CN113768518A - Electroencephalogram emotion recognition method and system based on multi-scale dispersion entropy analysis - Google Patents
Electroencephalogram emotion recognition method and system based on multi-scale dispersion entropy analysis Download PDFInfo
- Publication number
- CN113768518A CN113768518A CN202111031753.5A CN202111031753A CN113768518A CN 113768518 A CN113768518 A CN 113768518A CN 202111031753 A CN202111031753 A CN 202111031753A CN 113768518 A CN113768518 A CN 113768518A
- Authority
- CN
- China
- Prior art keywords
- electroencephalogram
- rhythm
- signal
- dispersion entropy
- entropy
- 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.)
- Pending
Links
- 239000006185 dispersion Substances 0.000 title claims abstract description 66
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000004458 analytical method Methods 0.000 title claims abstract description 30
- 230000008909 emotion recognition Effects 0.000 title claims abstract description 23
- 230000033764 rhythmic process Effects 0.000 claims abstract description 82
- 230000008451 emotion Effects 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims abstract description 15
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 238000001914 filtration Methods 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000001020 rhythmical effect Effects 0.000 claims description 10
- 239000000126 substance Substances 0.000 claims description 9
- 210000004556 brain Anatomy 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 6
- 230000033001 locomotion Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000012880 independent component analysis Methods 0.000 claims description 4
- 238000003066 decision tree Methods 0.000 claims description 3
- 230000007717 exclusion Effects 0.000 claims description 3
- 230000012010 growth Effects 0.000 claims description 3
- 230000002015 leaf growth Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims 1
- 230000002996 emotional effect Effects 0.000 abstract description 9
- 239000013598 vector Substances 0.000 abstract description 4
- 239000002131 composite material Substances 0.000 abstract description 2
- 230000010365 information processing Effects 0.000 abstract description 2
- 238000006073 displacement reaction Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000004397 blinking Effects 0.000 description 1
- 210000005252 bulbus oculi Anatomy 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000010247 heart contraction Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
- A61B5/372—Analysis of electroencephalograms
- A61B5/374—Detecting the frequency distribution of signals, e.g. detecting delta, theta, alpha, beta or gamma waves
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Psychiatry (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Medical Informatics (AREA)
- Signal Processing (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Heart & Thoracic Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physiology (AREA)
- Evolutionary Computation (AREA)
- Psychology (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Power Engineering (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention relates to the field of electroencephalogram information processing, and provides an electroencephalogram emotion recognition method and system based on multi-scale dispersion entropy analysis, wherein the method comprises the following steps: acquiring an original electroencephalogram signal, and preprocessing the original electroencephalogram signal to obtain a preprocessed electroencephalogram signal; the preprocessed electroencephalogram signals comprise: theta rhythm EEG signal, alpha rhythm EEG signal, beta rhythm EEG signal and gamma rhythm EEG signal; extracting the multi-scale dispersion entropy of the preprocessed electroencephalogram signal; inputting the multi-scale dispersion entropy into the trained light-weight gradient lifting tree to obtain emotion classification. The method can deeply analyze the emotional information contained in each rhythm electroencephalogram signal and extract the electroencephalogram emotional characteristics with stronger representation capability; the dispersion entropy is used as a feature extraction method, the distance between any two composite delay vectors with embedding dimensions s and s +1 respectively does not need to be calculated, the amplitude value of each embedded vector is not sequenced, and the feature extraction operation consumption can be effectively reduced.
Description
Technical Field
The invention relates to the field of electroencephalogram information processing, in particular to an electroencephalogram emotion recognition method and system based on multi-scale dispersion entropy analysis.
Background
At present, electroencephalogram emotional feature extraction is mainly carried out through methods such as time domain analysis, frequency domain analysis and time-frequency domain analysis. However, due to the strong randomness and non-stationarity of the electroencephalogram signals, analysis methods such as time domain, frequency domain and time-frequency domain analysis cannot comprehensively represent information of the electroencephalogram signals. The time domain analysis ignores the frequency domain information of the electroencephalogram signal, the frequency domain analysis loses the important phase characteristics of the signal, and the randomness and the non-stationarity of the electroencephalogram signal cannot be represented by the method and the time-frequency domain analysis method. The extracted features are not comprehensive enough in characterization of the electroencephalogram signals, and therefore the final emotion recognition result is influenced.
In recent years, with the successful application of more and more information entropy feature extraction methods in different physiological and psychological disease researches, the attention of students in the electroencephalogram emotion recognition research field to the information entropy extraction methods is also promoted. More and more researchers apply different information entropies to electroencephalogram emotion recognition research. At present, mainstream information entropy analysis methods applied to electroencephalogram emotion recognition mainly comprise sample entropy and displacement entropy analysis methods and the like, and a certain effect is achieved. However, these methods take too much time for feature extraction, and are unreliable in processing results of time-series signals with short length, and are easily interfered by noise. The representation capability of the methods on the emotion information contained in the electroencephalogram signals also needs to be further improved, and therefore the electroencephalogram emotion recognition rate is improved.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The main purpose of the invention is to solve the problems that the extracted features have weak representation capability on emotion information and low emotion recognition rate in the prior art; the calculation time consumption during the feature extraction is too long, the processing result of the time series signal with shorter length is unreliable, and the noise interference is easy to occur; the technical problem of insufficient representation capability of the emotional information contained in the electroencephalogram signal.
In order to achieve the purpose, the invention provides an electroencephalogram emotion recognition method based on multi-scale dispersion entropy analysis, which comprises the following steps:
s1: acquiring an original electroencephalogram signal, and preprocessing the original electroencephalogram signal to obtain a preprocessed electroencephalogram signal; the preprocessed electroencephalogram signals comprise: theta rhythm EEG signal, alpha rhythm EEG signal, beta rhythm EEG signal and gamma rhythm EEG signal;
s2: extracting the multi-scale dispersion entropy of the preprocessed electroencephalogram signal;
s3: inputting the multi-scale dispersion entropy into the trained light-weight gradient lifting tree to obtain emotion classification.
Preferably, step S1 is specifically:
s11: performing band-pass filtering on the original electroencephalogram signal to obtain an electroencephalogram signal after band-pass filtering;
s12: performing notch filtering on the EEG signal subjected to the band-pass filtering to obtain a notch-filtered EEG signal;
s13: removing electrocardio, electrooculogram, myoelectricity and motion artifacts in the EEG signals subjected to notch filtering by an independent component analysis method to obtain EEG signals subjected to artifact removal;
s14: and inputting the electroencephalogram signals without artifacts into a four-order Butterworth filter to obtain the preprocessed electroencephalogram signals.
Preferably, step S2 is specifically:
s21: selecting one rhythm electroencephalogram signal in the preprocessed electroencephalogram signals, and carrying out coarse graining processing on the rhythm electroencephalogram signal to obtain a coarse grained rhythm electroencephalogram signal xj(j ═ 1,2,. cndot, N); wherein j represents the time sequence of the rhythm electroencephalogram signal, and N represents the time length of the rhythm electroencephalogram signal;
s22: for the coarsely granulated rhythm electroencephalogram signal xjMapping to obtain mapped rhythm EEG signal YcExpressed as:
wherein the content of the first and second substances,representing the mapped rhythm electroencephalogram signals corresponding to each time sequence, and c representing the number of mapping categories;
s23: the mapped rhythm electroencephalogram signal YcConstructing a time series with a time delay parameter t and an embedding dimension sExpressed as:
wherein s represents the number of elements in the time series;
s24: obtaining the time seriesCorresponding dispersion modeWherein the content of the first and second substances,
S26: according to Shannon entropy theory, passing the potential relative probabilityCalculating to obtain the dispersion entropy of the rhythm electroencephalogram signal;
s27: repeating the steps S21-S26 for 4 times to obtain the multi-scale dispersion entropy, which comprises the following steps: the dispersion entropy of theta rhythm electroencephalogram signals, the dispersion entropy of alpha rhythm electroencephalogram signals, the dispersion entropy of beta rhythm electroencephalogram signals and the dispersion entropy of gamma rhythm electroencephalogram signals.
Preferably, the coarsely granulated electroencephalogram signal x of rhythm in step S21jThe expression of (a) is as follows:
wherein u isbRepresenting the rhythmic electroencephalogram signals, L representing the length of the rhythmic electroencephalogram signals, and tau representing the scale factor.
Preferably, the potential relative probabilities in step S25The calculation formula of (a) is as follows:
where, # denotes the calculation base.
Preferably, the calculation formula of the dispersion entropy of the rhythmic brain electrical signal in the step S26 is as follows:
wherein the content of the first and second substances,csrepresenting each time seriesU represents the rhythmic brain electrical signal.
Preferably, the training process of the trained lightweight gradient spanning tree in step S3 is as follows:
s31: acquiring a lightweight gradient lifting tree, and inputting training data into the lightweight gradient lifting tree;
s32: optimizing a layer growth strategy of a decision tree in the lightweight gradient spanning tree into a leaf growth strategy;
s33: reducing the storage space consumption during the operation of the lightweight gradient lifting tree by a histogram algorithm;
s34: reducing the data operation amount of the light-weight gradient lifting tree in operation through a unilateral gradient sampling algorithm and a mutual exclusion binding algorithm;
s35: optimizing the operation overhead and the operation efficiency of the lightweight gradient lifting tree in the classification process;
s36: and selecting the optimal model hyper-parameter of the lightweight gradient lifting tree by a Bayesian optimization method to obtain the trained lightweight gradient lifting tree.
A brain electric emotion recognition system based on multi-scale dispersion entropy analysis comprises:
the preprocessing module is used for acquiring an original electroencephalogram signal, preprocessing the original electroencephalogram signal and acquiring a preprocessed electroencephalogram signal; the preprocessed electroencephalogram signals comprise: theta rhythm EEG signal, alpha rhythm EEG signal, beta rhythm EEG signal and gamma rhythm EEG signal;
the multi-scale dispersion entropy extraction module is used for extracting the multi-scale dispersion entropy of the preprocessed electroencephalogram signal;
and the emotion classification module is used for inputting the multi-scale dispersion entropy into the trained light-weight gradient lifting tree to obtain emotion classification.
The invention has the following beneficial effects:
1. the emotion information contained in each rhythm electroencephalogram signal can be deeply analyzed, and electroencephalogram emotion characteristics with stronger representation capability are extracted;
2. the dispersion entropy is used as a feature extraction method, the distance between any two composite delay vectors with embedding dimensions s and s +1 respectively does not need to be calculated, the amplitude value of each embedded vector is not sequenced, and the feature extraction operation consumption can be effectively reduced.
Drawings
FIG. 1 is a flow chart of a method according to an embodiment of the present invention;
FIG. 2 is a system block diagram according to an embodiment of the present invention;
the implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
At present, electroencephalogram emotional feature extraction is mainly carried out by methods such as time domain analysis, frequency domain analysis, time-frequency domain analysis and the like, and the traditional methods have the following problems:
1. the electroencephalogram signals belong to low-frequency weak signals with strong nonlinearity and non-stationarity, and the existing time domain, frequency domain and time frequency domain electroencephalogram emotional feature extraction methods cannot well represent emotional information contained in the electroencephalogram signals, so that the extracted features have weak representation capability on the emotional information and low emotion recognition rate.
2. Analysis methods such as sample entropy, displacement entropy, dispersion entropy, multi-scale sample entropy and multi-scale displacement entropy are applied to electroencephalogram emotional feature extraction, and certain effects are achieved.
3. The representation capability of the sample entropy, the displacement entropy, the dispersion entropy and other methods on the emotion information contained in the electroencephalogram signal also needs to be further improved, and therefore the electroencephalogram emotion recognition rate is improved.
Referring to fig. 1, aiming at the above problems, the invention provides a method for recognizing electroencephalogram emotion based on multi-scale dispersion entropy analysis, which comprises the following steps:
s1: acquiring an original electroencephalogram signal, and preprocessing the original electroencephalogram signal to obtain a preprocessed electroencephalogram signal; the preprocessed electroencephalogram signals comprise: theta rhythm EEG signal, alpha rhythm EEG signal, beta rhythm EEG signal and gamma rhythm EEG signal;
s2: extracting the multi-scale dispersion entropy of the preprocessed electroencephalogram signal;
s3: inputting the multi-scale dispersion entropy into the trained light-weight gradient lifting tree to obtain emotion classification.
In this embodiment, step S1 specifically includes:
s11: performing band-pass filtering on the original electroencephalogram signal to obtain an electroencephalogram signal after band-pass filtering;
in the specific implementation, the main frequency range of the electroencephalogram signals subjected to band-pass filtering is 0.5-50 Hz, the frequency range acquired by electroencephalogram acquisition equipment is usually larger than the frequency range, a band-pass filter is required to perform band-pass filtering on the acquired original electroencephalogram signals, and the rest frequency components are removed to obtain the main frequency range of the electroencephalogram signals subjected to band-pass filtering;
s12: performing notch filtering on the EEG signal subjected to the band-pass filtering to obtain a notch-filtered EEG signal;
in the specific implementation, because of power frequency interference, the EEG signal after band-pass filtering still contains 50Hz power frequency noise, and the 50Hz power frequency noise needs to be filtered by a notch filter;
s13: removing electrocardio, electrooculogram, myoelectricity and motion artifacts in the EEG signals subjected to notch filtering by an independent component analysis method to obtain EEG signals subjected to artifact removal;
in the specific implementation, heart beating, eyeball movement, blinking, muscle activity and body movement of a person all affect the electroencephalogram signal, so that corresponding artifact components appear in the electroencephalogram signal, and various components can be separated by using an independent component analysis method, so that the artifacts are removed;
s14: inputting the electroencephalogram signals without artifacts into a four-order Butterworth filter to obtain the preprocessed electroencephalogram signals;
in the specific implementation, the preprocessed electroencephalogram signals mainly comprise: the four-step Butterworth filter is directly adopted to decompose the electroencephalogram signals to obtain theta rhythm electroencephalogram signals, alpha rhythm electroencephalogram signals, beta rhythm electroencephalogram signals and gamma rhythm electroencephalogram signals.
In this embodiment, step S2 specifically includes:
s21: selecting one rhythm electroencephalogram signal in the preprocessed electroencephalogram signals, and carrying out coarse graining processing on the rhythm electroencephalogram signal to obtain a coarse grained rhythm electroencephalogram signal xj(j ═ 1,2,. cndot, N); wherein j represents the time sequence of the rhythm electroencephalogram signal, and N represents the time length of the rhythm electroencephalogram signal;
s22: for the coarsely granulated rhythm electroencephalogram signal xjMapping, and using S-type mapping method to obtain coarse-grained rhythm EEG signal xjMapping to c different categories (the indexes of the c categories are integers from 1 to c respectively) to obtain mapped rhythm electroencephalogram signals YcExpressed as:
wherein the content of the first and second substances,representing the mapped rhythm electroencephalogram signals corresponding to each time sequence, and c representing the number of mapping categories;
s23: the mapped rhythm electroencephalogram signal YcConstructing a time series with a time delay parameter t and an embedding dimension sExpressed as:
wherein s represents the number of elements in the time series;
s24: obtaining the time seriesCorresponding dispersion modeWherein the content of the first and second substances,
in the calculation of step S24, each time series is assignedIs equal to cs(ii) a Because of the time sequenceHas s elements, and each element may be mapped into a category indexed by integers 1 through c;
S26: according to Shannon entropy theory, passing the potential relative probabilityCalculating to obtain Dispersion Entropy (Dispersion Entropy, Dispen) of the rhythm electroencephalogram signal;
s27: repeating the steps S21-S26 for 4 times to obtain the multi-scale dispersion entropy, which comprises the following steps: the dispersion entropy of theta rhythm electroencephalogram signals, the dispersion entropy of alpha rhythm electroencephalogram signals, the dispersion entropy of beta rhythm electroencephalogram signals and the dispersion entropy of gamma rhythm electroencephalogram signals.
In this embodiment, the coarsely granulated electroencephalogram signal x of rhythm described in step S21jThe expression of (a) is as follows:
wherein u isbRepresenting the rhythmic electroencephalogram signals, L representing the length of the rhythmic electroencephalogram signals, and tau representing the scale factor.
In this embodiment, the potential relative probability in step S25The calculation formula of (a) is as follows:
wherein, # denotes the calculation base; in practice, the amount of the liquid to be used,indicating assignment to time seriesDispersion mode ofIs divided by s as the embedding dimension, and thus,can also be expressed as:
in this embodiment, the formula for calculating the dispersion entropy of the rhythm electroencephalogram signal in step S26 is as follows:
wherein the content of the first and second substances,csrepresenting each time seriesU represents the rhythmic brain electrical signal.
In this embodiment, the training process of the trained lightweight gradient spanning tree in step S3 is as follows:
s31: acquiring a lightweight gradient lifting tree, and inputting training data into the lightweight gradient lifting tree;
s32: optimizing a layer growth strategy of a decision tree in the lightweight gradient spanning tree into a leaf growth strategy;
s33: reducing the storage space consumption during the operation of the lightweight gradient lifting tree by a histogram algorithm;
s34: reducing the data operation amount of the light-weight gradient lifting tree in operation through a unilateral gradient sampling algorithm and a mutual exclusion binding algorithm;
s35: optimizing the operation overhead and the operation efficiency of the lightweight gradient lifting tree in the classification process;
s36: and selecting the optimal model hyper-parameter of the lightweight gradient lifting tree by a Bayesian optimization method to obtain the trained lightweight gradient lifting tree.
Referring to fig. 2, the invention provides a brain emotion recognition system based on multi-scale dispersion entropy analysis, comprising:
the preprocessing module 10 is used for acquiring an original electroencephalogram signal, preprocessing the original electroencephalogram signal and acquiring a preprocessed electroencephalogram signal; the preprocessed electroencephalogram signals comprise: theta rhythm EEG signal, alpha rhythm EEG signal, beta rhythm EEG signal and gamma rhythm EEG signal;
a multi-scale dispersion entropy extraction module 20, configured to extract a multi-scale dispersion entropy of the preprocessed electroencephalogram signal;
and the emotion classification module 30 is configured to input the multi-scale dispersion entropy into the trained lightweight gradient lifting tree to obtain emotion classification.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third and the like do not denote any order, but rather the words first, second and the like may be interpreted as indicating any order.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (8)
1. A brain electric emotion recognition method based on multi-scale dispersion entropy analysis is characterized by comprising the following steps:
s1: acquiring an original electroencephalogram signal, and preprocessing the original electroencephalogram signal to obtain a preprocessed electroencephalogram signal; the preprocessed electroencephalogram signals comprise: theta rhythm EEG signal, alpha rhythm EEG signal, beta rhythm EEG signal and gamma rhythm EEG signal;
s2: extracting the multi-scale dispersion entropy of the preprocessed electroencephalogram signal;
s3: inputting the multi-scale dispersion entropy into the trained light-weight gradient lifting tree to obtain emotion classification.
2. The electroencephalogram emotion recognition method based on multi-scale dispersion entropy analysis, as claimed in claim 1, wherein step S1 specifically comprises:
s11: performing band-pass filtering on the original electroencephalogram signal to obtain an electroencephalogram signal after band-pass filtering;
s12: performing notch filtering on the EEG signal subjected to the band-pass filtering to obtain a notch-filtered EEG signal;
s13: removing electrocardio, electrooculogram, myoelectricity and motion artifacts in the EEG signals subjected to notch filtering by an independent component analysis method to obtain EEG signals subjected to artifact removal;
s14: and inputting the electroencephalogram signals without artifacts into a four-order Butterworth filter to obtain the preprocessed electroencephalogram signals.
3. The electroencephalogram emotion recognition method based on multi-scale dispersion entropy analysis, as claimed in claim 1, wherein step S2 specifically comprises:
s21: selecting one rhythm electroencephalogram signal in the preprocessed electroencephalogram signals, and carrying out coarse graining processing on the rhythm electroencephalogram signal to obtain a coarse grained rhythm electroencephalogram signal xj(j ═ 1,2,. cndot, N); wherein j represents the time sequence of the rhythm electroencephalogram signal, and N represents the time length of the rhythm electroencephalogram signal;
s22: for the coarsely granulated rhythm electroencephalogram signal xjMapping to obtain mapped rhythm EEG signal YcExpressed as:
wherein the content of the first and second substances,representing the mapped rhythm electroencephalogram signals corresponding to each time sequence, and c representing the number of mapping categories;
s23: the mapped rhythm electroencephalogram signal YcConstructing a time series with a time delay parameter t and an embedding dimension sExpressed as:
wherein s represents the number of elements in the time series;
s24: obtaining the time seriesCorresponding dispersion modeWherein the content of the first and second substances,
S26: according to Shannon entropy theory, passing the potential relative probabilityComputingObtaining the dispersion entropy of the rhythm electroencephalogram signal;
s27: repeating the steps S21-S26 for 4 times to obtain the multi-scale dispersion entropy, which comprises the following steps: the dispersion entropy of theta rhythm electroencephalogram signals, the dispersion entropy of alpha rhythm electroencephalogram signals, the dispersion entropy of beta rhythm electroencephalogram signals and the dispersion entropy of gamma rhythm electroencephalogram signals.
4. The electroencephalogram emotion recognition method based on multi-scale dispersion entropy analysis as claimed in claim 3, wherein the coarsely granulated rhythm electroencephalogram signal x in step S21jThe expression of (a) is as follows:
wherein u isbRepresenting the rhythmic electroencephalogram signals, L representing the length of the rhythmic electroencephalogram signals, and tau representing the scale factor.
6. The electroencephalogram emotion recognition method based on multi-scale dispersion entropy analysis, as claimed in claim 3, wherein the calculation formula of the dispersion entropy of the rhythm electroencephalogram signal in step S26 is as follows:
7. The electroencephalogram emotion recognition method based on multi-scale dispersion entropy analysis, as claimed in claim 1, wherein the training process of the trained lightweight gradient spanning tree in step S3 is as follows:
s31: acquiring a lightweight gradient lifting tree, and inputting training data into the lightweight gradient lifting tree;
s32: optimizing a layer growth strategy of a decision tree in the lightweight gradient spanning tree into a leaf growth strategy;
s33: reducing the storage space consumption during the operation of the lightweight gradient lifting tree by a histogram algorithm;
s34: reducing the data operation amount of the light-weight gradient lifting tree in operation through a unilateral gradient sampling algorithm and a mutual exclusion binding algorithm;
s35: optimizing the operation overhead and the operation efficiency of the lightweight gradient lifting tree in the classification process;
s36: and selecting the optimal model hyper-parameter of the lightweight gradient lifting tree by a Bayesian optimization method to obtain the trained lightweight gradient lifting tree.
8. The utility model provides a brain electricity emotion recognition system based on multiscale dispersion entropy analysis which characterized in that includes:
the preprocessing module is used for acquiring an original electroencephalogram signal, preprocessing the original electroencephalogram signal and acquiring a preprocessed electroencephalogram signal; the preprocessed electroencephalogram signals comprise: theta rhythm EEG signal, alpha rhythm EEG signal, beta rhythm EEG signal and gamma rhythm EEG signal;
the multi-scale dispersion entropy extraction module is used for extracting the multi-scale dispersion entropy of the preprocessed electroencephalogram signal;
and the emotion classification module is used for inputting the multi-scale dispersion entropy into the trained light-weight gradient lifting tree to obtain emotion classification.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111031753.5A CN113768518A (en) | 2021-09-03 | 2021-09-03 | Electroencephalogram emotion recognition method and system based on multi-scale dispersion entropy analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111031753.5A CN113768518A (en) | 2021-09-03 | 2021-09-03 | Electroencephalogram emotion recognition method and system based on multi-scale dispersion entropy analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113768518A true CN113768518A (en) | 2021-12-10 |
Family
ID=78841004
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111031753.5A Pending CN113768518A (en) | 2021-09-03 | 2021-09-03 | Electroencephalogram emotion recognition method and system based on multi-scale dispersion entropy analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113768518A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114209339A (en) * | 2021-12-14 | 2022-03-22 | 山东大学 | Real-time electroencephalogram monitoring system for cardio-pulmonary resuscitation |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110472649A (en) * | 2019-06-21 | 2019-11-19 | 中国地质大学(武汉) | Brain electricity sensibility classification method and system based on multiscale analysis and integrated tree-model |
KR20190130808A (en) * | 2018-05-15 | 2019-11-25 | 연세대학교 산학협력단 | Emotion Classification Device and Method using Convergence of Features of EEG and Face |
CN110866528A (en) * | 2019-10-28 | 2020-03-06 | 腾讯科技(深圳)有限公司 | Model training method, energy consumption use efficiency prediction method, device and medium |
CN110942103A (en) * | 2019-12-05 | 2020-03-31 | 桂林电子科技大学 | Training method of classifier and computer-readable storage medium |
-
2021
- 2021-09-03 CN CN202111031753.5A patent/CN113768518A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20190130808A (en) * | 2018-05-15 | 2019-11-25 | 연세대학교 산학협력단 | Emotion Classification Device and Method using Convergence of Features of EEG and Face |
CN110472649A (en) * | 2019-06-21 | 2019-11-19 | 中国地质大学(武汉) | Brain electricity sensibility classification method and system based on multiscale analysis and integrated tree-model |
CN110866528A (en) * | 2019-10-28 | 2020-03-06 | 腾讯科技(深圳)有限公司 | Model training method, energy consumption use efficiency prediction method, device and medium |
CN110942103A (en) * | 2019-12-05 | 2020-03-31 | 桂林电子科技大学 | Training method of classifier and computer-readable storage medium |
Non-Patent Citations (1)
Title |
---|
XUE-WEN DING等: "Electroencephalogram Emotion Recognition Based on Dispersion Entropy Feature Extraction Using Random Over-Sampling Imbalanced Data Processing", 《IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114209339A (en) * | 2021-12-14 | 2022-03-22 | 山东大学 | Real-time electroencephalogram monitoring system for cardio-pulmonary resuscitation |
CN114209339B (en) * | 2021-12-14 | 2024-03-29 | 山东大学 | Real-time electroencephalogram monitoring system for cardiopulmonary resuscitation |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Li et al. | Deep convolutional neural network based ECG classification system using information fusion and one-hot encoding techniques | |
CN106709469B (en) | Automatic sleep staging method based on electroencephalogram and myoelectricity multiple characteristics | |
CN111067514B (en) | Multi-channel electroencephalogram coupling analysis method based on multi-scale multivariable transfer entropy | |
CN109758145B (en) | Automatic sleep staging method based on electroencephalogram causal relationship | |
CN109674468A (en) | It is a kind of singly to lead brain electricity sleep mode automatically method by stages | |
CN113786204A (en) | Epilepsia intracranial electroencephalogram early warning method based on deep convolution attention network | |
CN111329474A (en) | Electroencephalogram identity recognition method and system based on deep learning and information updating method | |
CN104809434A (en) | Sleep staging method based on single-channel electroencephalogram signal ocular artifact removal | |
CN112022153B (en) | Electroencephalogram signal detection method based on convolutional neural network | |
CN111202512A (en) | Electrocardiogram classification method and device based on wavelet transformation and DCNN | |
CN113951903B (en) | High-speed railway dispatcher overload state identification method based on electroencephalogram data determination | |
CN109492546A (en) | A kind of bio signal feature extracting method merging wavelet packet and mutual information | |
CN115640827B (en) | Intelligent closed-loop feedback network method and system for processing electrical stimulation data | |
CN108647584B (en) | Arrhythmia identification and classification method based on sparse representation and neural network | |
CN111695500A (en) | Method and system for recognizing motor imagery task of stroke patient based on transfer learning | |
CN110490257A (en) | It is a kind of based on the electro-physiological signals entropy analysis method for removing trend term | |
Mishra et al. | Noise removal in EEG signals using SWT–ICA combinational approach | |
CN113768518A (en) | Electroencephalogram emotion recognition method and system based on multi-scale dispersion entropy analysis | |
Zachariah et al. | Automatic EEG artifact removal by independent component analysis using critical EEG rhythms | |
CN113269048B (en) | Motor imagery electroencephalogram signal classification method based on deep learning and mixed noise data enhancement | |
CN114387668A (en) | Classification method based on multi-level neuromuscular coupling characteristic information fusion | |
CN113842115A (en) | Improved EEG signal feature extraction method | |
CN111887811B (en) | Brain abnormal discharge detection method and system based on electroencephalogram signal characteristics | |
Upadhyay et al. | Application of wavelet fractal features for the automated detection of epileptic seizure using electroencephalogram signals | |
CN117520891A (en) | Motor imagery electroencephalogram signal classification method and system |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211210 |