-X for the left parietal lobe P3' and the right parietal lobe P4' ij , the two channels with the largest correlation coefficient are selected as the optimal channel of the left parietal lobe P3' and the optimal channel of the right parietal lobe P4' respectively, the EEG signal of the optimal channel of the left parietal lobe P3' is used as the optimal EEG signal of the left parietal lobe P3', and the EEG signal of the optimal channel of the right parietal lobe P4' is used as the optimal EEG signal of the right parietal lobe P4'; For the left occipital lobe O1' and the right occipital lobe O2' in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the left occipital lobe O1' and the right occipital lobe O2'. If yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals. If no, the preliminary EEG signals of each channel are used as normal EEG signals; For the normal EEG signals of each channel in the left occipital lobe O1' and the right occipital lobe O2': use the EEG signal of the i-th channel in the left occipital lobe O1' and the EEG signal of the j-th channel in the right occipital lobe O2' to calculate the correlation coefficient O'-X between the i-th channel and the j-th channel. ij , thereby obtaining 2M correlation coefficients O'-X for the left occipital lobe O1' and the right occipital lobe O2' ij , the two channels corresponding to the largest correlation coefficient are selected as the optimal channel of the left occipital lobe O1' and the optimal channel of the right occipital lobe O2' respectively, the EEG signal of the optimal channel of the left occipital lobe O1' is used as the optimal EEG signal of the left occipital lobe O1', and the EEG signal of the optimal channel of the right occipital lobe O2' is used as the optimal EEG signal of the right occipital lobe O2'.

5. The method for detecting attention state based on EEG signals as claimed in claim 1, characterized in that: In step S4, the attention time domain feature is the variance feature, and the attention frequency domain feature is the α wave power spectrum density in the α band and the β wave power spectrum density in the β band; For the connection feature L between any two brain functional areas Ea and Eb in the K brain functional areas in the non-interference detection mode E : Where H refers to the number of sampling time points, l is an imaginary number, φ Ea (h) represents the phase of the optimal EEG signal of the brain functional area Ea at the hth sampling time point, φ Eb (h) represents the phase of the optimal EEG signal of the brain functional area Eb at the hth sampling time point; For the connection feature L between any two brain functional areas Ea' and Eb' in the K brain functional areas in the interference detection mode E ': where φ E a'(h) represents the phase of the optimal EEG signal of the brain functional area Ea' at the hth sampling time point. E b'(h) represents the phase of the optimal EEG signal of the brain functional area Eb' at the hth sampling time point.

6. A system for detecting attention status based on EEG signals, characterized in that: It includes an EEG acquisition module, an EEG processing module, an optimal EEG screening module, an EEG feature extraction module and an attention state detection module; The EEG acquisition module is used to obtain basic information containing the age and identity of the person to be detected, and sequentially play videos in the non-interference detection mode and the interference detection mode that match the age of the person to be detected for the person to be detected to watch, the video in the non-interference detection mode contains N sample videos without interference, and the video in the interference detection mode contains N sample videos with interference. The playback time of each sample video in the non-interference detection mode is the same and the required attention level is different. The playback time of each sample video in the interference detection mode is the same and the required attention level is different. The attention level required for the nth sample video in the non-interference detection mode is the same as the attention level required for the nth sample video in the interference detection mode. The original EEG signals of M channels of each brain functional area in K brain functional areas during the viewing process of the person to be detected in the non-interference detection mode and the interference detection mode are respectively collected, where N is a positive integer and N≥2, 1≤n≤N, K is a positive integer and K≥2, and M is a positive integer and M≥2; The EEG processing module is used to perform preliminary processing of noise removal, filtering and artifact removal on the original EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode, respectively, to obtain preliminary EEG signals of each channel of each brain functional area in the non-interference detection mode and the interference detection mode, respectively; The optimal EEG screening module is used for selecting an optimal channel EEG signal from the M channels of the kth brain functional area in the non-interference detection mode as the optimal EEG signal of the kth brain functional area, based on which the optimal EEG signal of each brain functional area in the non-interference detection mode is selected; for the EEG signal of the M channels of the kth brain functional area in the interference detection mode, selecting an optimal channel EEG signal from the M channels as the optimal EEG signal of the kth brain functional area, based on which the optimal EEG signal of each brain functional area in the interference detection mode is selected; The EEG feature extraction module is used to extract features from the optimal EEG signals of K brain function areas in the non-interference detection mode, obtain the attention time domain features and attention frequency domain features corresponding to each sample video of each brain function area, and the connection features between any two brain function areas, and construct the EEG attention feature set in the non-interference detection mode from the attention time domain features and attention frequency domain features corresponding to each sample video of each brain function area arranged in sequence first and the connection features between two brain function areas arranged in sequence later; Extract features of the optimal EEG signals of K brain functional areas in the interference detection mode, obtain the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area, and the connection features between any two brain functional areas, and construct the EEG attention feature set in the interference detection mode by the attention time domain features and attention frequency domain features corresponding to each sample video of each brain functional area arranged in sequence first and the connection features between two brain functional areas arranged in sequence later; The attention state detection module is used to input the EEG attention feature set in the undisturbed detection mode and the EEG attention feature set in the interfered detection mode of the person to be detected into the trained attention state detection model constructed based on the convolutional neural network model for detection, so as to output the attention state detection result of the person to be detected. The attention state detection result is divided into 0 to 10 levels, respectively representing the degree of attention state 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100%, among which 0% is unconscious and 100% is over-concentration.

7. The attention state detection system based on EEG signals as claimed in claim 6, characterized in that: In the EEG acquisition module, the K brain functional areas correspond to each other in pairs, and K is an even number; In the optimal EEG screening module, for any two corresponding brain function areas Ea and Eb among the K brain function areas in the non-interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the brain function areas Ea and Eb, and if yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals, and if no, use the preliminary EEG signals of each channel as normal EEG signals; For the normal EEG signals of each channel in the brain functional areas Ea and Eb: the correlation coefficient EX between the i-th channel and the j-th channel is calculated using the EEG signal of the i-th channel in the brain functional area Ea and the EEG signal of the j-th channel in the brain functional area Eb. ij , thus obtaining 2M correlation coefficients EX for brain functional areas Ea and Eb ij , the two channels corresponding to the largest correlation coefficient are selected as the optimal channel of brain function area Ea and the optimal channel of brain function area Eb respectively, the EEG signal of the optimal channel of brain function area Ea is used as the optimal EEG signal of brain function area Ea, and the EEG signal of the optimal channel of brain function area Eb is used as the optimal EEG signal of brain function area Eb, based on which the optimal EEG signal of each brain function area in the interference-free detection mode is selected, 1≤i≤M, 1≤j≤M, abnormal signal refers to a signal with abnormal amplitude or a missing signal; For any two corresponding brain function areas Ea' and Eb' among the K brain function areas in the interference detection mode: determine whether there are abnormal signals in the preliminary EEG signals of each channel in the brain function areas Ea' and Eb', and if yes, perform interpolation processing on the abnormal signals to replace the abnormal signals to obtain normal EEG signals, and if no, use the preliminary EEG signals of each channel as normal EEG signals; For the normal EEG signals of each channel in the brain functional areas Ea' and Eb': use the EEG signal of the ith channel in the brain functional area Ea' and the EEG signal of the jth channel in the brain functional area Eb' to calculate the correlation coefficient E'-X between the ith channel and the jth channel ij , thus obtaining 2M correlation coefficients E'-X for brain functional areas Ea' and Eb' ij , the two channels corresponding to the largest correlation coefficient are selected as the optimal channel of the brain functional area Ea' and the optimal channel of the brain functional area Eb' respectively, the EEG signal of the optimal channel of the brain functional area Ea' is used as the optimal EEG signal of the brain functional area Ea', and the EEG signal of the optimal channel of the brain functional area Eb' is used as the optimal EEG signal of the brain functional area Eb'. Based on this, the optimal EEG signal of each brain functional area in the interference detection mode is selected.

8. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for detecting the attention state based on electroencephalogram signals as described in any one of claims 1 to 5 is implemented.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed, the method for detecting the attention state based on electroencephalogram signals as described in any one of claims 1 to 5 is implemented.

Images