CN117708682B - Intelligent brain wave acquisition and analysis system and method - Google Patents
Intelligent brain wave acquisition and analysis system and method Download PDFInfo
- Publication number
- CN117708682B CN117708682B CN202410169443.7A CN202410169443A CN117708682B CN 117708682 B CN117708682 B CN 117708682B CN 202410169443 A CN202410169443 A CN 202410169443A CN 117708682 B CN117708682 B CN 117708682B
- Authority
- CN
- China
- Prior art keywords
- brain wave
- waveform
- sequence
- local neighborhood
- feature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 210000004556 brain Anatomy 0.000 title claims abstract description 427
- 238000004458 analytical method Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 41
- 239000013598 vector Substances 0.000 claims abstract description 202
- 230000008451 emotion Effects 0.000 claims abstract description 50
- 230000003993 interaction Effects 0.000 claims abstract description 48
- 238000012549 training Methods 0.000 claims description 85
- 238000013135 deep learning Methods 0.000 claims description 25
- 230000006870 function Effects 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 8
- 230000002996 emotional effect Effects 0.000 abstract description 18
- 238000005516 engineering process Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 230000000694 effects Effects 0.000 description 8
- 230000008909 emotion recognition Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000003920 cognitive function Effects 0.000 description 4
- 210000004761 scalp Anatomy 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 3
- 230000007177 brain activity Effects 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000013136 deep learning model Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000000537 electroencephalography Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000004630 mental health Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 108010076504 Protein Sorting Signals Proteins 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000006722 reduction reaction Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000003860 sleep quality Effects 0.000 description 1
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The application discloses an intelligent brain wave acquisition and analysis system and method, and relates to the technical field of intelligent brain waves, wherein the intelligent brain wave acquisition and analysis system and method are used for acquiring brain wave signals of a tested person; extracting local neighborhood waveform characteristics of the brain wave signals to obtain a sequence of brain wave local neighborhood waveform association characteristic vectors; the sequence of the brain wave local neighborhood waveform association feature vector is passed through a global feature interaction module based on internal self-attention to obtain a sequence of global enhanced brain wave waveform association feature vector; and determining the emotion label of the tested person based on the sequence of the global enhanced brain wave waveform associated feature vector. Therefore, more accurate description and recognition of the emotional states of the testee can be realized, and more reliable technical support is provided for classification and recognition of the emotional states.
Description
Technical Field
The application relates to the technical field of intelligent brain waves, in particular to an intelligent brain wave acquisition and analysis system and method.
Background
Brain wave signals are important physiological signals reflecting human brain activities and have rich information content. Through acquisition and analysis of brain wave signals, the emotion state or cognitive function of a tested person can be known, and valuable references are provided for the fields of mental health, education, entertainment and the like.
However, brain wave signals have the characteristics of high dimensionality, high complexity, high noise, high nonlinearity and the like, and great challenges are brought to classification and identification of brain wave signals. Traditional brain wave signal classification and identification methods are generally based on manually extracted characteristic parameters, such as power spectrum, coherence, entropy and the like, which often cannot fully express the internal rules of brain wave signals, so that the classification and identification effects are not ideal.
Therefore, an optimized brain wave intelligent acquisition and analysis system and method are expected.
Disclosure of Invention
The application provides an intelligent brain wave acquisition and analysis system and method, which acquire brain wave signals of a tested person; extracting local neighborhood waveform characteristics of the brain wave signals to obtain a sequence of brain wave local neighborhood waveform association characteristic vectors; the sequence of the brain wave local neighborhood waveform association feature vector is passed through a global feature interaction module based on internal self-attention to obtain a sequence of global enhanced brain wave waveform association feature vector; and determining the emotion label of the tested person based on the sequence of the global enhanced brain wave waveform associated feature vector. Therefore, more accurate description and recognition of the emotional states of the testee can be realized, and more reliable technical support is provided for classification and recognition of the emotional states.
The application also provides an intelligent brain wave acquisition and analysis method, which comprises the following steps:
Acquiring brain wave signals of a tested person;
Extracting local neighborhood waveform characteristics of the brain wave signals to obtain a sequence of brain wave local neighborhood waveform association characteristic vectors;
The sequence of the brain wave local neighborhood waveform association feature vector is passed through a global feature interaction module based on internal self-attention to obtain a sequence of global enhanced brain wave waveform association feature vector;
determining an emotion tag of the tested person based on the sequence of the global enhanced brain wave waveform associated feature vector;
The sequence of the brain wave local neighborhood waveform association feature vector is passed through a global feature interaction module based on internal self-attention to obtain a sequence of global enhanced brain wave waveform association feature vector, which comprises the following steps: processing the sequence of brain wave local neighborhood waveform associated feature vectors by using the following global feature interaction formula to obtain the sequence of global enhanced brain wave waveform associated feature vectors; the global feature interaction formula is as follows:
；
wherein, Associating a first/>, in the sequence of feature vectors, for the brain wave local neighborhood waveformLocal neighborhood waveform of individual brain waves associates feature vectors with the/>Correlation matrix between local neighborhood waveform correlation feature vectors of individual brain waves,/>Associating a first/>, in the sequence of feature vectors, for the brain wave local neighborhood waveformLocal neighborhood waveform of individual brain waves associates feature vectors with the/>Correlation matrix between local neighborhood waveform correlation feature vectors of individual brain waves,/>For the number of the brain wave local neighborhood waveform associated feature vectors,/>The value of (2) is/>，/>For/>Local neighborhood waveform associated feature vector of brain wave,/>A/>, in the sequence of the global enhanced brain wave waveform associated feature vectorsGlobal enhanced brain wave waveform associated feature vector,/>Representing an exponential function operation.
In the above brain wave intelligent acquisition and analysis method, extracting the local neighborhood waveform characteristics of the brain wave signals to obtain a sequence of brain wave local neighborhood waveform association characteristic vectors comprises: segmenting the brain wave signals based on a preset scale to obtain a sequence of brain wave local signals; and extracting local neighborhood waveform characteristics of the sequence of the brain wave local signals by using a deep learning network model to obtain a sequence of brain wave local neighborhood waveform association characteristic vectors.
In the brain wave intelligent acquisition and analysis method, the deep learning network model is a brain wave local neighborhood waveform correlation feature extractor based on a convolution layer.
In the above brain wave intelligent acquisition and analysis method, the performing local neighborhood waveform feature extraction on the sequence of brain wave local signals by using a deep learning network model to obtain the sequence of brain wave local neighborhood waveform association feature vectors comprises: and passing the sequence of the brain wave local signals through the brain wave local neighborhood waveform association feature extractor based on the convolution layer to obtain the sequence of the brain wave local neighborhood waveform association feature vector.
In the above brain wave intelligent acquisition and analysis method, determining the emotion tag of the tested person based on the sequence of the global enhanced brain wave waveform associated feature vector includes: cascading the sequence of the global enhanced brain wave waveform association feature vector into a global enhanced brain wave waveform association cascading feature vector; and the global enhanced brain wave waveform associated cascade feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for representing the emotion label of the tested person.
In the above brain wave intelligent acquisition and analysis method, the global enhanced brain wave waveform association cascade feature vector is passed through a classifier to obtain a classification result, where the classification result is used to represent an emotion tag of the tested person, and the method includes: performing full-connection coding on the global enhanced brain wave waveform association cascade feature vector by using a plurality of full-connection layers of the classifier to obtain a coding classification feature vector; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
The intelligent brain wave acquisition and analysis method further comprises the training steps of: and training the brain wave local neighborhood waveform associated feature extractor based on the convolution layer, the global feature interaction module based on the internal self-focusing and the classifier.
In the above brain wave intelligent acquisition and analysis method, the training step includes: acquiring training data, wherein the training data comprises training brain wave signals of a tested person and true values of emotion labels of the tested person; segmenting the training brain wave signals based on a preset scale to obtain a sequence of training brain wave local signals; passing the sequence of the training brain wave local signals through the brain wave local neighborhood waveform associated feature extractor based on the convolution layer to obtain a sequence of training brain wave local neighborhood waveform associated feature vectors; passing the sequence of the training brain wave local neighborhood waveform associated feature vectors through the global feature interaction module based on internal self-attention to obtain a sequence of training global enhanced brain wave waveform associated feature vectors; after cascading the sequence of the training global enhanced brain wave waveform association feature vector into a training global enhanced brain wave waveform association cascading feature vector, the training global enhanced brain wave waveform association cascading feature vector passes through a classifier to obtain a classification loss function value; and training the brain wave local neighborhood waveform associated feature extractor based on the convolution layer, the global feature interaction module based on the internal self-attention and the classifier by using the classification loss function value, wherein in each round of iteration of training, the training global enhanced brain wave waveform associated cascade feature vector is optimized.
The application also provides an intelligent brain wave acquisition and analysis system, which comprises:
the brain wave signal acquisition module is used for acquiring brain wave signals of the tested person;
The local neighborhood waveform feature extraction module is used for extracting local neighborhood waveform features of the brain wave signals to obtain a sequence of brain wave local neighborhood waveform association feature vectors;
the global feature interaction module is used for enabling the sequence of the brain wave local neighborhood waveform association feature vector to pass through the global feature interaction module based on internal self-attention so as to obtain the sequence of the global enhanced brain wave waveform association feature vector;
The emotion label determining module is used for determining emotion labels of the testee based on the sequence of the global enhanced brain wave waveform associated feature vectors;
the global feature interaction module comprises:
Processing the sequence of brain wave local neighborhood waveform associated feature vectors by using the following global feature interaction formula to obtain the sequence of global enhanced brain wave waveform associated feature vectors; the global feature interaction formula is as follows:
；
wherein, Associating a first/>, in the sequence of feature vectors, for the brain wave local neighborhood waveformLocal neighborhood waveform of individual brain waves associates feature vectors with the/>Correlation matrix between local neighborhood waveform correlation feature vectors of individual brain waves,/>Associating a first/>, in the sequence of feature vectors, for the brain wave local neighborhood waveformLocal neighborhood waveform of individual brain waves associates feature vectors with the/>Correlation matrix between local neighborhood waveform correlation feature vectors of individual brain waves,/>For the number of the brain wave local neighborhood waveform associated feature vectors,/>The value of (2) is/>，/>For/>Local neighborhood waveform associated feature vector of brain wave,/>A/>, in the sequence of the global enhanced brain wave waveform associated feature vectorsGlobal enhanced brain wave waveform associated feature vector,/>Representing an exponential function operation.
Compared with the prior art, the brain wave intelligent acquisition and analysis system and the brain wave intelligent acquisition and analysis method can fully utilize the rich information of brain wave signals, and process and analyze the brain wave signals by combining with artificial intelligence technology based on deep learning, so that the intrinsic rules and waveform characteristic modes of the brain wave signals are described and depicted, and the classification and identification of the emotional states of a tested person are realized; the method can also realize more accurate description and recognition of the emotional states of the testee, and provide more reliable technical support for classification and recognition of the emotional states.
Drawings
In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
fig. 1 is a flowchart of an intelligent brain wave acquisition and analysis method according to an embodiment of the present application.
Fig. 2 is a schematic diagram of a system architecture of an intelligent brain wave acquisition and analysis method according to an embodiment of the present application.
Fig. 3 is a block diagram of an intelligent brain wave acquisition and analysis system according to an embodiment of the present application.
Fig. 4 is an application scenario diagram of an intelligent brain wave acquisition and analysis method provided in an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the embodiments of the present application will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present application and their descriptions herein are for the purpose of explaining the present application, but are not to be construed as limiting the application.
Unless defined otherwise, all technical and scientific terms used in the embodiments of the application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In describing embodiments of the present application, unless otherwise indicated and limited thereto, the term "connected" should be construed broadly, for example, it may be an electrical connection, or may be a communication between two elements, or may be a direct connection, or may be an indirect connection via an intermediate medium, and it will be understood by those skilled in the art that the specific meaning of the term may be interpreted according to circumstances.
It should be noted that, the term "first\second\third" related to the embodiment of the present application merely distinguishes similar objects, and does not represent a specific order for the objects, and it is understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that embodiments of the application described herein may be practiced in sequences other than those illustrated or described herein.
Brain waves (EEG), which are a physiological signal that records brain electrical activity, are the distributed potentials on the scalp of weak currents produced by the electrical activity of brain neurons, which can be detected and recorded by electrodes on the scalp. Brain wave signals can be divided into different frequency bands, including delta wave (0.5-4 Hz), theta wave (4-8 Hz), alpha wave (8-13 Hz), beta wave (13-30 Hz) and gamma wave (30-100 Hz), and brain waves of different frequency bands are associated with different states and functions of the brain.
The acquisition of brain wave signals is typically performed by means of electroencephalographic instruments comprising a plurality of electrodes, typically placed on the scalp of the subject, and the electroencephalographic records the changes in the electrical potentials captured by these electrodes, which reflect the electrical activity of the different areas of the brain. Analysis of brain wave signals may provide information about brain activity of the subject. For example, by analyzing brain waves in a specific frequency band, the attention level, the emotional state, the sleep quality, and the like of the subject can be known. In addition, brain wave signals can also be used in the fields of research of cognitive functions, nerve feedback training, brain-computer interfaces and the like.
Brain wave signals are used in a wide variety of applications including, but not limited to, medical diagnostics, mental health assessment, cognitive neuroscience research, brain-computer interface technology, sleep research, education, and entertainment. With the development of technology, the acquisition and analysis of brain wave signals will continue to provide important information and application value for human health and scientific research.
Traditional brain wave signal classification and identification methods are generally based on manually extracted characteristic parameters, such as power spectrum, coherence, entropy and the like, which often cannot fully express the internal rules of brain wave signals, so that the classification and identification effects are not ideal. Moreover, conventional brain wave signal classification and identification methods generally require manual selection and extraction of feature parameters, which may lead to subjective deviations in understanding and analyzing brain wave signals, while also limiting the diversity and richness of features. The characteristic parameters extracted by the traditional brain wave signal classification and identification method are generally based on the result of frequency domain or time domain analysis, and may not fully reflect the high dimensionality, high complexity and high nonlinearity of the brain wave signal, thereby limiting the performance of classification and identification models. The classification and recognition model established by the traditional brain wave signal classification and recognition method is sensitive to noise and data change, has limited generalization capability, and is difficult to adapt to brain wave signals of different individuals or different environments.
In recent years, with the development of technologies such as machine learning, deep learning and neural networks, people begin to try to classify and identify brain wave signals by using a method based on deep learning, so as to better mine potential rules and features in the brain wave signals and improve the accuracy and robustness of classification and identification.
The application provides an intelligent brain wave acquisition and analysis system and method, wherein the system comprises the following steps: the brain wave acquisition module is used for acquiring brain wave signals of the tested person; the signal processing module is used for filtering, reducing noise, amplifying and converting the acquired brain wave signals; the characteristic extraction module is used for extracting characteristic parameters from the processed brain wave signals; the classification and identification module is used for classifying and identifying brain wave signals according to the characteristic parameters and judging the emotion state or cognitive function of the tested person; and the output module is used for outputting the classification recognition result to the display device or the storage device. The method comprises the following steps: (1) Collecting brain wave signals of a tested person through a brain wave collecting module; (2) The method comprises the steps of filtering, noise reduction, amplification and conversion are carried out on acquired brain wave signals through a signal processing module, and processed brain wave signals are obtained; (3) Extracting characteristic parameters from the processed brain wave signals through a characteristic extraction module; (4) Classifying and identifying brain wave signals according to the characteristic parameters by a classification and identification module, and judging the emotion state or cognitive function of the tested person; (5) And outputting the classification recognition result to a display device or a storage device through an output module. The system and the method can realize intelligent acquisition and analysis of brain wave signals, improve the utilization rate and accuracy of brain wave signals, and provide an effective tool for brain science research and application.
In one embodiment of the present application, fig. 1 is a flowchart of an intelligent brain wave acquisition and analysis method provided in the embodiment of the present application. Fig. 2 is a schematic diagram of a system architecture of an intelligent brain wave acquisition and analysis method according to an embodiment of the present application. As shown in fig. 1 and 2, the brain wave intelligent acquisition and analysis method according to the embodiment of the application includes: 110, acquiring brain wave signals of a tested person; 120, extracting local neighborhood waveform characteristics of the brain wave signals to obtain a sequence of brain wave local neighborhood waveform association characteristic vectors; 130, passing the sequence of brain wave local neighborhood waveform association feature vectors through a global feature interaction module based on internal self-attention to obtain a sequence of global enhanced brain wave waveform association feature vectors; 140, determining the emotion label of the tested person based on the sequence of the global enhanced brain wave waveform associated feature vector.
In the step 110, brain wave signals of the subject are acquired, and a proper electroencephalogram instrument is ensured to be used and a standard brain wave acquisition program is followed so as to obtain high-quality brain wave signals. By obtaining accurate and reliable brain wave signals as input for subsequent processing, basic data is provided for determining emotion tags. In the step 120, the local neighborhood waveform feature of the brain wave signal is extracted to obtain a sequence of brain wave local neighborhood waveform association feature vectors, and a suitable method is selected to extract the local neighborhood waveform feature, which may involve filtering, time-frequency analysis, and other techniques in signal processing. By obtaining the local characteristics of brain wave signals, the method is beneficial to capturing the activity modes of local brain areas and provides input for subsequent global characteristic interaction. In the step 130, the sequence of brain wave local neighborhood waveform association feature vectors is passed through a global feature interaction module based on internal self-attention to obtain a sequence of global enhanced brain wave waveform association feature vectors, and an appropriate global feature interaction module is designed, which may involve an attention mechanism or other global feature integration method. Through global feature interaction, the features of each local neighborhood are integrated, the associated features of global brain wave waveforms can be captured better, and the expression capability of the features is improved. In the step 140, based on the sequence of the global enhanced brain wave waveform associated feature vector, determining an emotion tag of the testee, and selecting an appropriate emotion tag determination method, it may be necessary to establish a mapping model between the emotion tag and brain wave features. The emotion state of the tested person can be inferred more accurately by utilizing the globally enhanced brain wave waveform associated feature vector, and a more reliable basis is provided for emotion recognition.
Aiming at the technical problems, the technical concept of the application is to fully utilize the rich information of the brain wave signals, and process and analyze the brain wave signals by combining with the artificial intelligence technology based on deep learning, thereby describing and describing the inherent rules and waveform characteristic modes of the brain wave signals and realizing classification and identification of the emotional states of the testees.
The deep learning technology can learn complex nonlinear characteristics and rules from large-scale data, is beneficial to mining potential rules and characteristic modes hidden in brain wave signals, and improves understanding and describing ability of the brain wave signals. The deep learning model can automatically learn and extract abstract features in brain wave signals, and avoids the limitation that a feature extractor needs to be manually selected and designed in the traditional method, so that the inherent features of the brain wave signals are more comprehensively and efficiently described. The brain wave signal processing and analyzing method based on deep learning can better capture complex features in brain wave signals, so that classification and recognition accuracy of emotional states are improved, and more reliable technical support is provided for realizing accurate depiction of the emotional states of a tested person. The deep learning model can better adapt to data of different individuals and different environments when processing brain wave signals, improves generalization capability of the model, and enables classification and recognition of emotional states to be more universal and reliable. By combining with the deep learning technology, the brain wave signal can be monitored and analyzed in real time, thereby providing possibility for real-time emotion recognition and feedback and being beneficial to being applied to the fields of clinic, psychology, man-machine interaction and the like.
Based on this, in the technical scheme of the application, firstly, brain wave signals of a tested person are obtained. Here, brain wave signals are important physiological signals reflecting human brain activities, and have a rich information content. Specifically, acquisition of brain wave signals is generally performed by an electroencephalogram technique. Electroencephalography technology records and measures electrical activity of brain wave signals through a plurality of electrodes placed on the scalp of a subject. The electrodes are connected to an amplifier or data acquisition system through lead wires, and brain wave signals are converted into digital signals for storage and analysis. Then, the brain wave signals are segmented based on a predetermined scale to obtain a sequence of brain wave local signals. Here, the purpose of slicing the brain wave signal based on a predetermined scale is to acquire a sequence of local signals of brain waves. It should be understood that, because the brain wave signal has the characteristics of high dimensionality, high complexity, high noise, high nonlinearity, etc., directly analyzing the whole signal may cause problems of information mixing and excessive computational complexity. Therefore, in the technical scheme of the application, the segmentation of the brain wave signals into the sequences of shorter local signals is helpful for extracting local features and reducing the computational complexity so as to better represent and analyze the brain wave signals.
In a specific embodiment of the present application, extracting the local neighborhood waveform feature of the brain wave signal to obtain a sequence of brain wave local neighborhood waveform association feature vectors includes: segmenting the brain wave signals based on a preset scale to obtain a sequence of brain wave local signals; and extracting local neighborhood waveform characteristics of the sequence of the brain wave local signals by using a deep learning network model to obtain a sequence of brain wave local neighborhood waveform association characteristic vectors.
The deep learning network model is a brain wave local neighborhood waveform association feature extractor based on a convolution layer.
It should be understood that by slicing the brain wave signals, local signal sequences in different time periods can be obtained, and processing by using the deep learning network model can better capture the time sequence features, including waveform changes, frequency changes, and the like, so as to more fully describe the dynamic characteristics of the brain wave signals. Based on the preset scale, the brain wave signals are segmented, the feature extraction requirements under different scales can be met, the deep learning network model can learn and extract local neighborhood waveform features under different scales, and support is provided for analysis under different scales.
The deep learning network model can automatically learn abstract feature representation in brain wave local signals, and avoids the complicated process of manually designing a feature extractor in the traditional method, so that the local neighborhood waveform features of the brain wave signals are more comprehensively and efficiently described. The deep learning network model can better capture nonlinear and complex features in brain wave local signals, so that the expression capacity of the features is improved, and the brain wave local neighborhood waveform associated feature vectors are more differentiated and representative. The deep learning network model can better adapt to data of different individuals and different environments when processing brain wave signals, improves generalization capability of the model, and enables classification and identification of emotional states to be more universal and reliable.
And then, passing the sequence of the brain wave local signals through a brain wave local neighborhood waveform correlation feature extractor based on a convolution layer to obtain a sequence of brain wave local neighborhood waveform correlation feature vectors. It should be understood that each brain wave local signal has a certain local correlation in the time domain, i.e. the brain wave signals at adjacent time points have a correlation. The brain wave local neighborhood waveform correlation characteristic extractor based on the convolution layer can effectively capture local neighborhood waveform correlation characteristics in brain wave signals, including waveform shape, amplitude variation, frequency characteristics and the like.
In a specific embodiment of the present application, the extracting the local neighborhood waveform feature of the sequence of brain wave local signals by using a deep learning network model to obtain the sequence of brain wave local neighborhood waveform association feature vectors includes: and passing the sequence of the brain wave local signals through the brain wave local neighborhood waveform association feature extractor based on the convolution layer to obtain the sequence of the brain wave local neighborhood waveform association feature vector.
Here, the brain wave local neighborhood waveform association feature vector reflects the correlation of brain wave signals within a local time window, but the global feature and long-range dependency of brain wave signals may not be sufficiently captured by only relying on local features. In the technical scheme of the application, in order to obtain more comprehensive characteristic representation, the sequence of the brain wave local neighborhood waveform association characteristic vector is extracted and integrated with the association relation of each local brain wave signal by a global characteristic interaction module based on internal self-attention, so that the expression capacity of waveform association characteristics is enhanced, and the sequence of global enhanced brain wave waveform association characteristic vector is obtained.
The global feature interaction module based on the internal self-attention introduces an internal interaction mechanism, so that each brain wave local neighborhood waveform association feature vector can interact with other brain wave local neighborhood waveform association feature vectors, and the association and the dependency relationship between the brain wave local neighborhood waveform association feature vectors are captured. In the internal self-focusing process, each brain wave local neighborhood waveform associated feature vector is given a weight to represent the importance or contribution of the brain wave local neighborhood waveform associated feature vector in the global feature. The interaction and weight distribution mechanism is helpful for extracting global features, and representing and describing global dynamic features and long-range correlations of brain wave signals, so that the waveform modes of the brain wave signals can be described more accurately.
In a specific embodiment of the present application, the sequence of brain wave local neighborhood waveform association feature vectors is passed through a global feature interaction module based on internal self-attention to obtain a sequence of global enhanced brain wave waveform association feature vectors, including: processing the sequence of brain wave local neighborhood waveform associated feature vectors by using the following global feature interaction formula to obtain the sequence of global enhanced brain wave waveform associated feature vectors; the global feature interaction formula is as follows:
；
wherein, Associating a first/>, in the sequence of feature vectors, for the brain wave local neighborhood waveformLocal neighborhood waveform of individual brain waves associates feature vectors with the/>Correlation matrix between local neighborhood waveform correlation feature vectors of individual brain waves,/>Associating a first/>, in the sequence of feature vectors, for the brain wave local neighborhood waveformLocal neighborhood waveform of individual brain waves associates feature vectors with the/>Correlation matrix between local neighborhood waveform correlation feature vectors of individual brain waves,/>For the number of the brain wave local neighborhood waveform associated feature vectors,/>The value of (2) is/>，/>For/>Local neighborhood waveform associated feature vector of brain wave,/>A/>, in the sequence of the global enhanced brain wave waveform associated feature vectorsGlobal enhanced brain wave waveform associated feature vector,/>Representing an exponential function operation.
In one embodiment of the present application, determining the emotion tag of the subject based on the sequence of the global enhanced brain wave waveform-associated feature vectors includes: cascading the sequence of the global enhanced brain wave waveform association feature vector into a global enhanced brain wave waveform association cascading feature vector; and the global enhanced brain wave waveform associated cascade feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for representing the emotion label of the tested person.
And then, after cascading the sequence of the global enhanced brain wave waveform association feature vector into a global enhanced brain wave waveform association cascading feature vector, passing the global enhanced brain wave waveform association cascading feature vector through a classifier to obtain a classification result, wherein the classification result is used for representing the emotion label of the tested person. The emotion label of the testee is a label for describing the current emotion state of the testee. In emotion recognition tasks, emotions may be divided into categories, such as happy, sad, anger, etc., each category corresponding to one emotional state. In the emotion recognition task, the classifier can learn the mapping from the global enhanced brain wave waveform associated cascade feature vector to the emotion label, so that the automatic classification and recognition of the emotion state of the tested person are realized. Thus providing valuable information for applications such as emotion recognition and emotion analysis.
In a specific embodiment of the present application, the global enhanced brain wave waveform association cascade feature vector is passed through a classifier to obtain a classification result, where the classification result is used to represent an emotion tag of the tested person, and the method includes: performing full-connection coding on the global enhanced brain wave waveform association cascade feature vector by using a plurality of full-connection layers of the classifier to obtain a coding classification feature vector; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
The global enhanced brain wave waveform association cascade feature vector integrates the local neighborhood waveform features, and can more comprehensively represent the integral features of brain wave signals, thereby being beneficial to improving the classification and recognition accuracy of emotional states. By using the classifier to classify the global enhanced brain wave waveform association cascade feature vector, brain wave signals can be more effectively associated with different emotion states, classification of the emotion states of a tested person is realized, and reliable support is provided for determining emotion labels.
The classification result is used for representing the emotion label of the testee, so that the identification and description of the personalized emotion state of the testee can be realized, and technical support is provided for personalized emotion identification. The classification result can be used for real-time monitoring and feedback, for example, in the fields of clinic, psychology, man-machine interaction and the like, and the real-time monitoring and feedback of the emotion state of the tested person can be realized, so that the possibility is provided for personalized emotion adjustment. The comprehensive representation of the global enhanced brain wave waveform associated cascade feature vector and the acquisition of the classification result are expected to improve the accuracy and stability of emotion recognition, and provide technical support for describing the emotion state of a tested person more accurately.
In one embodiment of the present application, the brain wave intelligent acquisition and analysis method further includes a training step: and training the brain wave local neighborhood waveform associated feature extractor based on the convolution layer, the global feature interaction module based on the internal self-focusing and the classifier. The training step comprises the following steps: acquiring training data, wherein the training data comprises training brain wave signals of a tested person and true values of emotion labels of the tested person; segmenting the training brain wave signals based on a preset scale to obtain a sequence of training brain wave local signals; passing the sequence of the training brain wave local signals through the brain wave local neighborhood waveform associated feature extractor based on the convolution layer to obtain a sequence of training brain wave local neighborhood waveform associated feature vectors; passing the sequence of the training brain wave local neighborhood waveform associated feature vectors through the global feature interaction module based on internal self-attention to obtain a sequence of training global enhanced brain wave waveform associated feature vectors; after cascading the sequence of the training global enhanced brain wave waveform association feature vector into a training global enhanced brain wave waveform association cascading feature vector, the training global enhanced brain wave waveform association cascading feature vector passes through a classifier to obtain a classification loss function value; and training the brain wave local neighborhood waveform associated feature extractor based on the convolution layer, the global feature interaction module based on the internal self-attention and the classifier by using the classification loss function value, wherein in each round of iteration of training, the training global enhanced brain wave waveform associated cascade feature vector is optimized.
It should be appreciated that by training the convolution layer based brain wave local neighborhood waveform correlation feature extractor, a more efficient local waveform feature representation may be learned, facilitating capturing local structural information in brain wave signals. The global feature interaction module based on the internal self-attention can learn the association and interaction between different parts in the brain wave signals in the training process, so that the complex relationship between the global features is better captured, and the expression capability of the features is improved. By training the classifier, parameters of the classifier can be optimized to better adapt to characteristics of brain wave signals, so that the classifier can more accurately correlate the extracted features with the emotional states, and accuracy of classification of the emotional states is improved.
In the training process, the data characteristics of different testees and different situations can be learned, the generalization capability of the model is improved, and the model can be better adapted to brain wave signals in different situations in practical application, so that the universality and the reliability of emotion state classification are improved. The trained models may be used for real-time applications, such as in the areas of emotion monitoring, psychological intervention, etc., to provide support for real-time emotion recognition.
In the technical scheme of the application, each training brain wave local neighborhood waveform associated feature vector of the sequence of training brain wave local neighborhood waveform associated feature vectors expresses the local time domain image semantic features of the training brain wave signals in the global time domain under the preset scale local time domain determined by segmentation based on the preset scale, so that the sequence of training brain wave local neighborhood waveform associated feature vectors can extract the local inter-time domain image semantic interactive features in the global time domain based on the local time domain image semantic features by an internal self-focusing global feature interaction module.
Therefore, considering that the sequence of the training global enhanced brain wave waveform association feature vector is cascaded into the training global enhanced brain wave waveform association cascade feature vector, the training global enhanced brain wave waveform association cascade feature vector has a local time domain-global time domain image semantic feature expression difference, and the feature distribution information saliency of each training brain wave local neighborhood waveform association feature vector based on the local time domain image semantic feature distribution is also influenced, so that the training global enhanced brain wave waveform association cascade feature vector is difficult to stably focus on the salient local distribution of the feature in the training process, and the expression effect of the training global enhanced brain wave waveform association cascade feature vector and the training speed of a model are influenced.
Based on the above, the applicant optimizes the training global enhanced brain wave waveform associated cascade feature vector each time the training global enhanced brain wave waveform associated cascade feature vector performs an iteration of classification regression through a classifier, expressed as: optimizing the training global enhanced brain wave waveform association cascade feature vector in each iteration of the training by using the following optimization formula to obtain an optimized training global enhanced brain wave waveform association cascade feature vector; wherein, the optimization formula is:
；
wherein, And/>The training global enhanced brain wave waveform is associated with cascade feature vectors/>, respectivelySquare of 1-norm and 2-norm,/>Is the training global enhanced brain wave waveform associated cascade feature vector/>And/>Is a weight superparameter,/>Is the characteristic value of the cascade characteristic vector associated with the training global enhanced brain wave waveform,/>Is the characteristic value of the cascade characteristic vector associated with the global enhanced brain wave waveform after optimization training,/>Is the training global enhanced brain wave waveform associated cascade feature vector,/>A logarithmic function with a base of 2 is shown.
Specifically, by associating cascade feature vectors based on the training global enhanced brain wave waveformsGeometric registration of the high-dimensional characteristic manifold shape is carried out by the scale and structural parameters of the (4), and the training global enhanced brain wave waveform correlation cascade characteristic vector/>Features with rich feature semantic information in feature sets formed by feature values, namely distinguishable stable interest features representing dissimilarity based on local context information when the classifier classifies, so that the training global enhanced brain wave waveform associated cascade feature vector/>And the feature information significance is marked in the classification process, so that the training speed of the classifier is improved.
In summary, the brain wave intelligent acquisition and analysis method based on the embodiment of the application is explained, fully utilizes the rich information of brain wave signals, and processes and analyzes the brain wave signals by combining with artificial intelligence technology based on deep learning, thereby describing and describing the inherent rules and waveform characteristic modes of the brain wave signals and realizing classification and identification of the emotional states of the testees.
Fig. 3 is a block diagram of an intelligent brain wave acquisition and analysis system according to an embodiment of the present application. As shown in fig. 3, the brain wave intelligent acquisition and analysis system 200 includes: the brain wave signal acquisition module 210 is configured to acquire brain wave signals of a subject; the local neighborhood waveform feature extraction module 220 is configured to extract local neighborhood waveform features of the brain wave signal to obtain a sequence of brain wave local neighborhood waveform association feature vectors; the global feature interaction module 230 is configured to obtain a sequence of global enhanced brain wave waveform associated feature vectors by using the sequence of brain wave local neighborhood waveform associated feature vectors through the global feature interaction module based on internal self-focusing; the emotion label determining module 240 of the tested person is configured to determine an emotion label of the tested person based on the sequence of the global enhanced brain wave waveform associated feature vector.
It will be appreciated by those skilled in the art that the specific operations of the respective steps in the above-described brain wave intelligent acquisition and analysis system have been described in detail in the above description of the brain wave intelligent acquisition and analysis method with reference to fig. 1to 2, and thus, repetitive descriptions thereof will be omitted.
As described above, the brain wave intelligent acquisition and analysis system 200 according to the embodiment of the present application may be implemented in various terminal devices, such as a server for brain wave intelligent acquisition and analysis, and the like. In one example, the brain wave intelligent acquisition and analysis system 200 according to an embodiment of the present application may be integrated into a terminal device as a software module and/or a hardware module. For example, the brain wave intelligent acquisition and analysis system 200 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the brain wave intelligent acquisition and analysis system 200 can also be one of a plurality of hardware modules of the terminal device.
Alternatively, in another example, the brain wave intelligent acquisition and analysis system 200 and the terminal device may be separate devices, and the brain wave intelligent acquisition and analysis system 200 may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in a contracted data format.
Fig. 4 is an application scenario diagram of an intelligent brain wave acquisition and analysis method provided in an embodiment of the present application. As shown in fig. 4, in this application scenario, first, an electroencephalogram signal of a subject is acquired (e.g., C illustrated in fig. 4); then, the acquired brain wave signals are input to a server (e.g., S illustrated in fig. 4) in which a brain wave intelligent acquisition and analysis algorithm is deployed, wherein the server is capable of processing the brain wave signals based on the brain wave intelligent acquisition and analysis algorithm to determine an emotion tag of the subject.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the application, and is not meant to limit the scope of the application, but to limit the application to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the application are intended to be included within the scope of the application.
Claims (6)
1. An intelligent brain wave acquisition and analysis method is characterized by comprising the following steps:
Acquiring brain wave signals of a tested person;
Extracting local neighborhood waveform characteristics of the brain wave signals to obtain a sequence of brain wave local neighborhood waveform association characteristic vectors;
The sequence of the brain wave local neighborhood waveform association feature vector is passed through a global feature interaction module based on internal self-attention to obtain a sequence of global enhanced brain wave waveform association feature vector;
determining an emotion tag of the tested person based on the sequence of the global enhanced brain wave waveform associated feature vector;
The sequence of the brain wave local neighborhood waveform association feature vector is passed through a global feature interaction module based on internal self-attention to obtain a sequence of global enhanced brain wave waveform association feature vector, which comprises the following steps:
Processing the sequence of brain wave local neighborhood waveform associated feature vectors by using the following global feature interaction formula to obtain the sequence of global enhanced brain wave waveform associated feature vectors; the global feature interaction formula is as follows:
；
wherein, Associating a first/>, in the sequence of feature vectors, for the brain wave local neighborhood waveformLocal neighborhood waveform of individual brain waves associates feature vectors with the/>Correlation matrix between local neighborhood waveform correlation feature vectors of individual brain waves,/>Associating a first/>, in the sequence of feature vectors, for the brain wave local neighborhood waveformLocal neighborhood waveform of individual brain waves associates feature vectors with the/>Correlation matrix between local neighborhood waveform correlation feature vectors of individual brain waves,/>For the number of the brain wave local neighborhood waveform associated feature vectors,/>The value of (2) is/>，/>For/>Local neighborhood waveform associated feature vector of brain wave,/>A/>, in the sequence of the global enhanced brain wave waveform associated feature vectorsGlobal enhanced brain wave waveform associated feature vector,/>Representing an exponential function operation;
the extracting the local neighborhood waveform characteristic of the brain wave signal to obtain a sequence of brain wave local neighborhood waveform associated characteristic vectors comprises the following steps:
Segmenting the brain wave signals based on a preset scale to obtain a sequence of brain wave local signals;
Extracting local neighborhood waveform characteristics of the sequence of the brain wave local signals by using a deep learning network model to obtain a sequence of brain wave local neighborhood waveform association characteristic vectors;
the deep learning network model is a brain wave local neighborhood waveform association feature extractor based on a convolution layer;
the method for extracting the local neighborhood waveform characteristics of the sequence of the brain wave local signals by using a deep learning network model to obtain the sequence of the brain wave local neighborhood waveform association characteristic vectors comprises the following steps:
And passing the sequence of the brain wave local signals through the brain wave local neighborhood waveform association feature extractor based on the convolution layer to obtain the sequence of the brain wave local neighborhood waveform association feature vector.
2. The brain wave intelligent acquisition and analysis method according to claim 1, wherein determining the emotion tag of the subject based on the sequence of the globally enhanced brain wave waveform-associated feature vectors comprises:
cascading the sequence of the global enhanced brain wave waveform association feature vector into a global enhanced brain wave waveform association cascading feature vector;
And the global enhanced brain wave waveform associated cascade feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for representing the emotion label of the tested person.
3. The brain wave intelligent acquisition and analysis method according to claim 2, wherein the global enhanced brain wave waveform association cascade feature vector is passed through a classifier to obtain a classification result, and the classification result is used for representing an emotion tag of the tested person, and the method comprises the following steps:
Performing full-connection coding on the global enhanced brain wave waveform association cascade feature vector by using a plurality of full-connection layers of the classifier to obtain a coding classification feature vector;
And the coding classification feature vector is passed through a Softmax classification function of the classifier to obtain the classification result.
4. The brain wave intelligent acquisition and analysis method according to claim 3, further comprising a training step of: and training the brain wave local neighborhood waveform associated feature extractor based on the convolution layer, the global feature interaction module based on the internal self-focusing and the classifier.
5. The brain wave intelligent acquisition and analysis method according to claim 4, wherein the training step includes:
Acquiring training data, wherein the training data comprises training brain wave signals of a tested person and true values of emotion labels of the tested person;
segmenting the training brain wave signals based on a preset scale to obtain a sequence of training brain wave local signals;
Passing the sequence of the training brain wave local signals through the brain wave local neighborhood waveform associated feature extractor based on the convolution layer to obtain a sequence of training brain wave local neighborhood waveform associated feature vectors;
Passing the sequence of the training brain wave local neighborhood waveform associated feature vectors through the global feature interaction module based on internal self-attention to obtain a sequence of training global enhanced brain wave waveform associated feature vectors;
After cascading the sequence of the training global enhanced brain wave waveform association feature vector into a training global enhanced brain wave waveform association cascading feature vector, the training global enhanced brain wave waveform association cascading feature vector passes through a classifier to obtain a classification loss function value;
And training the brain wave local neighborhood waveform associated feature extractor based on the convolution layer, the global feature interaction module based on the internal self-attention and the classifier by using the classification loss function value, wherein in each round of iteration of training, the training global enhanced brain wave waveform associated cascade feature vector is optimized.
6. An intelligent brain wave acquisition and analysis system, which is characterized by comprising:
the brain wave signal acquisition module is used for acquiring brain wave signals of the tested person;
The local neighborhood waveform feature extraction module is used for extracting local neighborhood waveform features of the brain wave signals to obtain a sequence of brain wave local neighborhood waveform association feature vectors;
the global feature interaction module is used for enabling the sequence of the brain wave local neighborhood waveform association feature vector to pass through the global feature interaction module based on internal self-attention so as to obtain the sequence of the global enhanced brain wave waveform association feature vector;
The emotion label determining module is used for determining emotion labels of the testee based on the sequence of the global enhanced brain wave waveform associated feature vectors;
the global feature interaction module comprises:
Processing the sequence of brain wave local neighborhood waveform associated feature vectors by using the following global feature interaction formula to obtain the sequence of global enhanced brain wave waveform associated feature vectors; the global feature interaction formula is as follows:
；
wherein, Associating a first/>, in the sequence of feature vectors, for the brain wave local neighborhood waveformLocal neighborhood waveform of individual brain waves associates feature vectors with the/>Correlation matrix between local neighborhood waveform correlation feature vectors of individual brain waves,/>Associating a first/>, in the sequence of feature vectors, for the brain wave local neighborhood waveformLocal neighborhood waveform of individual brain waves associates feature vectors with the/>Correlation matrix between local neighborhood waveform correlation feature vectors of individual brain waves,/>For the number of the brain wave local neighborhood waveform associated feature vectors,/>The value of (2) is/>，/>For/>Local neighborhood waveform associated feature vector of brain wave,/>A/>, in the sequence of the global enhanced brain wave waveform associated feature vectorsGlobal enhanced brain wave waveform associated feature vector,/>Representing an exponential function operation;
the extracting the local neighborhood waveform characteristic of the brain wave signal to obtain a sequence of brain wave local neighborhood waveform associated characteristic vectors comprises the following steps:
Segmenting the brain wave signals based on a preset scale to obtain a sequence of brain wave local signals;
Extracting local neighborhood waveform characteristics of the sequence of the brain wave local signals by using a deep learning network model to obtain a sequence of brain wave local neighborhood waveform association characteristic vectors;
the deep learning network model is a brain wave local neighborhood waveform association feature extractor based on a convolution layer;
the method for extracting the local neighborhood waveform characteristics of the sequence of the brain wave local signals by using a deep learning network model to obtain the sequence of the brain wave local neighborhood waveform association characteristic vectors comprises the following steps:
And passing the sequence of the brain wave local signals through the brain wave local neighborhood waveform association feature extractor based on the convolution layer to obtain the sequence of the brain wave local neighborhood waveform association feature vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410169443.7A CN117708682B (en) | 2024-02-06 | 2024-02-06 | Intelligent brain wave acquisition and analysis system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410169443.7A CN117708682B (en) | 2024-02-06 | 2024-02-06 | Intelligent brain wave acquisition and analysis system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117708682A CN117708682A (en) | 2024-03-15 |
| CN117708682B true CN117708682B (en) | 2024-04-19 |
Family
ID=90148398
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410169443.7A Active CN117708682B (en) | 2024-02-06 | 2024-02-06 | Intelligent brain wave acquisition and analysis system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117708682B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103610461A (en) * | 2013-10-17 | 2014-03-05 | 杭州电子科技大学 | EEG noise elimination method based on dual-density wavelet neighborhood related threshold processing |
| CN111259761A (en) * | 2020-01-13 | 2020-06-09 | 东南大学 | Electroencephalogram emotion recognition method and device based on migratable attention neural network |
| CN112464768A (en) * | 2020-11-18 | 2021-03-09 | 中国石油大学(华东) | Fatigue detection method based on self-attention multi-feature fusion |
| CN113598774A (en) * | 2021-07-16 | 2021-11-05 | 中国科学院软件研究所 | Active emotion multi-label classification method and device based on multi-channel electroencephalogram data |
| CN114886383A (en) * | 2022-05-16 | 2022-08-12 | 中国科学院自动化研究所 | Electroencephalogram signal emotional feature classification method based on transfer learning |
| CN115414051A (en) * | 2021-05-12 | 2022-12-02 | 西安邮电大学 | Emotion classification and recognition method of electroencephalogram signal self-adaptive window |
| CN115804602A (en) * | 2022-12-21 | 2023-03-17 | 西京学院 | Electroencephalogram emotion signal detection method, equipment and medium based on attention mechanism and with multi-channel feature fusion |
| CN115981470A (en) * | 2022-12-29 | 2023-04-18 | 杭州叶蓁科技有限公司 | Gesture recognition method and system based on feature joint coding |
| CN116350222A (en) * | 2023-01-16 | 2023-06-30 | 武汉工商学院 | Emotion recognition method and device based on electroencephalogram signals |
| CN116671918A (en) * | 2023-05-31 | 2023-09-01 | 平安科技（深圳）有限公司 | Emotion recognition method and device based on electroencephalogram signals, electronic equipment and medium |
| CN116881762A (en) * | 2022-12-09 | 2023-10-13 | 江西索普信实业有限公司 | Emotion recognition method based on dynamic brain network characteristics |
-
2024
- 2024-02-06 CN CN202410169443.7A patent/CN117708682B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103610461A (en) * | 2013-10-17 | 2014-03-05 | 杭州电子科技大学 | EEG noise elimination method based on dual-density wavelet neighborhood related threshold processing |
| CN111259761A (en) * | 2020-01-13 | 2020-06-09 | 东南大学 | Electroencephalogram emotion recognition method and device based on migratable attention neural network |
| CN112464768A (en) * | 2020-11-18 | 2021-03-09 | 中国石油大学(华东) | Fatigue detection method based on self-attention multi-feature fusion |
| CN115414051A (en) * | 2021-05-12 | 2022-12-02 | 西安邮电大学 | Emotion classification and recognition method of electroencephalogram signal self-adaptive window |
| CN113598774A (en) * | 2021-07-16 | 2021-11-05 | 中国科学院软件研究所 | Active emotion multi-label classification method and device based on multi-channel electroencephalogram data |
| CN114886383A (en) * | 2022-05-16 | 2022-08-12 | 中国科学院自动化研究所 | Electroencephalogram signal emotional feature classification method based on transfer learning |
| CN116881762A (en) * | 2022-12-09 | 2023-10-13 | 江西索普信实业有限公司 | Emotion recognition method based on dynamic brain network characteristics |
| CN115804602A (en) * | 2022-12-21 | 2023-03-17 | 西京学院 | Electroencephalogram emotion signal detection method, equipment and medium based on attention mechanism and with multi-channel feature fusion |
| CN115981470A (en) * | 2022-12-29 | 2023-04-18 | 杭州叶蓁科技有限公司 | Gesture recognition method and system based on feature joint coding |
| CN116350222A (en) * | 2023-01-16 | 2023-06-30 | 武汉工商学院 | Emotion recognition method and device based on electroencephalogram signals |
| CN116671918A (en) * | 2023-05-31 | 2023-09-01 | 平安科技（深圳）有限公司 | Emotion recognition method and device based on electroencephalogram signals, electronic equipment and medium |
Non-Patent Citations (2)
| Title |
|---|
| "A Multi-Dimensional Graph Convolution Network for EEG Emotion Recognition";Guanglong Du等;《IEEE Transactions on Instrumentation and Measurement》;20221231;正文1-11页 * |
| "基于混合神经网络的脑电时空特征情感分类";陈景霞等;《软件学报》;20201202;3869-3883页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117708682A (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111134666B (en) | Emotion recognition method of multi-channel electroencephalogram data and electronic device | |
| CN109299751B (en) | EMD data enhancement-based SSVEP electroencephalogram classification method of convolutional neural model | |
| Shahnaz et al. | Emotion recognition based on wavelet analysis of Empirical Mode Decomposed EEG signals responsive to music videos | |
| Guo et al. | A hybrid fuzzy cognitive map/support vector machine approach for EEG-based emotion classification using compressed sensing | |
| CN109375776B (en) | Electroencephalogram action intention recognition method based on multi-task RNN model | |
| CN112800998A (en) | Multi-mode emotion recognition method and system integrating attention mechanism and DMCCA | |
| CN111407243A (en) | Pulse signal pressure identification method based on deep learning | |
| CN113069117A (en) | Electroencephalogram emotion recognition method and system based on time convolution neural network | |
| CN115414051A (en) | Emotion classification and recognition method of electroencephalogram signal self-adaptive window | |
| CN108470182B (en) | Brain-computer interface method for enhancing and identifying asymmetric electroencephalogram characteristics | |
| CN116230234A (en) | Multi-mode feature consistency psychological health abnormality identification method and system | |
| CN109993132B (en) | Pattern recognition generation method and system based on electroencephalogram signals | |
| CN113974627B (en) | Emotion recognition method based on brain-computer generated confrontation | |
| Gao et al. | EEG classification based on sparse representation and deep learning | |
| Jamal et al. | Cloud-based human emotion classification model from EEG signals | |
| CN106843509B (en) | Brain-computer interface system | |
| CN117708682B (en) | Intelligent brain wave acquisition and analysis system and method | |
| CN116531001A (en) | Method and device for generating multi-listener electroencephalogram signals and identifying emotion of cross-listener | |
| CN115969392A (en) | Cross-period brainprint recognition method based on tensor frequency space attention domain adaptive network | |
| Rudas et al. | On activity identification pipelines for a low-accuracy EEG device | |
| CN113057654B (en) | Memory load detection and extraction system and method based on frequency coupling neural network model | |
| CN112069897A (en) | Knowledge graph-based voice and micro-expression recognition suicide emotion sensing method | |
| Wu et al. | Design of gesture recognition system based on multi-channel myoelectricity correlation | |
| CN112244873B (en) | Electroencephalogram space-time feature learning and emotion classification method based on hybrid neural network | |
| CN115844421B (en) | Electroencephalogram emotion recognition method and equipment based on fractional Fourier transform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |