CN102499676B - Effective time sequence and electrode recombination based electroencephalograph signal categorizing system and method - Google Patents
Effective time sequence and electrode recombination based electroencephalograph signal categorizing system and method Download PDFInfo
- Publication number
- CN102499676B CN102499676B CN201110344311.6A CN201110344311A CN102499676B CN 102499676 B CN102499676 B CN 102499676B CN 201110344311 A CN201110344311 A CN 201110344311A CN 102499676 B CN102499676 B CN 102499676B
- Authority
- CN
- China
- Prior art keywords
- eeg signals
- eeg
- module
- signals
- electrode
- 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
Images
Abstract
The invention discloses an effective time sequence and electrode recombination based electroencephalograph signal categorizing system and a method, which can realize the identification to three types of character expression (happiness, normality and sadness) by acquiring and analyzing the electroencephalograph signals of the human brain, and mainly comprises an electroencephalograph signal acquisition process and an electroencephalograph signal analyzing process. According to the invention, the electroencephalograph signals are acquired by stimulating different expression of a testee; the feature space of the effective electroencephalograph signals is determined by the strong energy distribution of the full field of the electroencephalograph signals; then, the PCA (Principal Component Analysis) dimension reduction is carried out to the original electroencephalograph signals corresponding to the feature space, and the electroencephalograph signals with categorizing advantages are reconstructed; and finally, a linear discriminant function categorizer is selected to for categorizing. According to the invention, during expression identification, only target features are extracted from the acquired electroencephalograph signals, then, the acquired electroencephalograph signals are categorized, and the identification result can be determined; and the identification of the electroencephalograph signals can be realized based on the character expression stimulation. In the invention, the cognition of the human being is introduced, and the advantages of objectiveness and high efficiency are provided.
Description
Technical field
The present invention relates to a kind of processing and analytical method of EEG signals, particularly relate to a kind of eeg signal classification system and method based on effective time sequence and electrode restructuring.
Background technology
Social day by day closely at current this inter personal contact, the expression of correctly identifying other people has important significance of existence.This not only can make people regulate in time factum to conform, and can also effectively avoid unnecessary danger, is conducive to social communication and environmental adaptation.Meanwhile, normal person's research also be can be to clinical diagnosis and treatment provides reference, for prevention and treatment work.At present, the main application of human face expression recognition technology comprises man-machine interaction, safety, robot building, medical treatment, communication and automotive field etc.
In the document of relevant Expression Recognition, mainly by image Expression Recognition and speech signal analysis, judge expression, but the traditional method of these expression assessments has subjectivity, is easy to be denied by other people.Yet another kind of available Expression Recognition way is physiology electroencephalogramsignal signal analyzing, it is Expression Recognition means more directly perceived, effective, because expression state was reflected by neural activity originally.
Summary of the invention
The object of the invention is to stimulates for expression the EEG signals producing, and proposes a kind of eeg signal classification system and method based on effective time sequence and electrode restructuring.To avoid the subjectivity of feature, by adopting parallel computation strategy to improve execution efficiency.
The present invention adopts following technological means to realize:
An eeg signal classification system for effective time sequence and electrode restructuring, comprising: electroencephalogramsignal signal acquisition module, EEG signals pretreatment module, EEG signals feature selection module, EEG signals expression classification are implemented module.
EEG signals information acquisition module, gathers tested original EEG signals under glad, neutral and sad difference expression stimulates, and the EEG signals collecting is passed to EEG signals pretreatment module; EEG signals pretreatment module is carried out denoising (noise comprises level eye electricity and vertical eye electricity) by the original EEG signals collecting, and afterwards pure EEG signals is converted to overall field intensity and sends into brain electrical feature selection module; EEG signals feature selection module, by the peak value feature of overall field intensity EEG signals, determine the effective time region of EEG signals, in the EEG signals on effective time region, carry out electrode restructuring, by the EEG signals dimensionality reduction backsight after restructuring, be the final feature of EEG signals Expression Recognition, and this feature is sent to EEG signals expression classification enforcement module; EEG signals expression classification is implemented classical sorting algorithm (Fisher grader) for module and is carried out eeg signal classification.
A Method of EEG signals classification based on effective time sequence and electrode restructuring, comprises the following steps:
Step 5, in order to reduce the redundancy of EEG signals, carries out dimensionality reduction to the EEG signals after the resulting restructuring of step 4 by principal component analysis (PCA) method;
Step 6, the linear discriminant function grader (Fisher) that the EEG signals after characteristic extracting module is extracted is used EEG signals expression classification to implement in module carries out classification learning and test;
During test Expression Recognition, pass through electroencephalogramsignal signal acquisition module, gather tested EEG signals to be measured, EEG signals is sent into EEG signals pretreatment module, remove after noise, according to EEG feature extraction module, calculate and generate tested characteristic of correspondence vector again, then this characteristic vector is sent into EEG signals expression classification and implemented module, finally obtain the eeg signal classification result that expression stimulates.
The selection in the effective time region of EEG signals, comes definite effective time region according to the peak value of overall field intensity and high-energy value; The process of electrode restructuring is on the basis in effective time region, the process that Different electrodes is reconfigured; The process of EEG signals expression classification is on parallel basis, and the EEG signals feature of selecting is classified.
A kind of eeg signal classification system and method based on effective time sequence and electrode restructuring of the present invention, compared with prior art has the following advantages:
1, compare with traditional method, the present invention utilizes physiology EEG signals, has avoided the subjectivity of feature.
2, in step (3), according to the whole audience of EEG signals, to carry out by force feature selection be a kind of reasonable and effective new method in the present invention.
3, the principal component analytical method that the present invention uses in step (5) is the classical way in statistical learning, can find the implementation algorithm of comparative maturity in many numerical computations platforms.
4, main amount of calculation of the present invention concentrates on step (6), owing to can producing multiple combination of electrodes in step (4), therefore step (6) will be carried out grader training and evaluation to the brain electrical feature under every kind of combination, therefore can adopt parallel computation strategy to improve execution efficiency.
Accompanying drawing explanation
Fig. 1 is flow chart and the system module partition situation of method overall process involved in the present invention;
Fig. 2 is the experimental design flow chart of collection EEG signals involved in the present invention;
Fig. 3 is involved in the present invention based on the strong EEG signals figure of the whole audience;
Fig. 4 is the idiographic flow of " EEG Processing " part in Fig. 1 of the present invention.
The specific embodiment
Below in conjunction with the specific embodiment, the present invention is described further.
The step of the present invention when training Expression Recognition grader has following 6 steps:
First in step 1, according to the experiment designing, carry out the collection of EEG signals, in the process gathering in test, select three class human face expressions as stimulating picture, comprise glad expression, neutral expression and sad expression, each expression has 18 kinds of shapes of face, each subjects carries out 408 tests, and three generic tasks respectively account for 136 times.Each process of the test is as follows: first give subjects's display reminding language, after subjects presses space bar, show a forward or swing to, expression is the picture of one of glad, neutrality, sad three kinds of expressions, until subjects, expression is known and distinguished and press after corresponding button, represent that single test finishes, detailed process as shown in Figure 2.Tested data acquisition from 12 ages at the 20-30 Healthy People in year.
Next original EEG signals step 1 being collected is carried out pretreatment, and pretreatment module comprises 2 steps:
Step 2.1, because EEG signals is faint, is very easily subject to the impact of electro-ocular signal.Therefore, the noise of removing in EEG signals just seems particularly important, utilizes NeuroScan software to carry out denoising to the EEG signals collecting in the inventive method.
Step 2.2 obtains on the basis of clean EEG signals EEG in step 2.1, obtains the whole audience corresponding to the original EEG signals of 66 conductive electrode strong by NeuroScan software, i.e. GFP, and it is to obtain by the signal of each electrode is carried out to superposed average.
In step 4, in characteristic extracting module, we adopt heuristics method, and heuristic function is the performance of linear discriminant function grader, and we analyze the effectiveness of restructuring with performance quality.First by electrode, recombinated EEG signals feature selection is done to further extraction, in the determined validity feature of step 3 region, recombinate, the process of restructuring is to adopt exhaustive mode to carry out permutation and combination according to different time region and Different electrodes.Formula is as follows:
i∈{1,…,64},j∈{1,…,5},
Wherein, i is number of poles, and j is time series section number, E
ijrefer to the EEG signals of j time period of i electrode, work as α
i, represent that i unipolar EEG signals is used as feature, works as β at=1 o'clock
j, represent that the EEG signals of j time zone is used as feature at=1 o'clock.
Next in step 5 pair step 4, various combination of electrodes is carried out principal component analytical method dimensionality reduction.Because when EEG signals feature selection carries out, after electrode restructuring, all can producing the EEG signals of higher-dimension, thereby affected classification effectiveness and the result of EEG signals, so that the reduction process of EEG signals seems is particularly important.This patent drops to 400 dimensions EEG signals feature.
The linear discriminant function grader that EEG signals after step 6 pair dimensionality reduction is used EEG signals expression classification to implement in module is afterwards classified, because the process of EEG signals restructuring in step 4 is selected exhaustive mode, experience is known, exhaustive method has brought computer memory large, the problem that complexity is high, therefore, here we are for addressing this problem introducing parallel computation, parallel mainly for categorizing process, thus by concurrent operation, greatly improved classification time and speed.As Fig. 4, the idiographic flow of detailed step display 2-6.
The step of the present invention when test Expression Recognition is as follows:
Pass through electroencephalogramsignal signal acquisition module, gather tested EEG signals to be measured (method is consistent with above-mentioned corresponding step), EEG signals is sent into EEG signals pretreatment module, remove after noise, according to EEG feature extraction module, calculate and generate tested characteristic of correspondence vector (method is consistent with above-mentioned corresponding step) again, then this characteristic vector is sent into EEG signals expression classification and implemented module, finally generate the classification results stimulating based on expression.Result shows, the highest discrimination has surpassed 90%, and average recognition rate, in 85% left and right, can realize the EEG's Recognition that expression is stimulated.
Claims (5)
1. the eeg signal classification system based on effective time sequence and electrode restructuring, comprising: electroencephalogramsignal signal acquisition module, EEG signals pretreatment module, EEG feature extraction module, EEG signals expression classification are implemented module; It is characterized in that: described electroencephalogramsignal signal acquisition module, is mainly used to gather the original EEG signals under different expression stimulations, and the original EEG signals collecting is passed to EEG signals pretreatment module;
Described EEG signals pretreatment module is carried out denoising by the original EEG signals collecting, and converts pure EEG signals to overall field intensity afterwards and sends into characteristic extracting module;
Described overall field intensity is the superposed average value of all electrode EEG signals;
Described EEG feature extraction module, by the EEG signals overall situation peak value of field intensity and the time of energy, distribute and determine the effective time region of EEG signals, again the EEG signals on effective time region is carried out to electrode restructuring, the process of restructuring is to adopt exhaustive mode to carry out permutation and combination according to different time region and Different electrodes, EEG signals after restructuring is considered as to the feature of EEG signals Expression Recognition, and feature is sent to EEG signals expression classification enforcement module;
Described EEG signals expression classification is implemented module and is carried out eeg signal classification by classical sorting algorithm.
2. the eeg signal classification system based on effective time sequence and electrode restructuring according to claim 1, is characterized in that: described classical sorting algorithm adopts Fisher grader.
3. the Method of EEG signals classification based on the restructuring of effective time sequence and electrode, is characterized in that: comprise the following steps:
Step 1, experimenter is with upper electrode cap, and original EEG signals is to gather by amplifier, and chooses all electrode positions, gathers experimenter's EEG signals of different expression stimulating courses;
Step 2, is input to EEG feature extraction module by the EEG signals after EEG signals pretreatment module is processed, and pretreatment module comprises that the EEG signals to collecting carries out denoising, and obtains the overall field intensity of EEG signals;
Step 3, the analysis according to characteristic extracting module to EEG signals overall situation field intensity, determines the effective time region of EEG feature extraction;
Step 4, carries out electrode restructuring according to characteristic extracting module to the EEG signals of feature selection;
Step 5, carries out dimensionality reduction to the EEG signals after the resulting restructuring of step 4 by principal component analytical method;
Step 6, the linear discriminant function grader that the EEG signals after characteristic extracting module is extracted is used EEG signals expression classification to implement in module carries out classification learning and tests finally obtaining classification results.
4. the Method of EEG signals classification based on the restructuring of effective time sequence and electrode according to claim 3, the EEG amplifier of 10/20 method that the international electroencephalography of described amplifier Wei66Dao can be demarcated.
5. the Method of EEG signals classification based on the restructuring of effective time sequence and electrode according to claim 3, is characterized in that:
The selection in the effective time region of described EEG signals, refers to according to the peak value of overall field intensity and high-energy value and comes definite effective time region;
The process of described electrode restructuring is on the basis in effective time region, adopts exhaustive mode according to different time region and Different electrodes, to carry out the process of permutation and combination;
The process of described EEG signals expression classification is on parallel basis, and the EEG signals feature of selecting is classified.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110344311.6A CN102499676B (en) | 2011-11-03 | 2011-11-03 | Effective time sequence and electrode recombination based electroencephalograph signal categorizing system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110344311.6A CN102499676B (en) | 2011-11-03 | 2011-11-03 | Effective time sequence and electrode recombination based electroencephalograph signal categorizing system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102499676A CN102499676A (en) | 2012-06-20 |
CN102499676B true CN102499676B (en) | 2014-01-29 |
Family
ID=46211858
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201110344311.6A Active CN102499676B (en) | 2011-11-03 | 2011-11-03 | Effective time sequence and electrode recombination based electroencephalograph signal categorizing system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102499676B (en) |
Families Citing this family (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103413050B (en) * | 2013-08-20 | 2016-08-24 | 北京工业大学 | Mental imagery EEG signals temporal voting strategy sorting technique based on very fast learning machine |
CN103654773B (en) * | 2013-12-20 | 2016-02-03 | 北京飞宇星电子科技有限公司 | Electroencephalo experimental teaching unit |
CN103750844B (en) * | 2014-01-15 | 2015-07-29 | 杭州电子科技大学 | A kind of based on the phase locked personal identification method of brain electricity |
CN103876734B (en) * | 2014-03-24 | 2015-09-02 | 北京工业大学 | A kind of EEG signals feature selection approach based on decision tree |
CN104127179B (en) * | 2014-04-13 | 2016-04-06 | 北京工业大学 | The brain electrical feature extracting method of a kind of advantage combination of electrodes and empirical mode decomposition |
CN103971124B (en) * | 2014-05-04 | 2017-02-15 | 杭州电子科技大学 | Multi-class motor imagery brain electrical signal classification method based on phase synchronization |
CN104361345A (en) * | 2014-10-10 | 2015-02-18 | 北京工业大学 | Electroencephalogram signal classification method based on constrained extreme learning machine |
CN106338935A (en) * | 2015-04-23 | 2017-01-18 | 恒爱高科(北京)科技有限公司 | Robot emotion recognition method and system |
CN105395192A (en) * | 2015-12-09 | 2016-03-16 | 恒爱高科(北京)科技有限公司 | Wearable emotion recognition method and system based on electroencephalogram |
CN105894039A (en) * | 2016-04-25 | 2016-08-24 | 京东方科技集团股份有限公司 | Emotion recognition modeling method, emotion recognition method and apparatus, and intelligent device |
CN106725452A (en) * | 2016-11-29 | 2017-05-31 | 太原理工大学 | Based on the EEG signal identification method that emotion induces |
CN106778186A (en) * | 2017-02-14 | 2017-05-31 | 南方科技大学 | A kind of personal identification method and device for virtual reality interactive device |
CN109044350A (en) * | 2018-09-15 | 2018-12-21 | 哈尔滨理工大学 | A kind of eeg signal acquisition device and detection method |
CN111543988B (en) * | 2020-05-25 | 2021-06-08 | 五邑大学 | Adaptive cognitive activity recognition method and device and storage medium |
CN112241952B (en) * | 2020-10-22 | 2023-09-05 | 平安科技(深圳)有限公司 | Brain midline identification method, device, computer equipment and storage medium |
CN113197551B (en) * | 2021-05-07 | 2023-08-04 | 中国医学科学院生物医学工程研究所 | Multimode physiological nerve signal detection and experimental stimulation time alignment method |
CN113197573B (en) * | 2021-05-19 | 2022-06-17 | 哈尔滨工业大学 | Film watching impression detection method based on expression recognition and electroencephalogram fusion |
CN114021605A (en) * | 2021-11-02 | 2022-02-08 | 深圳市大数据研究院 | Risk prediction method, device and system, computer equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6266624B1 (en) * | 1996-03-19 | 2001-07-24 | Siemens Aktiengesellschaft | Method conducted in a computer for classification of a time series having a prescribable number of samples |
CN101339455A (en) * | 2008-08-07 | 2009-01-07 | 北京师范大学 | Brain machine interface system based on human face recognition specific wave N170 component |
CN101862194A (en) * | 2010-06-17 | 2010-10-20 | 天津大学 | Imagination action EEG identification method based on fusion feature |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6175762B1 (en) * | 1996-04-10 | 2001-01-16 | University Of Technology, Sydney | EEG based activation system |
-
2011
- 2011-11-03 CN CN201110344311.6A patent/CN102499676B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6266624B1 (en) * | 1996-03-19 | 2001-07-24 | Siemens Aktiengesellschaft | Method conducted in a computer for classification of a time series having a prescribable number of samples |
CN101339455A (en) * | 2008-08-07 | 2009-01-07 | 北京师范大学 | Brain machine interface system based on human face recognition specific wave N170 component |
CN101862194A (en) * | 2010-06-17 | 2010-10-20 | 天津大学 | Imagination action EEG identification method based on fusion feature |
Non-Patent Citations (1)
Title |
---|
JP特表2000-507867A 2000.06.27 |
Also Published As
Publication number | Publication date |
---|---|
CN102499676A (en) | 2012-06-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102499676B (en) | Effective time sequence and electrode recombination based electroencephalograph signal categorizing system and method | |
CN104586387B (en) | Method for extracting and fusing time, frequency and space domain multi-parameter electroencephalogram characters | |
CN101474070B (en) | Method for removing ocular artifacts in brain-electrical signal | |
CN110070105B (en) | Electroencephalogram emotion recognition method and system based on meta-learning example rapid screening | |
EP2027559B1 (en) | Method of processing multichannel and multivariate signals and method of classifying sources of multichannel and multivariate signals operating according to such processing method | |
Yang et al. | Comparison among driving state prediction models for car-following condition based on EEG and driving features | |
AlSharabi et al. | EEG signal processing for Alzheimer’s disorders using discrete wavelet transform and machine learning approaches | |
CN103610447A (en) | Mental workload online detection method based on forehead electroencephalogram signals | |
CN1883384A (en) | A method for automatically detecting and removing artifacts from EEG signal | |
CN111523601A (en) | Latent emotion recognition method based on knowledge guidance and generation counterstudy | |
CN112674782B (en) | Device and method for detecting epileptic-like electrical activity of epileptic during inter-seizure period | |
Prasanth et al. | Deep learning for interictal epileptiform spike detection from scalp EEG frequency sub bands | |
CN112884063B (en) | P300 signal detection and identification method based on multi-element space-time convolution neural network | |
CN113951903B (en) | High-speed railway dispatcher overload state identification method based on electroencephalogram data determination | |
Yang et al. | On the effectiveness of EEG signals as a source of biometric information | |
Parsaei et al. | SVM-based validation of motor unit potential trains extracted by EMG signal decomposition | |
CN108523883A (en) | A kind of continuous Mental imagery identifying system of left and right index finger based on actual act modeling | |
CN113017627A (en) | Depression and bipolar disorder brain network analysis method based on two-channel phase synchronization feature fusion | |
CN101433460B (en) | Spatial filtering method of lower limb imaginary action potential | |
CN106073767B (en) | Phase synchronization measurement, coupling feature extraction and the signal recognition method of EEG signal | |
CN113397482B (en) | Human behavior analysis method and system | |
CN111281382B (en) | Feature extraction and classification method based on electroencephalogram signals | |
Ahmed et al. | Effective hybrid method for the detection and rejection of electrooculogram (EOG) and power line noise artefacts from electroencephalogram (EEG) mixtures | |
Quatieri et al. | Using eeg to discriminate cognitive workload and performance based on neural activation and connectivity | |
CN116712086A (en) | Hidden Markov-based brain micro-state resolution method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |