Emotion recognition method, device and equipment based on electroencephalogram and storage medium

Abstract

The invention relates to the technical field of emotion recognition, in particular to an electroencephalogram-based emotion recognition method, device and equipment and a storage medium, and the method comprises the steps: executing model optimization processing on an electroencephalogram consistency detection model through an electroencephalogram information detection model to obtain a target consistency detection model associated with the electroencephalogram consistency detection model; acquiring first and second electroencephalogram information in an electroencephalogram information pair to be identified, and performing electroencephalogram similarity detection processing on the first and second electroencephalogram information according to a target similarity detection model to output a similarity calculation result between the first and second electroencephalogram information in the electroencephalogram information pair, characterizing whether the first electroencephalogram information and the second electroencephalogram information meet emotion consistency or characterizing whether to-be-identified personnel associated with the first electroencephalogram information and the second electroencephalogram information meet emotion consistency. The emotion stability consistency can be effectively detected based on the electroencephalogram of the person, emotion fluctuation is effectively coped, and reasonable countermeasures are taken in time.

Description

Technical Field

[0001] The present invention relates to the field of emotion recognition technology, and more specifically to an electroencephalogram-based emotion recognition method, device, equipment and storage medium. Background Art

[0002] Emotion recognition is used in many scenarios. For example, emotion recognition operations can be used to mine the features of corresponding user data, so that operations such as user data recommendations can be performed based on the mined features.

[0003] However, the existing technology of emotion recognition analysis based on human EEG is inaccurate and imprecise, and has a narrow coverage of emotion expression analysis. Therefore, we urgently need a visual perception emotion recognition method based on multi-dimensional data analysis of EEG that is more accurate and rigorous in emotion recognition analysis and has a wider coverage of emotion expression analysis. Summary of the invention

[0004] According to a first aspect of the present invention, the present invention claims protection for an electroencephalogram-based emotion recognition method, comprising: Through the EEG information detection model, Figure 1 The consistency detection model performs model optimization processing to obtain the EEG Figure 1 The target consistency detection model associated with the consistency detection model, the EEG information detection model and the EEG Figure 1 All consistency detection models are deep learning models; Obtaining first electroencephalogram information and second electroencephalogram information in a pair of electroencephalogram information to be identified, wherein the first electroencephalogram information and the second electroencephalogram information are used to characterize potential emotions of the corresponding person to be identified, and the first electroencephalogram information and the second electroencephalogram information belong to a brain wave time series feature sequence; According to the target similarity detection model, the first EEG information and the second EEG information are subjected to EEG similarity detection processing to output a similarity calculation result between the first EEG information and the second EEG information in the EEG information pair, wherein the similarity calculation result is used to characterize whether the first EEG information and the second EEG information satisfy emotional consistency or whether the person to be identified associated with the first EEG information and the person to be identified associated with the second EEG information satisfy emotional consistency.

[0005] The step of performing electroencephalogram similarity detection processing on the first electroencephalogram information and the second electroencephalogram information according to the target similarity detection model to output a similarity calculation result between the first electroencephalogram information and the second electroencephalogram information in the electroencephalogram information pair includes: According to the target similarity detection model, the first electroencephalogram information and the second electroencephalogram information are sequentially subjected to feature interval truncation processing to output interval truncation time-series brain waves associated with the first electroencephalogram information and interval truncation time-series brain waves associated with the second electroencephalogram information; Performing feature comparison processing on the interval truncated time-series EEG waves associated with the first EEG information and the interval truncated time-series EEG waves associated with the second EEG information to form a first full-cycle time-series EEG wave including full-cycle related information of the first EEG information, and forming a second full-cycle time-series EEG wave including full-cycle related information of the second EEG information; Performing cascade processing on the first complete cycle time-series brain wave and the second complete cycle time-series brain wave to form associated cascade complete cycle time-series brain wave; The cascaded full-cycle time-series electroencephalogram waves are subjected to detection processing of similarity calculation results to output similarity calculation results between the first electroencephalogram information and the second electroencephalogram information in the electroencephalogram information pair.

[0006] Furthermore, the EEG information detection model is used to detect the EEG Figure 1 The consistency detection model performs model optimization processing to obtain the EEG Figure 1 The steps of the target consistency detection model associated with the consistency detection model include: According to the electroencephalogram information detection model, electroencephalogram information detection processing is performed on the first reference electroencephalogram information in the reference electroencephalogram information pair to output an electroencephalogram detection result associated with the first reference electroencephalogram information, the first reference electroencephalogram information belongs to one of the reference electroencephalogram information in the reference electroencephalogram information pair, and the first reference electroencephalogram information is used to characterize the potential emotion of the corresponding reference electroencephalogram, the first reference electroencephalogram information belongs to the brain wave time series feature sequence, and the electroencephalogram detection result is used to characterize whether the first reference electroencephalogram information and the second reference electroencephalogram information in the reference electroencephalogram information pair meet emotion consistency or characterize whether the reference electroencephalogram associated with the first reference electroencephalogram information and the reference electroencephalogram associated with the second reference electroencephalogram information meet emotion consistency; According to the EEG extreme value oscillation scene of the reference EEG information pair sequence, performing risk analysis processing on the EEG detection result associated with the first reference EEG information to output a risk characterization index associated with the reference EEG information pair; According to the EEG similarity detection model, performing EEG similarity detection processing on the first reference EEG information and the second reference EEG information in the reference EEG information pair to output an EEG similarity detection result associated with the reference EEG information pair; If the electroencephalogram similarity detection result associated with the reference electroencephalogram information pair is the existence ratio of each current similarity data of the reference electroencephalogram information pair, a first parsing process is performed on the low similarity feature of the existence ratio of each current similarity data of the reference electroencephalogram information pair to form an associated target parsing output indicator; Performing a second parsing process on the target parsing output indicator and the risk characterization indicator associated with the reference EEG information pair to form a model optimization cost indicator associated with the EEG similarity detection model; A model optimization process is performed on the electroencephalogram similarity detection model so that the model optimization cost index is less than or equal to a pre-configured reference model optimization cost index, so as to obtain a target similarity detection model associated with the electroencephalogram similarity detection model.

[0007] Furthermore, before the step of performing electroencephalogram information detection processing on the first reference electroencephalogram information in the reference electroencephalogram information pair according to the electroencephalogram information detection model to output the electroencephalogram detection result associated with the first reference electroencephalogram information, the electroencephalogram information detection model is used to detect the electroencephalogram information. Figure 1 The consistency detection model performs model optimization processing to obtain the EEG Figure 1 The step of associating the target consistency detection model with the consistency detection model also includes: performing model optimization processing on a candidate electroencephalogram information detection model to form an electroencephalogram information detection model after model optimization based on current similarity data of third reference electroencephalogram information and fourth reference electroencephalogram information in a reference electroencephalogram information pair sequence and the third reference electroencephalogram information, wherein the third reference electroencephalogram information and the fourth reference electroencephalogram information belong to a reference electroencephalogram information pair in the reference electroencephalogram information pair sequence; The step of performing electroencephalogram information detection processing on the first reference electroencephalogram information in the reference electroencephalogram information pair according to the electroencephalogram information detection model to output an electroencephalogram detection result associated with the first reference electroencephalogram information comprises: According to the electroencephalogram information detection model optimized by the model, electroencephalogram information detection processing is performed on the first reference electroencephalogram information to output an electroencephalogram detection result associated with the first reference electroencephalogram information.

[0008] Further, the step of performing model optimization processing on the candidate EEG information detection model based on the current similarity data of the third reference EEG information and the fourth reference EEG information in the sequence and the third reference EEG information to form an optimized EEG information detection model comprises: Performing classification processing on the reference electroencephalogram information pair sequence to form a first number of reference electroencephalogram information pair subsequences associated with the reference electroencephalogram information pair sequence; For any one of the reference EEG information pair subsequences to be processed in the first number of reference EEG information pair subsequences, perform the following processing: Marking reference electroencephalogram information pair subsequences other than the reference electroencephalogram information pair subsequence to be processed to mark them as reference electroencephalogram information pair subsequences to be processed, and performing model optimization processing on candidate electroencephalogram information detection models based on the reference electroencephalogram information pair subsequence to be processed to form an electroencephalogram information detection model after model optimization; The step of performing electroencephalogram information detection processing on the first reference electroencephalogram information based on the electroencephalogram information detection model optimized by the model to output an electroencephalogram detection result associated with the first reference electroencephalogram information comprises: Based on the reference EEG information pair subsequence to be processed and the EEG information detection model optimized by the model, EEG information detection processing is performed on the reference EEG information pair subsequence to be processed to output the EEG detection result associated with the reference EEG information pair subsequence to be processed.

[0009] Furthermore, the step of performing electroencephalogram information detection processing on the first reference electroencephalogram information based on the electroencephalogram information detection model optimized by the model to output an electroencephalogram detection result associated with the first reference electroencephalogram information includes: According to the electroencephalogram information detection model optimized by the model, the first reference electroencephalogram information is subjected to characteristic interval truncation processing to output interval truncation time-series brain waves associated with the first reference electroencephalogram information; Perform feature comparison processing on the interval truncated time-series electroencephalogram associated with the first reference electroencephalogram information to form a whole-cycle time-series electroencephalogram associated with the first reference electroencephalogram information including whole-cycle related information; The whole cycle time-series electroencephalogram wave including the whole cycle related information is subjected to electroencephalogram information detection processing to output an electroencephalogram detection result associated with the first reference electroencephalogram information.

[0010] Further, the electroencephalogram detection result associated with the first reference electroencephalogram information is a ratio of each current similarity data of the first reference electroencephalogram information; The step of performing risk analysis processing on the electroencephalogram detection result associated with the first reference electroencephalogram information according to the electroencephalogram extreme value oscillation scene of the reference electroencephalogram information pair sequence to output the risk characterization index associated with the reference electroencephalogram information pair comprises: Analyzing the emotional fluctuation ratio of each current similarity data of the first reference electroencephalogram information and the second reference electroencephalogram information in the reference electroencephalogram information pair when the reference electroencephalogram information pair sequence is in an electroencephalogram extreme oscillation scene; Based on the ratio of each current similarity data of the first reference EEG information and the emotional fluctuation ratio of each current similarity data, the EEG detection result of the first reference EEG information is subjected to risk analysis processing to output a risk characterization index associated with the reference EEG information.

[0011] According to a second aspect of the present invention, the present invention claims protection for an electroencephalogram-based emotion recognition device, comprising: Model optimization module, used to detect the model through EEG information, Figure 1 The consistency detection model performs model optimization processing to obtain the EEG Figure 1 The target consistency detection model associated with the consistency detection model, the EEG information detection model and the EEG Figure 1 All consistency detection models are deep learning models; an electroencephalogram information extraction module, used to obtain first electroencephalogram information and second electroencephalogram information in an electroencephalogram information pair to be identified, wherein the first electroencephalogram information and the second electroencephalogram information are used to characterize the potential emotions of the corresponding person to be identified, and the first electroencephalogram information and the second electroencephalogram information belong to a brain wave time series feature sequence; An electroencephalogram similarity detection module is used to perform electroencephalogram similarity detection processing on the first electroencephalogram information and the second electroencephalogram information according to a target similarity detection model to output a similarity calculation result between the first electroencephalogram information and the second electroencephalogram information in the electroencephalogram information pair, wherein the similarity calculation result is used to characterize whether the first electroencephalogram information and the second electroencephalogram information meet emotional consistency or whether the person to be identified associated with the first electroencephalogram information and the person to be identified associated with the second electroencephalogram information meet emotional consistency.

[0012] According to the third aspect of the present invention, the present invention seeks protection for an electroencephalogram-based emotion recognition device, comprising a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to implement the electroencephalogram-based emotion recognition method.

[0013] According to a fourth aspect of the present invention, the present invention seeks protection for a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the electroencephalogram-based emotion recognition method.

[0014] The present invention relates to the field of emotion recognition technology, and in particular to an electroencephalogram-based emotion recognition method, device, equipment and storage medium, which detects the emotion of the electroencephalogram using an electroencephalogram information detection model. Figure 1 The consistency detection model performs model optimization processing to obtain EEG Figure 1 The target consistency detection model is associated with the consistency detection model; the first and second EEG information in the EEG information pair to be identified are obtained, and according to the target similarity detection model, the first and second EEG information are subjected to EEG similarity detection processing to output the similarity calculation result between the first and second EEG information in the EEG information pair, which characterizes whether the first EEG information and the second EEG information meet the emotional consistency or whether the person to be identified associated with the first and second EEG information meets the emotional consistency. The present invention can effectively detect the consistency of emotional stability based on the EEG of the person, and effectively respond to emotional fluctuations and make reasonable countermeasures in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flowchart of an electroencephalogram-based emotion recognition method claimed in an embodiment of the present application; Figure 2 This is a structural module diagram of an electroencephalogram-based emotion recognition device for which protection is sought in an embodiment of the present application. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions 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 only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0017] pass Figure 1 The embodiment of the present invention also provides an EEG-based emotion recognition method, which can be applied to the above-mentioned EEG-based emotion recognition system. The method steps defined in the process related to the EEG-based emotion recognition method can be implemented by the EEG-based emotion recognition system. Figure 1 The specific process shown is explained in detail.

[0018] Step S110, using the EEG information detection model to Figure 1 The consistency detection model performs model optimization processing to obtain the EEG Figure 1 The target consistency detection model associated with the consistency detection model.

[0019] In an embodiment of the present invention, the electroencephalogram-based emotion recognition system can detect the electroencephalogram information model and Figure 1 The consistency detection model performs model optimization processing to obtain the EEG Figure 1 The EEG information detection model and the EEG consistency detection model are associated with the target consistency detection model. Figure 1 All consistency detection models are deep learning models.

[0020] Step S120, obtaining the first electroencephalogram information and the second electroencephalogram information in the electroencephalogram information pair to be identified.

[0021] In an embodiment of the present invention, the electroencephalogram-based emotion recognition system can obtain the first electroencephalogram information and the second electroencephalogram information in the electroencephalogram information pair to be identified. The first electroencephalogram information and the second electroencephalogram information are used to characterize the potential emotions of the corresponding person to be identified, such as the characteristics of different periods. The first electroencephalogram information and the second electroencephalogram information belong to the brain wave time series feature sequence. For example, the first electroencephalogram information and the second electroencephalogram information can be the waveform of the same electroencephalogram, and the first electroencephalogram information and the second electroencephalogram information can also be different waveforms of the same electroencephalogram.

[0022] Step S130, performing electroencephalogram similarity detection processing on the first electroencephalogram information and the second electroencephalogram information according to the target similarity detection model to output a similarity calculation result between the first electroencephalogram information and the second electroencephalogram information in the electroencephalogram information pair.

[0023] In an embodiment of the present invention, the EEG-based emotion recognition system can perform EEG similarity detection processing on the first EEG information and the second EEG information according to a target similarity detection model to output a similarity calculation result between the first EEG information and the second EEG information in the EEG information pair. The similarity calculation result is used to characterize whether the first EEG information and the second EEG information meet emotional consistency or whether the person to be identified associated with the first EEG information and the person to be identified associated with the second EEG information meet emotional consistency, that is, the similarity calculation result can include a matching ratio index and a mismatching ratio index. For example, for the same EEG, it can be used to characterize whether the characteristics of the EEG at different periods meet emotional consistency, and for two different EEGs, it can be used to characterize whether the characteristics of the two EEGs meet emotional consistency.

[0024] Based on the above content, due to the EEG Figure 1During the process of model optimization, the EEG information detection model will be used to detect the target consistency detection model, so that the accuracy of the target consistency detection model formed can be higher. Therefore, the reliability of data matching can be improved to a certain extent.

[0025] It is understandable that, in some possible implementations, step S110 in the above description, i.e., detecting the EEG information using the EEG information detection model, Figure 1 The consistency detection model performs model optimization processing to obtain the EEG Figure 1 The step of associating the target consistency detection model with the consistency detection model may further include the specific implementation contents described in detail below: According to the electroencephalogram information detection model, electroencephalogram information detection processing is performed on the first reference electroencephalogram information in the reference electroencephalogram information pair to output an electroencephalogram detection result associated with the first reference electroencephalogram information, the first reference electroencephalogram information belongs to one of the reference electroencephalogram information in the reference electroencephalogram information pair, and the first reference electroencephalogram information is used to characterize the potential emotion of the corresponding reference electroencephalogram, the first reference electroencephalogram information belongs to the brain wave time series feature sequence, and the electroencephalogram detection result is used to characterize the first reference electroencephalogram information and the second one in the reference electroencephalogram information pair. Whether the reference EEG information satisfies the emotional consistency or whether the reference EEG associated with the first reference EEG information and the reference EEG associated with the second reference EEG information satisfy the emotional consistency, that is, the EEG information detection model can perform analysis and detection on only one reference EEG information in the reference EEG information pair to obtain the EEG detection result associated with the two reference EEG information in the reference EEG information pair, while the EEG similarity detection model performs analysis and detection on the two reference EEG information in the reference EEG information pair; According to the EEG extreme value oscillation scene of the reference EEG information pair sequence, performing risk analysis processing on the EEG detection result associated with the first reference EEG information to output a risk characterization index associated with the reference EEG information pair; According to the EEG similarity detection model, performing EEG similarity detection processing on the first reference EEG information and the second reference EEG information in the reference EEG information pair to output an EEG similarity detection result associated with the reference EEG information pair; If the EEG similarity detection result associated with the reference EEG information pair is the existence ratio of each current similarity data of the reference EEG information pair, such as a matching ratio index and an unmatched ratio index, a first parsing process is performed on a low similarity feature of the existence ratio of each current similarity data of the reference EEG information pair to form an associated target parsing output index, for example, the product between the low similarity feature and the existence ratio (index) is equal to a fixed value, such as a numerical value of 1, and the first parsing process may be a logarithmic process, etc.; Performing a second parsing process on the target parsing output indicator and the risk characterization indicator associated with the reference EEG information pair to form a model optimization cost indicator associated with the EEG similarity detection model, where, for reference, the risk characterization indicator associated with the reference EEG information pair may include the risk characterization indicator of the reference EEG information pair associated with the first current similarity data and the risk characterization indicator of the reference EEG information pair associated with the second current similarity data, and the target parsing output indicator may include a first target parsing output indicator associated with the first current similarity data and a second target parsing output indicator associated with the second current similarity data, so that a weighted summation may be performed on the first target parsing output indicator and the second target parsing output indicator based on the associated risk characterization indicator to obtain the model optimization cost indicator associated with the EEG similarity detection model; A model optimization process is performed on the electroencephalogram similarity detection model so that the model optimization cost index is less than or equal to a pre-configured reference model optimization cost index, so as to obtain a target similarity detection model associated with the electroencephalogram similarity detection model.

[0026] It can be understood that, in some possible implementations, before the step of performing EEG information detection processing on the first reference EEG information in the reference EEG information pair according to the EEG information detection model to output the EEG detection result associated with the first reference EEG information, the step of performing EEG information detection processing on the first reference EEG information in the reference EEG information pair according to the EEG information detection model to output the EEG detection result associated with the first reference EEG information Figure 1 The consistency detection model performs model optimization processing to obtain the EEG Figure 1 The step of associating the target consistency detection model with the consistency detection model, that is, step S110, may further include the specific implementation content described in detail below: Based on the current similarity data of the third reference EEG information and the fourth reference EEG information in the reference EEG information pair sequence and the third reference EEG information, a candidate EEG information detection model is subjected to model optimization processing to form a model-optimized EEG information detection model, wherein the third reference EEG information and the fourth reference EEG information belong to a reference EEG information pair in the reference EEG information pair sequence, that is, based on the candidate EEG information detection model, the third reference EEG information is subjected to analysis and detection to obtain an associated detection result, and then, based on the difference between the detection result and the current similarity data, a model optimization processing can be performed on the candidate EEG information detection model, such as performing model optimization processing along the path of reducing the energy of the difference to form a model-optimized EEG information detection model.

[0027] Based on the above content, the step of performing EEG information detection processing on the first reference EEG information in the reference EEG information pair according to the EEG information detection model to output the EEG detection result associated with the first reference EEG information may include: performing EEG information detection processing on the first reference EEG information according to the EEG information detection model optimized by the model to output the EEG detection result associated with the first reference EEG information.

[0028] It can be understood that, in some possible implementations, the step of performing model optimization processing on the candidate EEG information detection model based on the current similarity data of the third reference EEG information and the fourth reference EEG information in the sequence and the third reference EEG information to form an EEG information detection model after model optimization may further include the specific implementation contents described in detail below: Performing classification processing on the reference electroencephalogram information pair sequence to form a first number of reference electroencephalogram information pair subsequences associated with the reference electroencephalogram information pair sequence, wherein the classification processing may be random or may be configured according to requirements; For any one of the reference EEG information pair subsequences to be processed in the first number of reference EEG information pair subsequences, perform the following processing: The reference EEG information pair subsequences other than the reference EEG information pair subsequence to be processed are marked as reference EEG information pair subsequences to be processed, and, based on the reference EEG information pair subsequence to be processed, the candidate EEG information detection model is subjected to model optimization processing to form a model-optimized EEG information detection model.

[0029] Based on the above content, the step of performing EEG information detection processing on the first reference EEG information according to the EEG information detection model optimized by the model to output the EEG detection result associated with the first reference EEG information includes: performing EEG information detection processing on the subsequence of the reference EEG information pair to be processed according to the EEG information detection model optimized by the model to output the EEG detection result associated with the subsequence of the reference EEG information pair to be processed.

[0030] That is, a part of the data in the reference EEG information pair sequence is used to perform model optimization processing on the candidate EEG information detection model, and the other part of the data is used to perform model optimization processing on the candidate EEG information detection model. Figure 1 The consistency detection model performs model optimization processing.

[0031] It can be understood that, in some possible implementations, the step of performing electroencephalogram information detection processing on the first reference electroencephalogram information based on the electroencephalogram information detection model optimized by the model to output an electroencephalogram detection result associated with the first reference electroencephalogram information may further include the specific implementation contents described in detail below: According to the EEG information detection model optimized by the model, the first reference EEG information is subjected to characteristic interval truncation processing to output the interval truncation time-series EEG waves associated with the first reference EEG information, which may refer to the relevant description below; Performing feature comparison processing on the interval truncated time-series EEG waves associated with the first reference EEG information to form a whole-cycle time-series EEG wave associated with the first reference EEG information including whole-cycle related information, which may refer to the relevant description below; The whole cycle time-series electroencephalogram wave including the whole cycle related information is subjected to electroencephalogram information detection processing to output the electroencephalogram detection result associated with the first reference electroencephalogram information, and the relevant description below can be referred to.

[0032] It can be understood that, in some possible implementations, the electroencephalogram detection result associated with the first reference electroencephalogram information is the ratio of each current similarity data of the first reference electroencephalogram information. Based on this, the step of performing risk analysis processing on the electroencephalogram detection result associated with the first reference electroencephalogram information according to the brain wave extreme oscillation scene of the reference electroencephalogram information pair sequence to output the risk characterization index associated with the reference electroencephalogram information pair may further include the specific implementation content described in detail below: Analyze the emotional fluctuation ratio of each current similarity data of the first reference EEG information and the second reference EEG information in the reference EEG information pair when the reference EEG information pair sequence is in an EEG extreme oscillation scene, that is, the data distribution of the reference EEG information pair sequence is reasonable and the emotional fluctuation is high; Based on the ratio of each current similarity data of the first reference EEG information and the emotional fluctuation ratio of each current similarity data, the EEG detection result of the first reference EEG information is subjected to risk analysis processing (i.e., comparative analysis is performed on the ratio) to output the risk characterization index associated with the reference EEG information.

[0033] It can be understood that, in some possible implementations, the step of analyzing the emotional fluctuation ratio of each current similarity data of the first reference EEG information and the second reference EEG information in the reference EEG information pair when the reference EEG information pair sequence is in an EEG extreme oscillation scene may further include the specific implementation contents described in detail below: According to the ratio of occurrence of each current similarity data and the ratio of each current similarity data of the first reference electroencephalogram information, the to-be-determined emotion fluctuation ratio is iteratively calculated to form the emotion fluctuation ratio of each current similarity data of the first reference electroencephalogram information and the second reference electroencephalogram information in the reference electroencephalogram information pair, the emotion fluctuation ratio of each current similarity data including a first emotion fluctuation ratio of the first current similarity data representing a match and a second emotion fluctuation ratio of the second current similarity data representing a mismatch; for reference, the ratio of the occurrence ratio of the current similarity data representing a match to the occurrence ratio of the current similarity data representing a mismatch can be calculated first, and then, based on the ratio, the to-be-determined emotion fluctuation ratio is iteratively calculated.

[0034] It can be understood that, in some possible implementations, the step of performing risk analysis processing on the EEG detection result of the first reference EEG information based on the ratio of each current similarity data of the first reference EEG information and the emotional fluctuation ratio of each current similarity data to output the risk characterization index associated with the reference EEG information may further include the specific implementation content described in detail below: Obtaining a first ratio of first current similarity data of the first reference electroencephalogram information, and obtaining a second ratio of second current similarity data of the first reference electroencephalogram information, and obtaining a first emotion fluctuation ratio of the first current similarity data, and obtaining a second emotion fluctuation ratio of the second current similarity data; Performing compensation correction processing on a first multiplication correction index between the first ratio and the second emotion fluctuation ratio, and a second multiplication correction index between the second ratio and the first emotion fluctuation ratio, so as to output an associated target compensation correction index, that is, firstly performing multiplication on the first ratio and the second emotion fluctuation ratio, and then performing multiplication on the second ratio and the first emotion fluctuation ratio, and then performing addition on the two multiplication results, so as to obtain an associated target compensation correction index; Performing division analysis processing on the target compensation correction index and the first multiplication correction index to output a risk characterization index of the reference electroencephalogram information associated with the first current similarity data, for example, the target compensation correction index can be divided by the first multiplication correction index to obtain the risk characterization index; A division analysis process is performed on the target compensation correction index and the second multiplication correction index to output a risk characterization index of the reference electroencephalogram information associated with the second current similarity data. For example, the target compensation correction index can be divided by the second multiplication correction index to obtain the risk characterization index.

[0035] It can be understood that, in some possible implementations, the step of performing EEG similarity detection processing on the first reference EEG information and the second reference EEG information in the reference EEG information pair based on the EEG similarity detection model to output an EEG similarity detection result associated with the reference EEG information pair may further include the specific implementation contents described in detail below: The first reference electroencephalogram information and the second reference electroencephalogram information are sequentially subjected to feature interval truncation processing to output interval truncated time-series electroencephalogram waves associated with the first reference electroencephalogram information and interval truncated time-series electroencephalogram waves associated with the second reference electroencephalogram information. That is, the first reference electroencephalogram information and the second reference electroencephalogram information can be sequentially mapped to the feature space to perform representation in the form of vectors, thereby obtaining interval truncated time-series electroencephalogram waves associated with the first reference electroencephalogram information and interval truncated time-series electroencephalogram waves associated with the second reference electroencephalogram information. Performing feature comparison processing on the interval-truncated time-series EEG waves associated with the first reference EEG information and the interval-truncated time-series EEG waves associated with the second reference EEG information to form an example first full-cycle time-series EEG wave including the information related to the whole cycle of the first reference EEG information, and to form an example second full-cycle time-series EEG wave including the information related to the whole cycle of the second reference EEG information; Performing cascade pair processing on the example first full cycle time sequence brain wave and the example second full cycle time sequence brain wave to form associated example cascaded full cycle time sequence brain wave, where the example cascaded full cycle time sequence brain wave may be {the example first full cycle time sequence brain wave, the example second full cycle time sequence brain wave}; The example cascaded whole cycle time series EEG waves are subjected to detection processing of similarity calculation results to output the reference EEG information associated EEG similarity detection result. For example, the example cascaded whole cycle time series EEG waves can be first subjected to full connection processing to form an associated example full connection vector. Then, the example full connection vector can be processed by a classification function to obtain an associated EEG similarity detection result.

[0036] Among them, it can be understood that, in some possible implementations, the step of performing feature comparison processing on the interval truncated time-series EEG waves associated with the first reference EEG information and the interval truncated time-series EEG waves associated with the second reference EEG information to form an example first whole cycle time-series EEG wave including whole cycle related information of the first reference EEG information, and forming an example second whole cycle time-series EEG wave including whole cycle related information of the second reference EEG information, may further include the specific implementation content described in detail below: Performing a first energy feature comparison process on the interval truncated time-series electroencephalogram waves associated with the first reference electroencephalogram information to form a first energy time-series electroencephalogram wave associated with the first reference electroencephalogram information. For example, the interval truncated time-series electroencephalogram waves associated with the first reference electroencephalogram information may be subjected to a feature comparison process in a forward-to-backward order to form a first energy time-series electroencephalogram wave associated with the first reference electroencephalogram information. Performing feature comparison processing of a second energy on the interval truncated time-series electroencephalogram associated with the first reference electroencephalogram information to form a second energy time-series electroencephalogram associated with the first reference electroencephalogram information, wherein the feature comparison processing of the first energy is opposite to the feature comparison processing of the second energy, for example, the interval truncated time-series electroencephalogram associated with the first reference electroencephalogram information may be subjected to feature comparison processing in a backward order to form a first energy time-series electroencephalogram associated with the first reference electroencephalogram information; Aggregating the first energy time-series brainwave and the second energy time-series brainwave to form a first full-cycle time-series brainwave including full-cycle related information of the first reference electroencephalogram information. For example, the first energy time-series brainwave and the second energy time-series brainwave may be cascaded to form a first full-cycle time-series brainwave, that is, the first full-cycle time-series brainwave may be {the first energy time-series brainwave, the first energy time-series brainwave}; Performing a feature comparison process of a first energy on the interval truncated time-series electroencephalogram associated with the second reference electroencephalogram information to form a third energy time-series electroencephalogram associated with the second reference electroencephalogram information. For example, the interval truncated time-series electroencephalogram associated with the second reference electroencephalogram information may be subjected to a feature comparison process in a forward-to-backward order to form a third energy time-series electroencephalogram associated with the second reference electroencephalogram information. Performing a feature comparison process of a second energy on the interval truncated time-series electroencephalogram waves associated with the second reference electroencephalogram information to form a fourth energy time-series electroencephalogram waves associated with the second reference electroencephalogram information. For example, the interval truncated time-series electroencephalogram waves associated with the second reference electroencephalogram information may be subjected to a feature comparison process in a back-to-front order to form a fourth energy time-series electroencephalogram waves associated with the second reference electroencephalogram information. The third energy timing brain wave and the fourth energy timing brain wave are aggregated to form a second full-cycle timing brain wave including full-cycle related information of the second reference electroencephalogram information. For example, the third energy timing brain wave and the fourth energy timing brain wave can be cascaded to form a second full-cycle timing brain wave, that is, the second full-cycle timing brain wave can be {the third energy timing brain wave, the fourth energy timing brain wave}.

[0037] It can be understood that, in some possible implementations, the step of performing a first energy feature comparison process on the interval truncated time-series EEG waves associated with the first reference EEG information to form a first energy time-series EEG waves associated with the first reference EEG information may further include the specific implementation contents described in detail below: According to the associated waveform module (such as waveform sequence or waveform sequence row, which can be configured according to current needs), the interval truncated time-series brain waves associated with the first reference electroencephalogram information are split and sorted to form an interval truncated time-series brain wave sequence of associated intervals, wherein the interval truncated time-series brain wave sequence of the interval is composed of the interval truncated time-series brain waves of multiple intervals; According to the order of the interval truncated time-series brain waves of each interval in the interval truncated time-series brain wave sequence of the interval, the feature comparison processing is performed on the interval truncated time-series brain waves of each interval in turn to form the interval time-series brain waves associated with the interval truncated time-series brain waves of the interval; The interval time-series brain waves associated with the interval truncated time-series brain waves of each of the intervals are sorted to form a first energy time-series brain wave associated with the first reference electroencephalogram information.

[0038] Among them, it can be understood that, in some possible implementations, the step of performing feature comparison processing on the interval truncated time-series brain waves of each interval in turn according to the order of the interval truncated time-series brain waves of each interval in the interval truncated time-series brain wave sequence of the interval to form the interval time-series brain waves associated with the interval truncated time-series brain waves of the interval may further include the specific implementation content described in detail below: For the interval truncated time-series brain wave of the first interval in the interval truncated time-series brain wave sequence of the interval, the interval truncated time-series brain wave of the interval is directly used as the interval time-series brain wave associated with the interval truncated time-series brain wave of the interval; For the interval truncation timing brain waves of each other interval except the interval truncation timing brain waves of the first interval in the interval truncation timing brain wave sequence of the interval, based on the interval timing brain waves associated with the interval truncation timing brain waves of the previous interval of the interval truncation timing brain waves of the other interval, focusing feature analysis processing is performed on the interval truncation timing brain waves of the other intervals to form interval focusing timing brain waves associated with the interval truncation timing brain waves of the other intervals, and cascade pair processing is performed on the interval truncation timing brain waves of the other intervals and the interval focusing timing brain waves to form interval timing brain waves associated with the interval truncation timing brain waves of the other intervals.

[0039] It can be understood that, in some possible implementations, the step of performing aggregating processing on the first energy time-series brain wave and the second energy time-series brain wave to form a first full-cycle time-series brain wave including full-cycle related information of the first reference electroencephalogram information may further include specific implementation contents described in detail below: Extracting the last interval time-series brain wave (associated with the last waveform module, such as the last waveform sequence or waveform sequence row, etc.) from the first energy time-series brain wave, wherein the last interval time-series brain wave includes information of each previous interval time-series brain wave; Extracting the first interval time-series brain wave (associated with the first waveform module, such as the first waveform sequence or waveform sequence row, etc.) from the second energy time-series brain wave, wherein the first interval time-series brain wave includes information of each subsequent interval time-series brain wave; The last interval time-series brain wave in the first energy time-series brain wave and the first interval time-series brain wave in the second energy time-series brain wave are cascaded to form a first full-cycle time-series brain wave including full-cycle related information of the first reference electroencephalogram information.

[0040] Among them, it can be understood that, in some possible implementations, the step of performing model optimization processing on the EEG similarity detection model so that the model optimization cost index is less than or equal to the pre-configured reference model optimization cost index to obtain the target similarity detection model associated with the EEG similarity detection model further includes the specific implementation content described in detail below: When the model optimization cost indicator is greater than a pre-configured reference model optimization cost indicator, analyzing the associated model optimization indicator according to the model optimization cost indicator; In the electroencephalogram similarity detection model, reverse transfer processing is performed on the model optimization index, and during the transfer process, optimization adjustment is performed on the model index included in the electroencephalogram similarity detection model so that the model optimization cost index is less than or equal to the pre-configured reference model optimization cost index, thereby forming a target similarity detection model associated with the electroencephalogram similarity detection model.

[0041] It can be understood that, in some possible implementations, step S130 in the above description, i.e., the step of performing EEG similarity detection processing on the first EEG information and the second EEG information according to the target similarity detection model to output a similarity calculation result between the first EEG information and the second EEG information in the EEG information pair, may further include the specific implementation content described in detail below: According to the target similarity detection model, the first electroencephalogram information and the second electroencephalogram information are sequentially subjected to feature interval truncation processing to output interval truncated time-series brain waves associated with the first electroencephalogram information and interval truncated time-series brain waves associated with the second electroencephalogram information. That is, the first electroencephalogram information and the second electroencephalogram information can be sequentially mapped to the feature space to perform representation in the form of vectors, such as performing embedding processing, to obtain interval truncated time-series brain waves associated with the first electroencephalogram information and interval truncated time-series brain waves associated with the second electroencephalogram information; Performing feature comparison processing on the interval truncated time-series EEG waves associated with the first EEG information and the interval truncated time-series EEG waves associated with the second EEG information to form a first full-cycle time-series EEG wave including full-cycle related information of the first EEG information, and forming a second full-cycle time-series EEG wave including full-cycle related information of the second EEG information, as described above; Performing cascade processing on the first complete cycle time-series brain wave and the second complete cycle time-series brain wave to form associated cascaded complete cycle time-series brain waves. For example, the cascaded complete cycle time-series brain wave may be {the first complete cycle time-series brain wave, the second complete cycle time-series brain wave}; The similarity calculation results of the cascaded whole cycle time-series EEG waves are detected and processed to output the similarity calculation results between the first EEG information and the second EEG information in the EEG information pair. For example, the cascaded whole cycle time-series EEG waves can be fully connected first to form an associated fully connected vector, and then the fully connected vector can be processed by a classification function to obtain an associated similarity calculation result.

[0042] pass Figure 2 The embodiment of the present invention further provides an EEG-based emotion recognition device, which can be applied to the above-mentioned EEG-based emotion recognition system. The EEG-based emotion recognition device may include: Model optimization module, used to detect the model through EEG information, Figure 1 The consistency detection model performs model optimization processing to obtain the EEG Figure 1 The target consistency detection model associated with the consistency detection model, the EEG information detection model and the EEG Figure 1 All consistency detection models are deep learning models; an electroencephalogram information extraction module, used to obtain first electroencephalogram information and second electroencephalogram information in an electroencephalogram information pair to be identified, wherein the first electroencephalogram information and the second electroencephalogram information are used to characterize the potential emotions of the corresponding person to be identified, and the first electroencephalogram information and the second electroencephalogram information belong to a brain wave time series feature sequence; An electroencephalogram similarity detection module is used to perform electroencephalogram similarity detection processing on the first electroencephalogram information and the second electroencephalogram information according to a target similarity detection model to output a similarity calculation result between the first electroencephalogram information and the second electroencephalogram information in the electroencephalogram information pair, wherein the similarity calculation result is used to characterize whether the first electroencephalogram information and the second electroencephalogram information meet emotional consistency or whether the person to be identified associated with the first electroencephalogram information and the person to be identified associated with the second electroencephalogram information meet emotional consistency.

[0043] In summary, the present invention provides an EEG-based emotion recognition method and system, which can first detect the EEG information model and then Figure 1 The consistency detection model performs model optimization processing to obtain EEG Figure 1 The target consistency detection model is associated with the consistency detection model; the first EEG information and the second EEG information in the EEG information pair to be identified are obtained; according to the target similarity detection model, the first EEG information and the second EEG information are subjected to EEG similarity detection processing to output the similarity calculation result between the first EEG information and the second EEG information in the EEG information pair. Based on the above content, since in the EEG Figure 1 During the process of model optimization, the EEG information detection model will be used to detect the target consistency detection model, so that the accuracy of the target consistency detection model formed can be higher. Therefore, the reliability of data matching can be improved to a certain extent.

[0044] According to the third aspect of the present invention, the present invention seeks protection for an electroencephalogram-based emotion recognition device, comprising a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to implement the electroencephalogram-based emotion recognition method.

[0045] According to a fourth aspect of the present invention, the present invention seeks protection for a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the electroencephalogram-based emotion recognition method.

[0046] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An electroencephalogram-based emotion recognition method, characterized in that: include: Performing model optimization processing on the EEG consistency detection model through the EEG information detection model to obtain a target consistency detection model associated with the EEG consistency detection model, wherein both the EEG information detection model and the EEG consistency detection model are deep learning models; Obtaining first electroencephalogram information and second electroencephalogram information in a pair of electroencephalogram information to be identified, wherein the first electroencephalogram information and the second electroencephalogram information are used to characterize potential emotions of the corresponding person to be identified, and the first electroencephalogram information and the second electroencephalogram information belong to a brain wave time series feature sequence; According to the target similarity detection model, the first EEG information and the second EEG information are subjected to EEG similarity detection processing to output a similarity calculation result between the first EEG information and the second EEG information in the EEG information pair, wherein the similarity calculation result is used to characterize whether the first EEG information and the second EEG information satisfy emotional consistency or whether the person to be identified associated with the first EEG information and the person to be identified associated with the second EEG information satisfy emotional consistency.

2. The method for emotion recognition based on electroencephalogram according to claim 1, characterized in that: The step of performing electroencephalogram similarity detection processing on the first electroencephalogram information and the second electroencephalogram information according to the target similarity detection model to output a similarity calculation result between the first electroencephalogram information and the second electroencephalogram information in the electroencephalogram information pair includes: According to the target similarity detection model, the first electroencephalogram information and the second electroencephalogram information are sequentially subjected to feature interval truncation processing to output interval truncation time-series brain waves associated with the first electroencephalogram information and interval truncation time-series brain waves associated with the second electroencephalogram information; Performing feature comparison processing on the interval truncated time-series EEG waves associated with the first EEG information and the interval truncated time-series EEG waves associated with the second EEG information to form a first full-cycle time-series EEG wave including full-cycle related information of the first EEG information, and forming a second full-cycle time-series EEG wave including full-cycle related information of the second EEG information; Performing cascade processing on the first complete cycle time-series brain wave and the second complete cycle time-series brain wave to form associated cascade complete cycle time-series brain wave; The cascaded full-cycle time-series electroencephalogram waves are subjected to detection processing of similarity calculation results to output similarity calculation results between the first electroencephalogram information and the second electroencephalogram information in the electroencephalogram information pair.

3. The method for emotion recognition based on electroencephalogram according to claim 1, characterized in that: The step of performing model optimization processing on the EEG consistency detection model through the EEG information detection model to obtain a target consistency detection model associated with the EEG consistency detection model includes: According to the electroencephalogram information detection model, electroencephalogram information detection processing is performed on the first reference electroencephalogram information in the reference electroencephalogram information pair to output an electroencephalogram detection result associated with the first reference electroencephalogram information, the first reference electroencephalogram information belongs to one of the reference electroencephalogram information in the reference electroencephalogram information pair, and the first reference electroencephalogram information is used to characterize the potential emotion of the corresponding reference electroencephalogram, the first reference electroencephalogram information belongs to the brain wave time series feature sequence, and the electroencephalogram detection result is used to characterize whether the first reference electroencephalogram information and the second reference electroencephalogram information in the reference electroencephalogram information pair meet emotion consistency or characterize whether the reference electroencephalogram associated with the first reference electroencephalogram information and the reference electroencephalogram associated with the second reference electroencephalogram information meet emotion consistency; According to the EEG extreme value oscillation scene of the reference EEG information pair sequence, performing risk analysis processing on the EEG detection result associated with the first reference EEG information to output a risk characterization index associated with the reference EEG information pair; According to the EEG similarity detection model, performing EEG similarity detection processing on the first reference EEG information and the second reference EEG information in the reference EEG information pair to output an EEG similarity detection result associated with the reference EEG information pair; If the electroencephalogram similarity detection result associated with the reference electroencephalogram information pair is the existence ratio of each current similarity data of the reference electroencephalogram information pair, a first parsing process is performed on the low similarity feature of the existence ratio of each current similarity data of the reference electroencephalogram information pair to form an associated target parsing output indicator; Performing a second parsing process on the target parsing output indicator and the risk characterization indicator associated with the reference EEG information pair to form a model optimization cost indicator associated with the EEG similarity detection model; A model optimization process is performed on the electroencephalogram similarity detection model so that the model optimization cost index is less than or equal to a pre-configured reference model optimization cost index, so as to obtain a target similarity detection model associated with the electroencephalogram similarity detection model.

4. The method for emotion recognition based on electroencephalogram according to claim 3, characterized in that: Before the step of performing electroencephalogram information detection processing on the first reference electroencephalogram information in the reference electroencephalogram information pair according to the electroencephalogram information detection model to output the electroencephalogram detection result associated with the first reference electroencephalogram information, the step of performing model optimization processing on the electroencephalogram consistency detection model through the electroencephalogram information detection model to obtain the target consistency detection model associated with the electroencephalogram consistency detection model further includes: performing model optimization processing on a candidate electroencephalogram information detection model to form an electroencephalogram information detection model after model optimization based on current similarity data of third reference electroencephalogram information and fourth reference electroencephalogram information in a reference electroencephalogram information pair sequence and the third reference electroencephalogram information, wherein the third reference electroencephalogram information and the fourth reference electroencephalogram information belong to a reference electroencephalogram information pair in the reference electroencephalogram information pair sequence; The step of performing electroencephalogram information detection processing on the first reference electroencephalogram information in the reference electroencephalogram information pair according to the electroencephalogram information detection model to output an electroencephalogram detection result associated with the first reference electroencephalogram information comprises: According to the electroencephalogram information detection model optimized by the model, electroencephalogram information detection processing is performed on the first reference electroencephalogram information to output an electroencephalogram detection result associated with the first reference electroencephalogram information.

5. The method for emotion recognition based on electroencephalogram according to claim 4, characterized in that: The step of performing model optimization processing on the candidate EEG information detection model based on the current similarity data of the third reference EEG information and the fourth reference EEG information in the sequence and the third reference EEG information to form an optimized EEG information detection model comprises: Performing classification processing on the reference electroencephalogram information pair sequence to form a first number of reference electroencephalogram information pair subsequences associated with the reference electroencephalogram information pair sequence; For any one of the reference EEG information pair subsequences to be processed in the first number of reference EEG information pair subsequences, perform the following processing: Marking reference electroencephalogram information pair subsequences other than the reference electroencephalogram information pair subsequence to be processed to mark them as reference electroencephalogram information pair subsequences to be processed, and performing model optimization processing on candidate electroencephalogram information detection models based on the reference electroencephalogram information pair subsequence to be processed to form an electroencephalogram information detection model after model optimization; The step of performing electroencephalogram information detection processing on the first reference electroencephalogram information based on the electroencephalogram information detection model optimized by the model to output an electroencephalogram detection result associated with the first reference electroencephalogram information comprises: Based on the reference EEG information pair subsequence to be processed and the EEG information detection model optimized by the model, EEG information detection processing is performed on the reference EEG information pair subsequence to be processed to output the EEG detection result associated with the reference EEG information pair subsequence to be processed.

6. The method for emotion recognition based on electroencephalogram according to claim 4, characterized in that: The step of performing electroencephalogram information detection processing on the first reference electroencephalogram information based on the electroencephalogram information detection model optimized by the model to output an electroencephalogram detection result associated with the first reference electroencephalogram information comprises: According to the electroencephalogram information detection model optimized by the model, the first reference electroencephalogram information is subjected to characteristic interval truncation processing to output interval truncation time-series brain waves associated with the first reference electroencephalogram information; Perform feature comparison processing on the interval truncated time-series electroencephalogram associated with the first reference electroencephalogram information to form a whole-cycle time-series electroencephalogram associated with the first reference electroencephalogram information including whole-cycle related information; The whole cycle time-series electroencephalogram wave including the whole cycle related information is subjected to electroencephalogram information detection processing to output an electroencephalogram detection result associated with the first reference electroencephalogram information.

7. The method for emotion recognition based on electroencephalogram according to claim 3, characterized in that: The electroencephalogram detection result associated with the first reference electroencephalogram information is the ratio of each current similarity data of the first reference electroencephalogram information; The step of performing risk analysis processing on the electroencephalogram detection result associated with the first reference electroencephalogram information according to the electroencephalogram extreme value oscillation scene of the reference electroencephalogram information pair sequence to output the risk characterization index associated with the reference electroencephalogram information pair comprises: Analyzing the emotional fluctuation ratio of each current similarity data of the first reference electroencephalogram information and the second reference electroencephalogram information in the reference electroencephalogram information pair when the reference electroencephalogram information pair sequence is in an electroencephalogram extreme oscillation scene; Based on the ratio of each current similarity data of the first reference EEG information and the emotional fluctuation ratio of each current similarity data, the EEG detection result of the first reference EEG information is subjected to risk analysis processing to output a risk characterization index associated with the reference EEG information.

8. An electroencephalogram-based emotion recognition device, characterized in that: include: A model optimization module, used for performing model optimization processing on the EEG consistency detection model through the EEG information detection model to obtain a target consistency detection model associated with the EEG consistency detection model, wherein both the EEG information detection model and the EEG consistency detection model are deep learning models; an electroencephalogram information extraction module, used to obtain first electroencephalogram information and second electroencephalogram information in an electroencephalogram information pair to be identified, wherein the first electroencephalogram information and the second electroencephalogram information are used to characterize the potential emotions of the corresponding person to be identified, and the first electroencephalogram information and the second electroencephalogram information belong to a brain wave time series feature sequence; An electroencephalogram similarity detection module is used to perform electroencephalogram similarity detection processing on the first electroencephalogram information and the second electroencephalogram information according to a target similarity detection model to output a similarity calculation result between the first electroencephalogram information and the second electroencephalogram information in the electroencephalogram information pair, wherein the similarity calculation result is used to characterize whether the first electroencephalogram information and the second electroencephalogram information meet emotional consistency or whether the person to be identified associated with the first electroencephalogram information and the person to be identified associated with the second electroencephalogram information meet emotional consistency.

9. An electroencephalogram-based emotion recognition device, characterized in that: It comprises a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to implement an electroencephalogram-based emotion recognition method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for emotion recognition based on electroencephalogram according to any one of claims 1 to 7 is implemented.

Images