CN108520239A - A kind of Method of EEG signals classification and system - Google Patents
A kind of Method of EEG signals classification and system Download PDFInfo
- Publication number
- CN108520239A CN108520239A CN201810317158.XA CN201810317158A CN108520239A CN 108520239 A CN108520239 A CN 108520239A CN 201810317158 A CN201810317158 A CN 201810317158A CN 108520239 A CN108520239 A CN 108520239A
- Authority
- CN
- China
- Prior art keywords
- eeg signals
- waves
- pretreated
- classification
- carried out
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
- G06F2218/06—Denoising by applying a scale-space analysis, e.g. using wavelet analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- 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
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- Complex Calculations (AREA)
Abstract
The present invention discloses a kind of Method of EEG signals classification and system.The method includes：Obtain EEG signals；EEG signals are pre-processed by independent component analysis；Feature extraction is carried out to pretreated EEG signals by dual-tree complex wavelet transform and cospace pattern；Classification processing is carried out to the EEG signals after feature extraction by Multiple Kernel Learning support vector machines.Using this method and system in the present invention, can be rapidly and efficiently classification difference is carried out to EEG signals.
Description
Technical field
The present invention relates to EEG signals control fields, more particularly to a kind of Method of EEG signals classification and system.
Background technology
Brain-computer interface (brain-computer interface, BCI) is in human brain and computer or other electronic equipments
Between establish it is direct exchange and control channel, idea or manipulation directly can be expressed by brain by this channel people
Equipment, without language or action, this patient that can effectively enhance body handicap exchanges or controls with the external world external
The ability of environment, to improve the quality of life of patient.Brain-computer interface technology is that one kind being related to Neuscience, signal detection, signal
The multi-disciplinary interleaving techniques such as processing, pattern-recognition.
BCI technologies provide a kind of new rehabilitation maneuver for dyskinesia, it directly provides for brain a kind of new
Communication and controlling soil moist, and do not depend on the normal output channel of brain.But technical research is still in initial stage of development, not enough
Maturation, for example, the communication speed of BCI is not also especially desirable, and individual difference is larger.Now, it more and more scholars and grinds
Study carefully mechanism is to the interested major reason of BCI technologies, it may cannot be by speaking or limb action is thought to express for those
Method or the people for operating equipment provide a kind of new communication and control mode.Many traffic accidents are destructive, are had the honor under survival
The people come are much severe disabilities, and even hand, eye, facial muscles, nerve fiber etc. is all damaged.Further more, many diseases
Disease, such as amyotrophic lateral sclerosis, brain paralysis, brain stem apoplexy, spinal cord injury, may all cause neuron death or
Apoptosis causes the damage of nerve pathway, makes brain that can not be exchanged with external environment by normal neuromuscular.This feelings
Under condition, traditional disabled person's ancillary technique is required for some form of muscular movement to participate in due to more or less, using by
Limitation, therefore it is then one kind can that a kind of direct communication channel is established between brain and external environment using BCI technologies
The selection of energy.
The NeuroSky companies in the U.S. develop brain wave remote control mobile phone, pass through the sensor technology and software of NeuroSky
Algorithm derives the psychological condition of user.The product of the said firm's research and development before is chiefly used in medical application, but the cost of sensor
Height is not easy to popularize, and is frequently used in medical field, predominantly detects defective EEG signals.
The neural scientific ＆ technical corporation " Emotiv Systems " in California, USA San Francisco is proposed a entitled " Emotiv
" nerve head is blue " of EPOC ", user need to only play thought aroused in interest, so that it may which to manipulate computer at the moment, but it is expensive, makes
With being limited in scope.
Tsinghua University is dedicated to one machine interface system of real-time brain based on Steady State Visual Evoked Potential, achieves field neck
The first high rate of information throughput, establishes a set of advanced nerve signal process and method for classifying modes.Tsinghua University god at present
The research of brain-computer interface is mainly promoted in terms of two through engineering experiment room：It is filled first, developing the brain-computer interface with practicability
It sets；Second is that research brain-computer interface control during neuromechanism to realize the brain-computer interface algorithm with coadaptation ability,
But due to being real-time brain machine interface system, the rate of information throughput is high so that data processing occupies more resource, and counts
Excessive according to amount, the requirement to hardware is also higher and higher, and corresponding cost can increase.
Whether Shanghai Communications University can be generated by study subject continuous or lasting can be counted as the stable state of motion
The ERD/ERS phenomenons of imagination task are similar to Steady State Visual Evoked Potential (steady-state visual evoked
Potential, SSVEP) or auditory steady state (auditory steady-state responses, ASSR), it devises
A kind of new experimental paradigm.When imagination task is continuous and repeats, phenomenon can also continue to occur.It can be on the phenomenological theory
The Mental imagery task for being construed to repeat causes cortical reaction and enters a kind of " stable state ", and is moved accordingly in cerebral cortex
Control zone generates the lasting phenomenon that desynchronizes, but the Mental imagery of single direction is more, and the time needed is long, and demand takes " steady
State " is more.
Invention content
The object of the present invention is to provide a kind of Method of EEG signals classification and system, come rapidly and efficiently to EEG signals into
Row difference classification.
To achieve the above object, the present invention provides following schemes：
A kind of Method of EEG signals classification, the method includes：
Obtain EEG signals；
EEG signals are pre-processed by independent component analysis；
Feature extraction is carried out to pretreated EEG signals by dual-tree complex wavelet transform and cospace pattern；
Classification processing is carried out to the EEG signals after feature extraction by Multiple Kernel Learning support vector machines.
Optionally, the acquisition EEG signals specifically include：
EEG signals are obtained by mindset.
Optionally, it is described by independent component analysis to EEG signals carry out pretreatment specifically include：
Centralization processing is carried out to the EEG signals；
Whitening processing is carried out to the EEG signals after the centralization, obtains orthogonal matrix Z；
According to the orthogonal matrix Z, an initial weight W is randomly choosed, W is enabled*=E { Zg (WTZ)}-E{g’(WTZ) } W,
Wherein W*For the adjoint matrix of W, g is nonlinear function, a1For constant, 1≤a1≤2；
Judge W=W*/||W*| | value whether be 1, obtain judging result, if judging result indicate W=W*/||W*| |
When value is 1, W is exported, if judging result indicates W=W*/||W*| | value when not being 1, then execute described according to the orthogonal moment
Battle array randomly chooses an initial weight W, enables W*=E { Zg (WTZ)}-E{g’(WTZ) } W, wherein W*For the adjoint matrix of W, g is
Nonlinear function, a1For constant, 1≤a1≤2。
Optionally, described that feature is carried out to pretreated EEG signals by dual-tree complex wavelet transform and cospace pattern
Extraction specifically includes：
The pretreated EEG signals are sampled；
Even numbers complex wavelet transform decomposition is carried out to the EEG signals after the sampling, chooses the frequency of δ waves, θ waves, α waves and β waves
The corresponding frequency range of rate range, and the δ waves, θ waves, α waves and β waves are reconstructed；
To the frequency range progress cospace Pattern Filter of δ waves, θ waves, α waves and β waves after the reconstruct, and then pre-processed
The feature vector of EEG signals afterwards.
Optionally, it is described to the pretreated EEG signals carry out sampling specifically include：
According to frequency fsPretreated EEG signals are sampled, fs=30Hz2l, it is small that wherein l indicates that even numbers is answered
The number of plies in wave conversion.
The present invention still further provides a kind of eeg signal classification system, the system comprises：
Acquisition module, for obtaining EEG signals；
Preprocessing module pre-processes EEG signals for passing through independent component analysis；
Characteristic extracting module, for by dual-tree complex wavelet transform and cospace pattern to pretreated EEG signals into
Row feature extraction；
Sort module, for being carried out at classification to the EEG signals after feature extraction by Multiple Kernel Learning support vector machines
Reason.
Optionally, which is characterized in that the preprocessing module specifically includes：
Centralization processing unit, for carrying out centralization processing to the EEG signals；
Whitening processing unit obtains orthogonal matrix Z for carrying out whitening processing to the EEG signals after the centralization；
Initial weight acquiring unit, for according to the orthogonal matrix Z, randomly choosing an initial weight W, enabling W*=E
{Zg(WTZ)}-E{g’(WTZ) } W, wherein W*For the adjoint matrix of W, g is nonlinear function, a1For constant, 1≤a1≤2；
Judging unit, for judging W=W*/||W*| | value whether be 1, obtain judging result；
Output unit, for indicating W=W when judging result*/||W*| | value be 1 when, export W；
Execution unit, for indicating W=W according to judging result*/||W*| | value when not being 1, then execute described according to institute
Orthogonal matrix is stated, an initial weight W is randomly choosed, enables W*=E { Zg (WTZ)}-E{g’(WTZ) } W, wherein W*For the adjoint of W
Matrix, g are nonlinear function, a1For constant, 1≤a1≤2。
Optionally, the characteristic extracting module specifically includes：
Sampling unit, for being sampled to the pretreated EEG signals；
Even numbers complex wavelet transform resolving cell, for carrying out even numbers complex wavelet transform decomposition to the EEG signals after sampling,
Choose the corresponding frequency range of frequency range of δ waves, θ waves, α waves and β waves；
Reconfiguration unit, for the δ waves, θ waves, α waves and β waves to be reconstructed；
Filter unit is used for the frequency range progress cospace Pattern Filter to δ waves, θ waves, α waves and β waves after the reconstruct,
And then obtain the feature vector of pretreated EEG signals.
According to specific embodiment provided by the invention, the invention discloses following technique effects：
EEG signals are pre-processed by using independent component analysis in the present invention, it is more quickly steady.
Carrying out feature extraction to pretreated EEG signals by using even numbers complex wavelet transform and cospace pattern can
More effectively to reduce the redundancy that the EEG signals frequency unrelated to Mental imagery and some uncorrelated or noise pollutions provide
The influences such as information so that the characteristic value gap extracted is big as far as possible, helps to distinguish every type games imagination task.
It can be classified with the simplified mode by using the multitask classification of Multiple Kernel Learning support vector machines, can be rapidly and efficiently
Classification difference is carried out to EEG signals.
Description of the drawings
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example, for those of ordinary skill in the art, without having to pay creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is Method of EEG signals classification flow chart of the embodiment of the present invention；
Fig. 2 is eeg signal classification system structure diagram of the embodiment of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The object of the present invention is to provide a kind of Method of EEG signals classification and system, come rapidly and efficiently to EEG signals into
Row difference classification.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, below in conjunction with the accompanying drawings and specific real
Applying mode, the present invention is described in further detail.
Fig. 1 is Method of EEG signals classification flow chart of the embodiment of the present invention, as shown in Figure 1, the method includes：
Step 101：Obtain EEG signals；
Step 102：EEG signals are pre-processed by independent component analysis；
Step 103：Feature is carried out by dual-tree complex wavelet transform and cospace pattern to pretreated EEG signals to carry
It takes；
Step 104：Classification processing is carried out to the EEG signals after feature extraction by Multiple Kernel Learning support vector machines.
It describes in detail below to each step：
Specifically, obtaining EEG signals in the step 101, obtained by mindset.
Mindset is derived from brain-computer interface technology years of researches achievement.People is under different psychological conditions, brain wave meeting
Different wave characters is shown, brain-computer interface technology decodes the idea of wearer with these features, to realize
Acquisition to EEG signals.
The step 102 pre-processes EEG signals by independent component analysis, specifically includes：
Step 1021：Centralization processing is carried out to the EEG signals so that the mean value of EEG signals is 0；
Step 1022：Whitening processing is carried out to the EEG signals after the centralization, obtains orthogonal matrix Z；
Step 1023：According to the orthogonal matrix Z, an initial weight W is randomly choosed, W is enabled*=E { Zg (WTZ)}-E{g’
(WTZ) } W, wherein W*For the adjoint matrix of W, g is nonlinear function, a1For constant, 1≤a1≤2；Wherein, the dimension of W matrixes with
The dimension of matrix Z is identical, and is the similar matrix of matrix Z；
Step 1024：Judge W=W*/||W*| | value whether be 1, obtain judging result, if judging result indicate W=W*/
||W*| | value be 1 when, export W, if judging result indicate W=W*/||W*| | value when not being 1, then execute described in the basis
Orthogonal matrix randomly chooses an initial weight W, enables W*=E { Zg (WTZ)}-E{g’(WTZ) } W, wherein W*For the adjoint matrix of W
Battle array, g is nonlinear function, a1For constant, 1≤a1≤2。
Wherein W=W*/||W*| | it is Rule of judgment, works as W=W*/||W*| | value be 1 when, then illustrate to restrain, output at this time
W matrixes, if W=W*/||W*| | value be 1 when, then illustrate not restrain, choose W again at this time, until convergence.
Specifically, g is nonlinear function, g=tanh (a can be chosen1, y) or g=y exp (- y2/ 2) or g=y3。
The step 103 carries out pretreated EEG signals by dual-tree complex wavelet transform and cospace pattern special
Sign extraction, specifically includes：
Step 1031：The pretreated EEG signals are sampled；
Step 1032：Even numbers complex wavelet transform decomposition is carried out to the EEG signals after the sampling, chooses δ waves, θ waves, α waves
Frequency range corresponding with the frequency range of β waves, and the δ waves, θ waves, α waves and β waves are reconstructed；Wherein in frequency fsUnder to pre-
Treated, and EEG signals are sampled, fs=30Hz2l, wherein l indicate even numbers complex wavelet transform in the number of plies.
Step 1033：To the frequency range progress cospace Pattern Filter of δ waves, θ waves, α waves and β waves after the reconstruct, in turn
Obtain the feature vector of pretreated EEG signals.
Fig. 2 is eeg signal classification system structure diagram of the embodiment of the present invention, as shown in Fig. 2, the system comprises：
Acquisition module 201, for obtaining EEG signals；
Preprocessing module 202 pre-processes EEG signals for passing through independent component analysis；
Characteristic extracting module 203, for passing through dual-tree complex wavelet transform and cospace pattern to pretreated brain telecommunications
Number carry out feature extraction；
Sort module 204, for being classified to the EEG signals after feature extraction by Multiple Kernel Learning support vector machines
Processing.
Specifically, the preprocessing module specifically includes：
Centralization processing unit 2021, for carrying out centralization processing to the EEG signals；
Whitening processing unit 2022 obtains orthogonal moment for carrying out whitening processing to the EEG signals after the centralization
Battle array Z；
Initial weight acquiring unit 2023, for according to the orthogonal matrix Z, randomly choosing an initial weight W, enabling W*
=E { Zg (WTZ)}-E{g’(WTZ) } W, wherein W* are the adjoint matrix of W, and g is nonlinear function, a1For constant, 1≤a1≤2；
Judging unit 2024, for judging W=W*/||W*| | value whether be 1, obtain judging result；
Output unit 2025, for indicating W=W when judging result*/||W*| | value be 1 when, export W；
Execution unit 2026, for indicating W=W according to judging result*/||W*| | value when not being 1, then execute described
According to the orthogonal matrix, an initial weight W is randomly choosed, W is enabled*=E { Zg (WTZ)}-E{g’(WTZ) } W, wherein W*For W's
Adjoint matrix, g are nonlinear function, a1For constant, 1≤a1≤2。
Specifically, the characteristic extracting module specifically includes：
Sampling unit 2031, for being sampled to the pretreated EEG signals；
Even numbers complex wavelet transform resolving cell 2032, for carrying out even numbers complex wavelet transform point to the EEG signals after sampling
Solution chooses the corresponding frequency range of frequency range of δ waves, θ waves, α waves and β waves；
Reconfiguration unit 2033, for the δ waves, θ waves, α waves and β waves to be reconstructed；
Filter unit 2034, for the frequency range progress cospace pattern filter to δ waves, θ waves, α waves and β waves after the reconstruct
Wave, and then obtain the feature vector of pretreated EEG signals.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with other
The difference of embodiment, just to refer each other for identical similar portion between each embodiment.
Principle and implementation of the present invention are described for specific case used herein, and above example is said
The bright method and its core concept for being merely used to help understand the present invention；Meanwhile for those of ordinary skill in the art, foundation
The thought of the present invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.
Claims (8)
1. a kind of Method of EEG signals classification, which is characterized in that the method includes：
Obtain EEG signals；
EEG signals are pre-processed by independent component analysis；
Feature extraction is carried out to pretreated EEG signals by dual-tree complex wavelet transform and cospace pattern；
Classification processing is carried out to the EEG signals after feature extraction by Multiple Kernel Learning support vector machines.
2. a kind of Method of EEG signals classification according to claim 1, which is characterized in that the acquisition EEG signals are specific
Including：
EEG signals are obtained by mindset.
3. a kind of Method of EEG signals classification according to claim 1, which is characterized in that described to pass through independent component analysis
Pretreatment is carried out to EEG signals to specifically include：
Centralization processing is carried out to the EEG signals；
Whitening processing is carried out to the EEG signals after the centralization, obtains orthogonal matrix Z；
According to the orthogonal matrix Z, an initial weight W is randomly choosed, W is enabled*=E { Zg (WTZ)}-E{g’(WTZ) } W, wherein W*
For the adjoint matrix of W, g is nonlinear function, a1For constant, 1≤a1≤2；
Judge W=W*/||W*| | value whether be 1, obtain judging result, if judging result indicate W=W*/||W*| | value be 1
When, W is exported, if judging result indicates W=W*/||W*| | value when not being 1, then execute it is described according to the orthogonal matrix, at random
An initial weight W is selected, W is enabled*=E { Zg (WTZ)}-E{g,(WTZ) } W, wherein W*For the adjoint matrix of W, g is non-linear letter
Number, a1For constant, 1≤a1≤2。
4. a kind of Method of EEG signals classification according to claim 1, which is characterized in that described to be become by dual-tree complex wavelet
It changes and the progress feature extraction of pretreated EEG signals is specifically included with cospace pattern：
The pretreated EEG signals are sampled；
Even numbers complex wavelet transform decomposition is carried out to the EEG signals after the sampling, chooses the frequency model of δ waves, θ waves, α waves and β waves
Corresponding frequency range is enclosed, and the δ waves, θ waves, α waves and β waves are reconstructed；
To the frequency range progress cospace Pattern Filter of δ waves, θ waves, α waves and β waves after the reconstruct, and then obtain pretreated
The feature vector of EEG signals.
5. a kind of Method of EEG signals classification according to claim 4, which is characterized in that described to described pretreated
EEG signals carry out sampling and specifically include：
According to frequency fsPretreated EEG signals are sampled, fs=30Hz2l, wherein l expression even numbers Phase information changes
The number of plies in changing.
6. a kind of eeg signal classification system, which is characterized in that the system comprises：
Acquisition module, for obtaining EEG signals；
Preprocessing module pre-processes EEG signals for passing through independent component analysis；
Characteristic extracting module, it is special for being carried out to pretreated EEG signals by dual-tree complex wavelet transform and cospace pattern
Sign extraction；
Sort module, for carrying out classification processing to the EEG signals after feature extraction by Multiple Kernel Learning support vector machines.
7. a kind of eeg signal classification system according to claim 6, which is characterized in that the preprocessing module is specifically wrapped
It includes：
Centralization processing unit, for carrying out centralization processing to the EEG signals；
Whitening processing unit obtains orthogonal matrix Z for carrying out whitening processing to the EEG signals after the centralization；
Initial weight acquiring unit, for according to the orthogonal matrix Z, randomly choosing an initial weight W, enabling W*=E { Zg
(WTZ)}-E{g’(WTZ) } W, wherein W*For the adjoint matrix of W, g is nonlinear function, a1For constant, 1≤a1≤2；
Judging unit, for judging W=W*/||W*| | value whether be 1, obtain judging result；
Output unit, for indicating W=W when judging result*/||W*| | value be 1 when, export W；
Execution unit, for indicating W=W according to judging result*/||W*| | value when not being 1, then execute it is described according to it is described just
Matrix is handed over, an initial weight W is randomly choosed, enables W*=E { Zg (WTZ)}-E{g,(WTZ) } W, wherein W*For the adjoint matrix of W,
G is nonlinear function, a1For constant, 1≤a1≤2。
8. a kind of eeg signal classification system according to claim 6, which is characterized in that the characteristic extracting module is specific
Including：
Sampling unit, for being sampled to the pretreated EEG signals；
Even numbers complex wavelet transform resolving cell chooses δ for carrying out even numbers complex wavelet transform decomposition to the EEG signals after sampling
The corresponding frequency range of frequency range of wave, θ waves, α waves and β waves；
Reconfiguration unit, for the δ waves, θ waves, α waves and β waves to be reconstructed；
Filter unit, for the frequency range progress cospace Pattern Filter to δ waves, θ waves, α waves and β waves after the reconstruct, in turn
Obtain the feature vector of pretreated EEG signals.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810317158.XA CN108520239B (en) | 2018-04-10 | 2018-04-10 | Electroencephalogram signal classification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810317158.XA CN108520239B (en) | 2018-04-10 | 2018-04-10 | Electroencephalogram signal classification method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108520239A true CN108520239A (en) | 2018-09-11 |
| CN108520239B CN108520239B (en) | 2021-05-07 |
Family
ID=63430828
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810317158.XA Expired - Fee Related CN108520239B (en) | 2018-04-10 | 2018-04-10 | Electroencephalogram signal classification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108520239B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109711278A (en) * | 2018-12-07 | 2019-05-03 | 浙江大学 | A kind of the eeg signal compression and classification method of low complex degree |
| CN109784233A (en) * | 2018-12-29 | 2019-05-21 | 佛山科学技术学院 | A kind of locking phase value weighted space filtering method and device based on Lp- norm |
| CN111035459A (en) * | 2020-01-03 | 2020-04-21 | 新疆医科大学第一附属医院 | Neurosurgical head support positioning control system and method |
| CN112450950A (en) * | 2020-12-10 | 2021-03-09 | 南京航空航天大学 | Brain-computer aided analysis method and system for aviation accident |
| CN113011493A (en) * | 2021-03-18 | 2021-06-22 | 华南理工大学 | Electroencephalogram emotion classification method, device, medium and equipment based on multi-kernel width learning |
| CN113220127A (en) * | 2021-05-25 | 2021-08-06 | 南昌达甘科技有限公司 | Learning system based on brain wave control |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006072150A1 (en) * | 2005-01-07 | 2006-07-13 | K.U. Leuven Research And Development | Muscle artifact removal from encephalograms |
| US20080262371A1 (en) * | 2004-09-16 | 2008-10-23 | Elvir Causevic | Method for Adaptive Complex Wavelet Based Filtering of Eeg Signals |
| CN101596101A (en) * | 2009-07-13 | 2009-12-09 | 北京工业大学 | Judge the method for fatigue state according to EEG signals |
| CN104091172A (en) * | 2014-07-04 | 2014-10-08 | 北京工业大学 | Characteristic extraction method of motor imagery electroencephalogram signals |
| US20150018704A1 (en) * | 2013-07-10 | 2015-01-15 | National Chiao Tung University | Real-time multi-channel automatic eye blink artifact eliminator |
| CN106600140A (en) * | 2016-12-07 | 2017-04-26 | 河海大学常州校区 | Improved support vector machine-based gas pipeline fault prediction and early warning system and method |
| CN107157477A (en) * | 2017-05-24 | 2017-09-15 | 上海交通大学 | EEG signals Feature Recognition System and method |
| CN107260166A (en) * | 2017-05-26 | 2017-10-20 | 昆明理工大学 | A kind of electric artefact elimination method of practical online brain |
-
2018
- 2018-04-10 CN CN201810317158.XA patent/CN108520239B/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080262371A1 (en) * | 2004-09-16 | 2008-10-23 | Elvir Causevic | Method for Adaptive Complex Wavelet Based Filtering of Eeg Signals |
| WO2006072150A1 (en) * | 2005-01-07 | 2006-07-13 | K.U. Leuven Research And Development | Muscle artifact removal from encephalograms |
| CN101596101A (en) * | 2009-07-13 | 2009-12-09 | 北京工业大学 | Judge the method for fatigue state according to EEG signals |
| US20150018704A1 (en) * | 2013-07-10 | 2015-01-15 | National Chiao Tung University | Real-time multi-channel automatic eye blink artifact eliminator |
| CN104091172A (en) * | 2014-07-04 | 2014-10-08 | 北京工业大学 | Characteristic extraction method of motor imagery electroencephalogram signals |
| CN106600140A (en) * | 2016-12-07 | 2017-04-26 | 河海大学常州校区 | Improved support vector machine-based gas pipeline fault prediction and early warning system and method |
| CN107157477A (en) * | 2017-05-24 | 2017-09-15 | 上海交通大学 | EEG signals Feature Recognition System and method |
| CN107260166A (en) * | 2017-05-26 | 2017-10-20 | 昆明理工大学 | A kind of electric artefact elimination method of practical online brain |
Non-Patent Citations (2)
| Title |
|---|
| 昌风玲: "多类运动想象脑电模式识别及其在电动轮椅控制上的应用", 《中国优秀硕士学位论文全文数据库信息科技辑（月刊）》 * |
| 王超: "基于独立分量分析的脑电信号的分离", 《中国优秀硕士学位论文全文数据库信息科技辑（月刊）》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109711278A (en) * | 2018-12-07 | 2019-05-03 | 浙江大学 | A kind of the eeg signal compression and classification method of low complex degree |
| CN109711278B (en) * | 2018-12-07 | 2020-06-23 | 浙江大学 | Low-complexity brain wave signal compression and classification method |
| CN109784233A (en) * | 2018-12-29 | 2019-05-21 | 佛山科学技术学院 | A kind of locking phase value weighted space filtering method and device based on Lp- norm |
| CN111035459A (en) * | 2020-01-03 | 2020-04-21 | 新疆医科大学第一附属医院 | Neurosurgical head support positioning control system and method |
| CN112450950A (en) * | 2020-12-10 | 2021-03-09 | 南京航空航天大学 | Brain-computer aided analysis method and system for aviation accident |
| CN113011493A (en) * | 2021-03-18 | 2021-06-22 | 华南理工大学 | Electroencephalogram emotion classification method, device, medium and equipment based on multi-kernel width learning |
| CN113220127A (en) * | 2021-05-25 | 2021-08-06 | 南昌达甘科技有限公司 | Learning system based on brain wave control |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108520239B (en) | 2021-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108520239A (en) | A kind of Method of EEG signals classification and system | |
| Zhang et al. | Motor imagery recognition with automatic EEG channel selection and deep learning | |
| Tang et al. | Single-trial EEG classification of motor imagery using deep convolutional neural networks | |
| Jin et al. | An improved P300 pattern in BCI to catch user’s attention | |
| Pfurtscheller et al. | Discrimination of motor imagery‐induced EEG patterns in patients with complete spinal cord injury | |
| Wang et al. | Analysis and classification of hybrid BCI based on motor imagery and speech imagery | |
| Tang et al. | Single-trial classification of different movements on one arm based on ERD/ERS and corticomuscular coherence | |
| CN102613972A (en) | Extraction method of characteristics of electroencephalogram signals based on motor imagery | |
| CN101352337A (en) | Method for capturing signals and extracting characteristics of stand imagination action brain wave | |
| CN106362287A (en) | Novel MI-SSSEP mixed brain-computer interface method and system thereof | |
| Yao et al. | A novel calibration and task guidance framework for motor imagery BCI via a tendon vibration induced sensation with kinesthesia illusion | |
| CN108280414A (en) | A kind of recognition methods of the Mental imagery EEG signals based on energy feature | |
| Gao et al. | Multi-ganglion ANN based feature learning with application to P300-BCI signal classification | |
| Zheng et al. | Concurrent prediction of finger forces based on source separation and classification of neuron discharge information | |
| Delisle-Rodriguez et al. | EEG changes during passive movements improve the motor imagery feature extraction in BCIs-based sensory feedback calibration | |
| Wei et al. | A novel multi-dimensional features fusion algorithm for the EEG signal recognition of brain's sensorimotor region activated tasks | |
| Geng et al. | [Retracted] A Fusion Algorithm for EEG Signal Processing Based on Motor Imagery Brain‐Computer Interface | |
| Mao et al. | Effects of Skin Friction on Tactile P300 Brain‐Computer Interface Performance | |
| Li et al. | A study of action difference on motor imagery based on delayed matching posture task | |
| Hu et al. | A Survey on Brain-Computer Interface-Inspired Communications: Opportunities and Challenges | |
| Risangtuni et al. | Towards online application of wireless EEG-based open platform Brain Computer Interface | |
| Yu et al. | The research of sEMG movement pattern classification based on multiple fused wavelet function | |
| CN207101480U (en) | Upper limbs ectoskeleton control system based on Mental imagery | |
| Petoku et al. | Object movement motor imagery for EEG based BCI system using convolutional neural networks | |
| Hu et al. | A Multi-feature Fusion Transformer Neural Network for Motor Imagery EEG Signal Classification. |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210507 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |