CN107437011A - The method and apparatus of identification based on EEG signals - Google Patents

The method and apparatus of identification based on EEG signals Download PDF

Info

Publication number
CN107437011A
CN107437011A CN201610361544.XA CN201610361544A CN107437011A CN 107437011 A CN107437011 A CN 107437011A CN 201610361544 A CN201610361544 A CN 201610361544A CN 107437011 A CN107437011 A CN 107437011A
Authority
CN
China
Prior art keywords
sections
signals
ssvep signals
ssvep
frequency
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
Application number
CN201610361544.XA
Other languages
Chinese (zh)
Inventor
袁鹏
薛希俊
姚骏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CN201610361544.XA priority Critical patent/CN107437011A/en
Priority to PCT/CN2016/104445 priority patent/WO2017201972A1/en
Publication of CN107437011A publication Critical patent/CN107437011A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31—User authentication
    • G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Electrotherapy Devices (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

The present embodiments relate to the method and apparatus of the identification based on EEG signals.This method includes：Determine goal stimulus frequency sequence；N section stimulus signals corresponding to the goal stimulus frequency sequence are shown for user to be detected；The user to be detected is obtained due to n sections Steady State Visual Evoked Potential SSVEP signals caused by the n section stimulus signals；When the similarity that the n section SSVEP signals and n sections preset SSVEP signals is more than or equal to threshold value, determine that the user identity to be detected is correct；Otherwise, it determines the user identity mistake to be detected.The method and apparatus of the identification based on EEG signals of the embodiment of the present invention, identification is carried out by SSVEP signals, it can shorten stimulation duration compared to existing brain electricity personal identification method, signal characteristic is relatively more stable, identification system based on this structure more maintains secrecy, and is not easy to be replicated forgery.

Description

The method and apparatus of identification based on EEG signals
Technical field
The present invention relates to areas of information technology, the method and apparatus of the identification based on EEG signals.
Background technology
In the epoch interconnected into all things on earth, due to the exponential growth of the smart machine number of connection, information security is asked Topic becomes particularly important.In information security field, identification is a technology being widely used.With information security issue More and more important, the research of various personal identification methods also turns into study hotspot.
Traditional personal identification method is typically by key, identity document, username and password and other items or information password To realize.However, this traditional personal identification method is easily stolen, loses or forgets, people can not be fully met Expectation and requirement.Then people start to resort to using biological characteristic to carry out identification.Biometrics identification technology It is often referred to carry out identification using some intrinsic physiological properties of human body or behavioural characteristic.The physiological property of human body is typically wrapped Include：Face, fingerprint, palm shape, iris etc.；The behavioural characteristic of people may include：Notes, gait etc..
However, existing biometrics identification technology is also faced with some problems at present.For example, for being carried out by fingerprint Identification, the prosthetic finger can made of gelatin are successfully out-tricked fingerprint recognition system；For carrying out identity by iris Identification, the false iris feature etched on contact lenses can allow iris authentication system not distinguish the true from the false.These problems New challenge is proposed to biometrics identification technology, also inspires people constantly to explore new biological feather recognition method.
In recent years, people start to consider to apply among identification using brain electricity as a kind of new biological characteristic.Grind Study carefully and show, even if under same outside stimulus or people think deeply it is same the problem of when, the brain of different subjects is induced Caused EEG signals are also different, i.e. brain electricity has significant individual difference.At the same time, brain electricity, which has, is difficult to replicate With forge, can be by numerous advantages such as main body autonomous notice modulation.Therefore, occur at present a series of based on various pattern brains electricity Personal identification method.Such as personal identification method based on resting electroencephalogramidentification, the identity based on imagination motion state brain electrical feature are known Other method, brain electricity personal identification method based on P300 event related potentials etc..
But above-mentioned EEG signals signal to noise ratio used, typically than relatively low, signal characteristic is not sufficiently stable, it usually needs collection is more The EEG signals and the substantial amounts of training sample of needs of individual lead, it is not convenient enough to use.For example, the identity based on resting electroencephalogramidentification Recognition methods, the spontaneous brain electricity of quiescent condition has the non-stationary of height, also easily influenceed by individual state, intraindividual Mobility is larger；And in test, user generally requires the multiple electrodes collection EEG signals for covering full brain, and user is not using Just.For another example the brain electricity personal identification method based on P300 event related potentials, the signal to noise ratio of P300 event related potentials is very It is low, it is necessary to can just obtain stable waveform after substantial amounts of repetitive stimulation and superposed average, certification duration is longer, inconvenient for use； Also, P300 event related potentials involve the higher cognitive effect of people, are induced more by novel stimulus, easily by the essence of user agent Refreshing state influences, and prolonged stimulate can allow user adaptation reaction occur, causes the P300 event related potentials decay induced, It is unfavorable for the extraction of signal characteristic, poor reliability.
The content of the invention
This application provides a kind of method and apparatus of the identification based on EEG signals, can be lured according to stable state vision Generating position SSVEP signals carry out identification, improve the reliability of identification.
First aspect, there is provided a kind of method of the identification based on EEG signals, this method include：Determine that target is pierced Swash frequency sequence [f1,f2,f3,……,fn], n is positive integer；The goal stimulus frequency sequence [f is shown for user to be detected1, f2,f3,……,fn] corresponding to n section stimulus signals, the display frequency of i-th section of stimulus signal is the target in the n section stimulus signals Frequency of stimulation sequence [f1,f2,f3,……,fn] in i-th of frequency fi, i takes 1,2,3 ... n；Obtain the user to be detected due to N sections Steady State Visual Evoked Potential SSVEP signals caused by the n section stimulus signals；When the n section SSVEP signals are preset with n sections When the similarity of SSVEP signals is more than or equal to threshold value, determine that the user identity to be detected is correct；When the n section SSVEP signals with When the similarity that the n sections preset SSVEP signals is less than the threshold value, the user identity mistake to be detected is determined.
Therefore, the method for the identification based on EEG signals of the application, it is to be checked according to goal stimulus frequency sequence Survey user and show stimulus signal, so as to gather SSVEP signals caused by user to be detected, by the SSVEP signals and default SSVEP Signal is contrasted, and carries out the identification of user to be detected.Because the characteristics of SSVEP signal high s/n ratios, can cause signal It is more readily detected, can shortens stimulation duration compared to existing brain electricity personal identification method；SSVEP signals are concentrated mainly on people's Brain occipital region, it is only necessary to which less electrode, such as an electrode can collect the signal of abundant information amount, easy to use； The signal that SSVEP signals induce as a kind of primary visual cortex, it is not necessary to which the higher cognitive activity of people participates in, therefore it is by people The state of mind influence it is smaller, signal characteristic is relatively more stable；SSVEP signals abundant amplitude-frequency response and phase frequency Response can be directly corresponding with the system physiological property of people's primary visual cortex so that the identification system based on this structure is more The secrecy added, it is not easy to be replicated forgery.
It should be understood that carrying out identification using SSVEP signals, the SSVEP signals are pierced by stable periodic vision Swash and induce human brain generation, that is, the either default SSVEP signals of the prior typing of original user, or user to be detected is in body Collected SSVEP signals during part identification, it is required for stable periodic visual to stimulate to induce human brain generation SSVEP signals. It is thus necessary to determine that the frequency of at least one visual stimulus.
It should be understood that during collection SSVEP signals, user can be allowed to take wearable dry electrode cap, the electrode cap comprises at least One dry electrode being placed at user occipital region scalp, in order to gather the SSVEP signals of user.In the SSVEP letters of collection user Number when, user is with eye gaze stimulus signal and keeps notice.Now, the SSVEP signals that user's brain occipital region induces Collected by wearable brain electric equipment (i.e. wearable dry electrode cap).The SSVEP signals collected can be transmitted wirelessly to phase The signal analysis region answered, identified for subsequent analysis processing.
With reference in a first aspect, in a kind of implementation of first aspect, the determination goal stimulus frequency sequence [f1,f2, f3,……,fn], including：Obtain the frequency of stimulation sequence [f to be detected of user's input to be detected1′,f2′,f3′,……, fn′]；As the frequency of stimulation sequence [f to be detected1′,f2′,f3′,……,fn'] and the goal stimulus frequency sequence [f1,f2, f3,……,fn] it is identical when, obtain the goal stimulus frequency sequence [f1,f2,f3,……,fn] corresponding to the n section stimulus signals； As the frequency of stimulation sequence [f to be detected1′,f2′,f3′,……,fn'] and the goal stimulus frequency sequence [f1,f2,f3,……, fn] it is different when, determine the user identity mistake to be detected.
Therefore, for the goal stimulus frequency sequence of setting, identification is carried out, so, exclusive stimulus sequence adds The pattern of the double certifications of SSVEP signals, improves the grade of equipment secrecy, is less susceptible to be broken into.
Alternatively, the goal stimulus frequency sequence [f that user is set1,f2,f3,……,fn] in each frequency scope Typically in 6Hz between 100Hz, and the n frequency can arrange according to size order, or random alignment, in the n frequency At least two identical frequencies can be included.
Alternatively, due to the goal stimulus frequency sequence [f1,f2,f3,……,fn] in each frequency can be in sequence Arrangement or random alignment, therefore, it is determined that frequency of stimulation sequence [f1′,f2′,f3′,……,fn'] and goal stimulus sequence [f1,f2,f3,……,fn] it is identical when, can include determining that the order of accordingly each frequency in two sequences is identical, corresponding The size of each frequency is identical and two sequences in frequency number it is identical.
Alternatively, can not also consider it is above-mentioned on whole factors in order, size and number, it is defeated in the user to be detected Enter frequency of stimulation sequence [f1′,f2′,f3′,……,fn'] it is goal stimulus sequence [f1,f2,f3,……,fn] subset when, User's input stimulus frequency sequence [f to be detected can be determined1′,f2′,f3′,……,fn'] and goal stimulus sequence [f1,f2, f3,……,fn] identical.
For example, user's input stimulus frequency sequence [f to be detected1′,f2′,f3′,……,fn'], it is determined that frequency of stimulation sequence Arrange [f1′,f2′,f3′,……,fn'] and goal stimulus sequence [f1,f2,f3,……,fn] whether it is identical when, can not consider defeated Enter order, i.e., it is identical and incoming frequency with respective frequencies size in goal stimulus sequence in the size of each frequency of input When number and also identical goal stimulus sequence, frequency of stimulation sequence [f is determined1′,f2′,f3′,……,fn'] and goal stimulus sequence Arrange [f1,f2,f3,……,fn] identical.
For another example user's input stimulus frequency sequence [f to be detected1′,f2′,f3′,……,fn'], it is determined that stimulating frequency Rate sequence [f1′,f2′,f3′,……,fn'] and goal stimulus sequence [f1,f2,f3,……,fn] whether mutually at the same time it can also not Consider input number, i.e., the size in each frequency of input is identical with respective frequencies size in goal stimulus sequence and inputs The order of frequency with when frequency order is also identical in goal stimulus sequence, or only input each frequency size and target Respective frequencies size is identical in stimulus sequence, can determine that frequency of stimulation sequence [f1′,f2′,f3′,……,fn'] pierced with target Swash sequence [f1,f2,f3,……,fn] identical.
It should be understood that the goal stimulus frequency sequence [f1,f2,f3,……,fn] can also not as identification foundation, when When user to be detected needs to carry out identification, the goal stimulus frequency set in advance by user is obtained automatically by identity recognition device Rate sequence [f1,f2,f3,……,fn], or obtain or obtained in order in advance by a series of frequencies of user's setting at random Component frequency, form goal stimulus frequency sequence [f1,f2,f3,……,fn], pass through the goal stimulus frequency sequence got [f1,f2,f3,……,fn] stimulus signal is shown to user to be detected, and user to be detected is gathered because caused by the stimulus signal SSVEP signals, carry out the identification of SSVEP signals.
It is use to be detected in another implementation of first aspect with reference to first aspect and its above-mentioned implementation Family shows the goal stimulus frequency sequence [f1,f2,f3,……,fn] corresponding to n section stimulus signals, including：For user to be detected N section stimulus signals are shown, wherein, first paragraph stimulus signal is with f1A period of time is shown for frequency, second segment stimulus signal is with f2For Frequency display a period of time, the 3rd section of stimulus signal is with f3A period of time is shown for frequency, until n-th section of stimulus signal is with fn A period of time is shown for frequency.
It should be understood that after showing above-mentioned n sections stimulus signal for user to be detected, then user to be detected can be collected according to this Corresponding n sections SSVEP signals caused by n section stimulus signals.Likewise, when user sets default SSVEP signals, and basis The goal stimulus frequency sequence [f is shown successively1,f2,f3,……,fn] corresponding to n section stimulus signals, produce n sections SSVEP letter Number, gather the n section SSVEP signals and preserved as default SSVEP signals.
It should be understood that the stimulus signal can be the picture with certain pattern, the picture is according to corresponding frequency fiDisplay； Or the stimulus signal can also be light, the alternate frequency of switch or light and shade of light is display frequency fi；Or the thorn Energizing signal can also be other any forms that can be induced human brain and produce SSVEP signals, pass through frequency fiShown to user, with It is easy to obtain the user to be detected for SSVEP signals caused by the reaction of stimulus signal.
With reference to first aspect and its above-mentioned implementation, in another implementation of first aspect, this method is also wrapped Include：SSVEP signals are preset with the n sections in corresponding at least one domain according to the n section SSVEP signals at least one domain, it is determined that The n section SSVEP signals preset the coefficient correlation vector of SSVEP signals with the n sectionsCoefficient correlation vectorIn each member Element represents that the n section SSVEP signals preset the coefficient correlation of SSVEP signals with the n sections；WillIt is defined as n sections SSVEP letters The similarity of SSVEP signals number is preset with the n sections, wherein,Represent weight parameter vector, weight parameter vectorIn it is every Individual element representation coefficient correlation vectorThe weighted value of middle corresponding element.
Specifically, weight parameter vectorIn each element, corresponding to coefficient correlation vectorIn each element representation The weight of coefficient correlation.For example, the weight parameter is vectorialIn each element be disposed as being equal toWherein,Represent coefficient correlation vectorThe number of the element included.Or when coefficient correlation vectorIn some members When element is relatively small, illustrate the real user reflected and the distinction of non-real real user that the element can be stronger, therefore can It is bigger than normal to modulate weighted value corresponding to the element.
It should be understood that presetting the similarity of SSVEP signals according to the n section SSVEP signals of the user to be detected and n sections, carry out Identification.When the similarity is more than or equal to threshold value, illustrates that the user identity to be detected is correct, i.e., be based on by this The identification of SSVEP signals；When the similarity is less than threshold value, illustrates the user identity mistake to be detected, i.e., can not pass through This identification based on SSVEP signals.Alternatively, the threshold value can be configured according to actual conditions, and can root According to multiple test, the threshold value is constantly updated.
With reference to first aspect and its above-mentioned implementation, in another implementation of first aspect, the determination n sections SSVEP signals preset the coefficient correlation vector of SSVEP signals with the n sectionsIncluding：I-th section is determined in the n section SSVEP signals SSVEP signals Xi(fi,Ti) with the n sections preset the X in SSVEP signalsi′(fi,Ti) time domain coefficient correlation be rX(fi), i takes 1, 2、3……n；According to the time domain correlation coefficient rX(fi), determine coefficient correlation vector
With reference to first aspect and its above-mentioned implementation, in another implementation of first aspect, the determination n sections SSVEP signals preset the coefficient correlation vector of SSVEP signals with the n sectionsIncluding：By i-th section in the n section SSVEP signals SSVEP signals Xi(fi,Ti) enter line translation and obtain Zi(fi,Yi)；The n sections are preset into the X in SSVEP signalsi′(fi,Ti) become Get Z in returni′(fi,Yi)；Determine Zi(fi,Yi) and Zi′(fi,Yi) transform domain coefficient correlation be rY(fi), i takes 1,2,3 ... n； According to the transform domain correlation coefficient rY(fi), determine coefficient correlation vector
With reference to first aspect and its above-mentioned implementation, in another implementation of first aspect, the determination n sections SSVEP signals preset the coefficient correlation vector of SSVEP signals with the n sectionsIncluding：By i-th section in the n section SSVEP signals SSVEP signals Xi(fi,Ti) carry out Fourier transformation obtain Zi(fi,Fi)；The n sections are preset into the X in SSVEP signalsi′(fi,Ti) Carry out Fourier transformation and obtain Zi′(fi,Fi)；Determine Zi(fi,Fi) and Zi′(fi,Fi) frequency domain correlation coefficient be rF(fi), i takes 1、2、3……n；According to frequency domain correlation coefficient rF(fi), determine coefficient correlation vector
It should be understood that when the SSVEP signals collected are the signals of multiple electrodes, Xi(fi,Ti) or Xi′(fi,Ti) can Think a matrix.At this point it is possible to the classical Method of Data with Adding Windows such as principal component analysis (PCA) or canonical correlation analysis (CCA) is used, By Xi(fi,Ti) or Xi′(fi,Ti) it is reduced to one-dimensional time-domain signal.The application is with Xi(fi,Ti) and Xi′(fi,Ti) it is one-dimensional time domain Illustrated exemplified by signal.
It should be understood that the transform domain coefficient correlation is rY(fi) can be r including frequency domain correlation coefficientF(fi)。
Alternatively, the time domain coefficient correlation is rX(fi) and transform domain coefficient correlation be rY(fi), can be linear correlation Coefficient, for example, Pearson's linearly dependent coefficient.
Alternatively, according to n time domain correlation coefficient rX(fi) and/or n transform domain correlation coefficient rY(fi), it is determined that related Coefficient vectorCoefficient correlation vectorIn element can represent time domain coefficient correlation, frequency domain phase relation can also be represented Number.For example, the coefficient correlation is vectorialN element can be included, the n element can include i time domain correlation coefficient rX (fi), and including remaining (n-i) individual time domain correlation coefficient rX(fi) corresponding to (n-i) individual transform domain coefficient correlation be rY(fi)； For another example the 2n element includes n time domain correlation coefficient rX(fi) and n transform domain coefficient correlation be rY(fi)。
Second aspect, there is provided a kind of device of the identification based on EEG signals, for performing above-mentioned first aspect Or the method in any possible implementation of first aspect.Specifically, the device includes being used to perform above-mentioned first aspect Or the unit of the method in any possible implementation of first aspect.
The third aspect, there is provided a kind of device of the identification based on EEG signals, including：Memory cell and processing Device, the memory cell are used for store instruction, and the processor is used for the instruction for performing the memory storage, and when the processor is held During the instruction of the row memory storage, the execution causes any possible reality of the computing device first aspect or first aspect Method in existing mode.
Fourth aspect, there is provided a kind of computer-readable medium, for storing computer program, the computer program includes The instruction of the method in any possible implementation for performing first aspect or first aspect.
Brief description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, it will make below to required in the embodiment of the present invention Accompanying drawing is briefly described, it should be apparent that, drawings described below is only some embodiments of the present invention, for For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings Accompanying drawing.
Fig. 1 is the indicative flowchart of the method for the identification according to embodiments of the present invention based on EEG signals.
Fig. 2 is the schematic block diagram of the device of the identification according to embodiments of the present invention based on EEG signals.
Fig. 3 is the schematic block diagram of the device of the identification according to another embodiment of the present invention based on EEG signals.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete Site preparation describes, it is clear that described embodiment is the part of the embodiment of the present invention, rather than whole embodiments.Based on this hair Embodiment in bright, the every other reality that those of ordinary skill in the art are obtained on the premise of creative work is not made Example is applied, should all belong to the scope of protection of the invention.
Fig. 1 shows the schematic stream of the method 100 of the identification according to embodiments of the present invention based on EEG signals Cheng Tu.This method 100 can be by needing the equipment for carrying out authentication to perform, for example, the identification dress installed on safety cabinet Authentication can be carried out with the method 100 by putting.As shown in figure 1, this method 100 includes：
S110, determine goal stimulus frequency sequence [f1,f2,f3,……,fn], n is positive integer.
It should be understood that the method for the identification based on EEG signals of the embodiment of the present invention, need in advance gather and preserve use The EEG signals at family, when carrying out authentication or identification, gather the brain electricity of user to be detected as default EEG signals Signal, the EEG signals are contrasted with the default EEG signals preserved, if being consistent with default EEG signals, illustrated to be detected User identity is correct, and authentication passes through；If not being consistent with default brain electricity, illustrate user identity mistake to be detected, authentication It can not pass through.
Specifically, the EEG signals of progress of embodiment of the present invention authentication collection refer to Steady State Visual Evoked Potential (Steady-State Visual Evoked Potentials, SSVEP) signal, identity knowledge is carried out according to the SSVEP signals Not.The SSVEP signals typically induce human brain by stable periodic visual stimulus and produced, typically can in the occipital region of the brain of people Recorded the SSVEP signals.Consist predominantly of the brain electricity composition with stimulus signal same frequency in the SSVEP signals, while by Also comprising each harmonic composition and other frequencies in the nonlinear characteristic of vision system and the influence of spontaneous brain electricity, its signal Rate composition.Due to the high complexity and otherness in the physiological brain structure of people, for example, the difference of spontaneous background brain electricity, depending on Feel the delay variance, the response difference of vision system etc. of path, there is very big difference for the SSVEP signals that different people induces.Phase It is compared to other EEG signals, such as event related potential (Event Related Potentials, ERP) signal, or vision Evoked ptential (Visual Evoked Potentials, VEP) signal, SSVEP signals are more stable and signal to noise ratio is higher.
In embodiments of the present invention, SSVEP signals are to induce human brain by stable periodic visual stimulus to produce, also It is the default SSVEP signals of either user's typing, or the SSVEP signals that user to be detected is collected, is required for stable Periodic visual stimulates produces SSVEP signals to induce human brain.It is thus necessary to determine that the frequency of at least one visual stimulus.
Specifically, for user when setting default SSVEP signals, it is necessary to first selection target frequency of stimulation sequence [f1,f2, f3,……,fn], n is positive integer, and each frequency corresponds to the display frequency of one section of stimulus signal in the goal stimulus frequency sequence Rate, for example, the frequency f in goal stimulus frequency sequenceiThe display frequency for referring to i-th section of stimulus signal is fi, so as to obtain user For the default SSVEP signals of i-th section of stimulus signal.It should be understood that original user when user here refers to identification, example Such as, identification is carried out during opening safety cabinet, the holder of the safety cabinet can think the user in the embodiment of the present invention, should User can set foundation of the default SSVEP signals as identification.
Alternatively, the goal stimulus frequency sequence [f that user is set1,f2,f3,……,fn] in each frequency scope Typically in 6Hz between 100Hz, and the n frequency can arrange according to size order, or random alignment, for example, user can To set the goal stimulus frequency sequence, as [20,57,38,64], the embodiment of the present invention is not limited to this.Alternatively, this n frequency At least two identical frequency values can be included in rate, i.e. the n frequency values can repeat.
In embodiments of the present invention, user to be detected first determines the goal stimulus frequency when needing to carry out identification Sequence [f1,f2,f3,……,fn].Because the goal stimulus frequency sequence can be set in advance by user, therefore can be according to the mesh Mark frequency of stimulation sequence first carries out an identification.Specifically, user's input stimulus frequency sequence [f to be detected1′,f2′, f3′,……,fn'], as the frequency of stimulation sequence [f1′,f2′,f3′,……,fn'] and goal stimulus sequence [f1,f2, f3,……,fn] it is identical when, illustrate the user identity identification to be detected by the way that the identity on SSVEP signals can be continued Identification；As the frequency of stimulation sequence [f1′,f2′,f3′,……,fn'] and goal stimulus sequence [f1,f2,f3,……,fn] different When, illustrate that the user identity identification to be detected can not be by identification failure.Alternatively, it is determined that during goal stimulus sequence When having determined user identity identification failure to be detected, it still can continue the identification subsequently with respect to SSVEP signals, still Recognition result is identity mistake, can not be passed through；Or the identification on SSVEP signals after can also no longer carrying out, Directly determine identification failure.
Alternatively, due to the goal stimulus frequency sequence [f1,f2,f3,……,fn] in each frequency can be in sequence Arrangement or random alignment, therefore, it is determined that frequency of stimulation sequence [f1′,f2′,f3′,……,fn'] and goal stimulus sequence [f1,f2,f3,……,fn] whether it is identical when, can include determining that each frequency and goal stimulus frequency of user to be detected input The size frequency corresponding with goal stimulus frequency sequence for each frequency that whether order of respective frequencies is identical in rate sequence, inputs Whether the size of rate identical and number of incoming frequency and frequency number in goal stimulus frequency sequence it is whether identical.For example, During user's Initialize installation identification system, user chooses [12,8,20] Hz as goal stimulus sequence, and by three frequencies Oneself true EEG signals X (12, t), X (8, t) under stimulating are saved in safety box identification system with X (20, t).When When user to be detected asks authentication, system requirements user's input stimulus frequency sequence instruction to be detected is bonded beam with #.If treat Detect user and input non-setpoint frequency size sequences such as [6,12,15#], or the sequence of the sequence error such as input [12,20,8#] Row, or the sequence of the number mistake such as input [12,8#], illustrate subscriber authentication failure to be detected, terminate identification.Only Have when user correctly enters [12,8,20#], then illustrating user to be detected, this is tested on the identity of frequency of stimulation sequence inputting Demonstrate,prove successfully, the authentication on SSVEP signals can be continued.
Furthermore it is also possible to not consider above-mentioned on whole factors in order, size and number, inputted in the user to be detected Frequency of stimulation sequence [f1′,f2′,f3′,……,fn'] it is goal stimulus sequence [f1,f2,f3,……,fn] subset when, also may be used To determine user's input stimulus frequency sequence [f to be detected1′,f2′,f3′,……,fn'] and goal stimulus sequence [f1,f2, f3,……,fn] identical.
Alternatively, user's input stimulus frequency sequence [f to be detected1′,f2′,f3′,……,fn'], it is determined that frequency of stimulation Sequence [f1′,f2′,f3′,……,fn'] and goal stimulus sequence [f1,f2,f3,……,fn] whether it is identical when, can not consider Input sequence, i.e., input each frequency size with respective frequencies size in goal stimulus sequence identical and incoming frequency Number and goal stimulus sequence it is also identical when, determine frequency of stimulation sequence [f1′,f2′,f3′,……,fn'] and goal stimulus Sequence [f1,f2,f3,……,fn] identical.Specifically, during user's Initialize installation identification system, user's selection [12,8, 20] Hz is as goal stimulus sequence, and by oneself true EEG signals X (12, t), X (8, t) and the X under three frequency stimulations (20, t) are saved in safety box identification system.When user to be detected asks authentication, system requirements use to be detected Family input stimulus frequency sequence instruction is bonded beam with #.If user to be detected inputs the big forewords of non-setpoint frequency such as [6,12,15#] Row, or the sequence of the number mistake such as input [12,8#], illustrate subscriber authentication failure to be detected, terminate identification.When When user correctly enters [12,8,20#], or input [12,20,8#] etc. is only the sequence of sequence error, it may be said that bright This is successful on the authentication of frequency of stimulation sequence inputting by user to be detected, can continue on SSVEP signals Authentication.
Alternatively, user's input stimulus frequency sequence [f to be detected1′,f2′,f3′,……,fn'], it is determined that frequency of stimulation Sequence [f1′,f2′,f3′,……,fn'] and goal stimulus sequence [f1,f2,f3,……,fn] whether mutually at the same time it can also not examining Consider input number, i.e., the size in each frequency of input is identical with respective frequencies size in goal stimulus sequence and inputs frequency The order of rate in size and the target of each frequency of input with when frequency order is also identical in goal stimulus sequence, or only piercing Respective frequencies size is identical in sharp sequence, can determine that frequency of stimulation sequence [f1′,f2′,f3′,……,fn'] and goal stimulus Sequence [f1,f2,f3,……,fn] identical.Specifically, user can be set a large amount of when setting initial frequency of stimulation sequence Frequency, such as set 10 frequencies to form initial impulse frequency sequence., can be only when user to be detected will carry out authentication The frequency of stimulation sequence that a small amount of frequency is formed is inputted, if the frequency that the frequency of stimulation sequence includes belongs to the initial of user's setting Frequency of stimulation sequence when, otherwise identification is not by passing through.
For example, user sets the initial impulse frequency sequence that 10 frequencies are formed, when user to be detected carries out identification When, input stimulus frequency sequence [f1′,f2′,f3′,……,fn'] it is [12,8,20], if searching initial impulse frequency sequence, really When including three frequencies surely, that is, goal stimulus frequency sequence [f be present1,f2,f3,……,fn] it is [12,8,20], the target Stimulus sequence is the subset for 10 frequency sets that user is set, should [f1′,f2′,f3′,……,fn'] and [f1,f2, f3,……,fn] identical, then identification is by continuing the identification on SSVEP signals；If search initial impulse Frequency sequence, it is determined that when not including three frequencies or only including wherein one or two frequency, for example, only existing goal stimulus Frequency sequence [f1,f2,f3,……,fn] it is [12,8], then should [f not including frequency 20Hz1′,f2′,f3′,……,fn'] with [f1,f2,f3,……,fn] different, identification is not by that can terminate identification.
For another example in the above-described example, in order to reduce error, can also further consideration order it is whether correct.Likewise, User sets the initial impulse frequency sequence that 10 frequencies are formed, when user to be detected carries out identification, input stimulus frequency Rate sequence [f1′,f2′,f3′,……,fn'] it is [12,8,20], if searching initial impulse frequency sequence, it is determined that including continuous Three frequencies, and when three frequency orders are identical, i.e. a band frequency sequence, i.e. goal stimulus frequency be present in 10 frequencies Sequence [f1,f2,f3,……,fn] it is [12,8,20], should [f1′,f2′,f3′,……,fn'] and [f1,f2,f3,……,fn] phase Together, then identification by continuing the identification on SSVEP signals；If searching initial impulse frequency sequence, it is determined that Do not include continuous three frequencies, or only include wherein one or two frequency, or when order is different, for example, only existing mesh Mark frequency of stimulation sequence [f1,f2,f3,……,fn] it is [12,8], not including frequency 20Hz, or goal stimulus frequency sequence be present For [12,20,8], i.e., input sequence is different, or exist goal stimulus frequency sequence be [12,11,20,8], i.e. input sequence not Continuously, then all explanation should [f1′,f2′,f3′,……,fn'] and [f1,f2,f3,……,fn] different, identification is not by knot Beam identification.
In embodiments of the present invention, the goal stimulus frequency sequence [f1,f2,f3,……,fn] can also be not as identity Basis of characterization, when user to be detected needs to carry out identification, obtained by identification system and set in advance by user automatically Goal stimulus frequency sequence [f1,f2,f3,……,fn], or obtain at random or obtain what is set in advance by user in order A series of component frequency in frequencies, form goal stimulus frequency sequence [f1,f2,f3,……,fn], pass through the target got Frequency of stimulation sequence [f1,f2,f3,……,fn] to the identification to be detected for being used to carry out SSVEP signals.
S120, the goal stimulus frequency sequence [f is shown for user to be detected1,f2,f3,……,fn] corresponding to n sections stimulate Signal, the display frequency of i-th section of stimulus signal is the goal stimulus frequency sequence [f in the n section stimulus signals1,f2,f3,……, fn] in i-th of frequency fi, i takes 1,2,3 ... n.
S130, the user to be detected is obtained due to n section Steady State Visual Evoked Potentials caused by the n sections stimulus signal SSVEP signals.
In embodiments of the present invention, when user to be detected carries out identification, goal stimulus frequency sequence [f is determined1,f2, f3,……,fn] after, according to the goal stimulus frequency sequence [f1,f2,f3,……,fn] it is that user to be detected shows that n sections are right successively The stimulus signal answered, the display frequency of i-th section of stimulus signal is the goal stimulus frequency sequence [f in the n section stimulus signals1,f2, f3,……,fn] in i-th of frequency fi, i takes 1,2,3 ... n successively.Specifically, first paragraph stimulus signal is with f1Shown for frequency For a period of time, then second segment stimulus signal with f2A period of time is shown for frequency, the 3rd section of stimulus signal is with f3Shown for frequency For a period of time, until n-th section of stimulus signal is with fnFor frequency show a period of time, then can collect user to be detected according to Corresponding n sections SSVEP signals caused by the n section stimulus signals.
It should be understood that similar, user is when setting default SSVEP signals, and according to showing goal stimulus frequency successively Rate sequence [f1,f2,f3,……,fn] corresponding to n section stimulus signals, produce n section SSVEP signals, gather the n section SSVEP signals Preserved as default SSVEP signals, in order to be according to the SSVEP signals of default SSVEP signal detections user to be detected It is no correct, so as to carry out authentication.
Alternatively, the display frequency of the stimulus signal shown for the original user of user to be detected or setting SSVEP signals For fi, stimulus signal here can be the picture with certain pattern, and the picture is according to corresponding frequency fiDisplay；Or should Stimulus signal can also be light, and the alternate frequency of switch or light and shade of light is display frequency fi；Or the stimulus signal It can also be other any forms that can be induced human brain and produce SSVEP signals, pass through frequency fiShown to user, in order to obtain The user to be detected or original user are taken for SSVEP signals caused by the reaction of stimulus signal.
It should be understood that the stimulus signal type for setting the original user of default SSVEP signals to use uses with user to be detected Stimulus signal type it is consistent, can so avoid due to stimulus signal different band come error.
It should be understood that capture setting preset SSVEP signals original user or user to be detected SSVEP signals when, can be with User is allowed to take wearable dry electrode cap, the electrode cap comprises at least a dry electrode being placed at user occipital region scalp, so as to In the SSVEP signals of collection user.Gathering user to be detected or original user SSVEP signals according to caused by stimulus signal When, user is with eye gaze stimulus signal and keeps notice.Now, the SSVEP signals that user's brain occipital region induces can lead to Wearable brain electric equipment (i.e. wearable dry electrode cap) is crossed to collect.The SSVEP signals collected can be transmitted wirelessly to accordingly Signal analysis region, for subsequent analysis processing identify.
In embodiments of the present invention, when every section of stimulus signal starts, can be sent out to the corresponding module of collection SSVEP signals Send synchronizing information, in order to when gathering SSVEP signals, can according to corresponding to the synchronizing signal intercepts different frequency every section of thorn The SSVEP signals of energizing signal, i.e. n sections stimulus signal correspond to n section SSVEP signals, can be intercepted out respectively by the synchronizing signal The n section SSVEP signals.
S140, when the similarity of the n section SSVEP signals and the default SSVEP signals of n sections is more than or equal to threshold value, it is determined that The user identity to be detected is correct；The similarity that SSVEP signals are preset when the n section SSVEP signals and the n sections is less than the threshold value When, determine the user identity mistake to be detected.
In embodiments of the present invention, letter is stimulated for the n section SSVEP signals collected, corresponding one of every section of SSVEP signal Number frequency, i.e. i-th section of SSVEP signal can be expressed as Xi(fi,Ti), fiRepresent that i-th section of SSVEP signal is according to display Frequency is fiStimulus signal obtain, i.e. goal stimulus frequency sequence [f1,f2,f3,……,fn] corresponding n sections user to be detected SSVEP signals X1(f1,T1)、X2(f2,T2)……Xn(fn,Tn).Correspondingly, the default SSVEP signals that user is set in advance can To be expressed as Xi′(fi,Ti), i.e. goal stimulus frequency sequence [f1,f2,f3,……,fn] correspond to the default SSVEP signals X of n sections1′ (f1,T1)、X2′(f2,T2)……Xn′(fn,Tn)。
It should be understood that when the SSVEP signals collected are the signals of multiple electrodes, Xi(fi,Ti) or Xi′(fi,Ti) can Think a matrix.At this point it is possible to the classical Method of Data with Adding Windows such as principal component analysis (PCA) or canonical correlation analysis (CCA) is used, By Xi(fi,Ti) or Xi′(fi,Ti) it is reduced to one-dimensional time-domain signal.Below by Xi(fi,Ti) and Xi′(fi,Ti) it is considered as one-dimensional time domain Handled exemplified by signal.
In embodiments of the present invention, the SSVEP signals of the user to be detected collected and default SSVEP signals are carried out pair Than analysis, and then determine whether user identity to be detected is correct.Specifically, can be by calculating the n sections of the user to be detected SSVEP signals preset the similarity of SSVEP signals with n sections, determine identification result.When the n section SSVEP signals and n sections are pre- If the similarity of SSVEP signals is more than or equal to threshold value, illustrates that the identity of user to be detected is correct, then can be known by identity Not；When the similarity that the n section SSVEP signals and n sections preset SSVEP signals is less than threshold value, illustrate the body of the user to be detected Part mistake, identification do not pass through.For example, by the identification opening safety cabinet, then identification passes through, you can to open Safety cabinet；But identification is obstructed out-of-date, then cannot opening safety cabinet, it is and further, obstructed out-of-date in identification, Warning device can also be started.
It should be understood that the similarity of the two can be calculated by the coefficient correlation of n section SSVEP signals and n section SSVEP signals. It is alternatively possible to the coefficient correlation of SSVEP signals is preset by time domain angle calculation n section SSVEP signals and n sections.Specifically, may be used To calculate i-th section of SSVEP signals Xi(fi,Ti) and i-th section of default SSVEP signals Xi′(fi,Ti) time domain correlation coefficient rX(fi), The coefficient correlation can be linearly dependent coefficient, for example, Pearson's linearly dependent coefficient.Pearson's linearly dependent coefficient, can be with For weighing the similarity degree between input signal and real user signal.Under identical frequency of stimulation, different people by In the difference of physiological brain structure, for example, pathways for vision delay is different, spectral response difference of vision system etc., it is each lured There is larger difference for meeting between the SSVEP signals of hair.Therefore, only when the SSVEP signals of input are real users During EEG signals, coefficient correlation just can be larger.
Alternatively, the transform domain of SSVEP signals can also be preset by transform domain angle calculation n section SSVEP signals and n sections Coefficient correlation.Specifically, can be by i-th section of SSVEP signals X in n section SSVEP signalsi(fi,Ti) enter line translation and obtain Zi(fi, Yi)；N sections are preset into the X in SSVEP signalsi′(fi,Ti) enter line translation and obtain Zi′(fi,Yi)；Calculate and determine Zi(fi,Yi) and Zi′ (fi,Yi) transform domain coefficient correlation be rY(fi), i takes 1,2,3 ... n；According to the transform domain correlation coefficient rY(fi), it is determined that The coefficient correlation vector
It should be understood that transform domain refers to other domains different from time domain (T domains), for example, frequency domain (i.e. F domains), S domains, Z domains etc.. Can be by algorithm by the SSVEP signals X of time domaini(fi,Ti) transform domain is transformed to, obtain Zi(fi,Yi)；For example, it can pass through Fourier algorithm is by SSVEP signals Xi(fi,Ti) frequency domain is transformed to, obtain Zi(fi,Fi), fast Fourier algorithm can also be passed through By SSVEP signals Xi(fi, t) and frequency domain is transformed to, obtain another Zi(fi,Fi)；For another example will by Laplacian algorithm SSVEP signals Xi(fi, t) and S domains are transformed to, obtain Zi(fi,Si)。
For example, the transform domain correlation coefficient rY(fi) frequency domain correlation coefficient r can be includedF(fi), you can to pass through frequency domain angle Degree calculates the frequency domain correlation coefficient that n section SSVEP signals preset SSVEP signals with n sections.For example, to calculate the similar of amplitude-frequency response Exemplified by property.Specifically, by i-th section of SSVEP signals Xi(fi,Ti) Fourier transformation is carried out, obtain its corresponding amplitude-frequency response signal Zi(fi,Fi), likewise, by i-th section of default SSVEP signals Xi′(fi,Ti) Fourier transformation is also carried out, obtain Zi′(fi,Fi), Calculate Zi(fi,Fi) and Zi′(fi,Fi) frequency domain correlation coefficient be rF(fi).Similar, n section SSVEP signals and n can also be calculated The similarity factor of the phase-frequency response of the default SSVEP signals of section, the embodiment of the present invention are not limited to this.
In embodiments of the present invention, the time domain correlation coefficient r that can be determined according to above two methodX(fi) and conversion Domain coefficient correlation is rY(fi), determine that the n section SSVEP signals preset the similarity of SSVEP signals with n sections.Specifically, Ke Yigen According to n time domain correlation coefficient rX(fi) and n transform domain coefficient correlation be rY(fi), determine coefficient correlation vectorThe correlation Coefficient vectorIn element can represent time domain coefficient correlation, transform domain coefficient correlation can also be represented, wherein, the transform domain Coefficient correlation can be frequency domain correlation coefficient, or can also be other coefficient correlations.For example, the coefficient correlation is vectorialCan be with Including n element, the n element can be n time domain correlation coefficient rX(fi), or n transform domain coefficient correlation is rY (fi), or, the n element can include i time domain correlation coefficient rX(fi), and including remaining (n-i) individual time domain phase relation Number rX(fi) corresponding to (n-i) individual transform domain coefficient correlation be rY(fi)；For another example the coefficient correlation is vectorial2n can also be included Individual element, the 2n element include n time domain correlation coefficient rX(fi) and n transform domain coefficient correlation be rY(fi), the present invention It is not limited to this.
In embodiments of the present invention, according to coefficient correlation vectorDetermine that the n section SSVEP signals are preset with n sections The similarity of SSVEP signals.Specifically, formula can be passed throughDetermine that n section SSVEP signals preset SSVEP signals with n sections Similarity, wherein,Weight parameter vector is represented, i.e. n section SSVEP signals and n sections presets similarity of SSVEP signals etc. In weight parameter vectorTransposition and coefficient correlation vectorInner product.
Specifically, weight parameter vectorIn each element, corresponding to coefficient correlation vectorIn each element representation The weight of coefficient correlation.For example, the weight parameter is vectorialIn each element be disposed as being equal toWherein,Represent coefficient correlation vectorThe number of the element included.Or when statistics shows coefficient correlation vector In some element, then can will when can preferably reflect the difference of real user and non-real real user compared to other elements Weighted value corresponding to the element tunes up.
It should be understood that the element here can preferably reflect the difference of real user and non-real real user, refer to non- When real user is compared with the coefficient correlation of original real user, the difference of the element is larger compared to other elements.Tool Body, original user can carry out multi collect when setting default SSVEP signals, it is determined that after default SSVEP signals, every time The SSVEP signals of the original user collected can determine a coefficient correlation vector with default SSVEP signalsEnter Row multi collect, you can to determine multigroup coefficient correlation vectorMultigroup coefficient correlation vectorAverage value can be used as original use The original coefficient correlation vector at familySimilitude of the original user with every section of default SSVEP signal in itself can be reflected, i.e.,. Correspondingly, when user to be detected carries out identification, determine the SSVEP signals of user to be detected and default SSVEP signals it Between coefficient correlation vectorWhen statistics shows non-real real user's coefficient correlation vectorIn some element and original phase relation Number vectorThe difference of corresponding element, more than other elements and original coefficient correlation vectorDifference between corresponding element, Illustrate the original user reflected and the distinction of user to be detected that the element can be stronger, therefore the element pair can be modulated The weighted value answered is bigger than normal.
For example, when user to be detected carries out identification, goal stimulus frequency sequence is [12,8], then during corresponding calculating Domain correlation coefficient r (12) and r (8), make coefficient correlation vectorialIn only include two coefficient correlations, i.e. r (12) and r (8), then It can setAlternatively, if statistics shows the non-real real user's of calculating In r (12) and r (8), r (12) and true original user correlation coefficient value R (12) difference are bigger, then can change weighted value, OrderWherein it is determined that the r (12) differs greatly and referred to, for example, original user is being set When putting default SSVEP signals, it is determined that R (12) and R (8) be 0.9, but calculate user r (8) to be detected be 0.5, r (12) it is more than r (8) and R (8) difference 0.4 for 0.2, r (12) and R (12) difference 0.7, then it is corresponding can increases the r (12) Weight, evenBut if likewise, the r (12) of user to be detected be 0.2, But original user is when setting default SSVEP signals, it is determined that R (12) be 0.8, and the r (8) of user to be detected is 0.3, former Beginning user when setting default SSVEP signals, it is determined that R (8) be 0.9, now, the r (12) of user to be detected differs with R (12) It is also 0.6 that 0.6, r (8) differ with R (8), therefore can not increase the r (12) of the user to be detected weighted value, is made
In embodiments of the present invention, the phase of SSVEP signals is preset with n sections according to the n section SSVEP signals of the user to be detected Like degree, identification is carried out.When the similarity is more than or equal to threshold value, illustrates that the user identity to be detected is correct, that is, pass through This identification based on SSVEP signals；When the similarity is less than threshold value, illustrate the user identity mistake to be detected, i.e., This identification based on SSVEP signals can not be passed through.Alternatively, the threshold value can be configured according to actual conditions, and And the threshold value can be constantly updated according to multiple test.
It should be understood that in various embodiments of the present invention, the size of the sequence number of above-mentioned each process is not meant to perform suitable The priority of sequence, the execution sequence of each process should be determined with its function and internal logic, without the implementation of the reply embodiment of the present invention Process forms any restriction.
Therefore, the method for the identification based on EEG signals of the embodiment of the present invention, by setting goal stimulus frequency Sequence, it is that user to be detected shows stimulus signal according to certain frequency, so as to gather SSVEP signals caused by user to be detected, The SSVEP signals and default SSVEP signals are contrasted, carry out the identification of user to be detected.Because SSVEP signals are high The characteristics of signal to noise ratio, can cause signal to be more readily detected, when can shorten stimulation compared to existing brain electricity personal identification method It is long；SSVEP signals are concentrated mainly on the brain occipital region of people, it is only necessary to which less electrode, such as an electrode can collect The signal of abundant information amount, it is easy to use；The signal that SSVEP signals induce as a kind of primary visual cortex, it is not necessary to people's Higher cognitive activity is participated in, therefore it is influenceed smaller by the state of mind of people, and signal characteristic is relatively more stable；SSVEP signals are rich Rich amplitude-frequency response and phase-frequency response can be directly corresponding with the system physiological property of people's primary visual cortex, makes It must more be maintained secrecy based on the identification system of this structure, be not easy to be replicated forgery.In addition, the goal stimulus frequency for setting Rate sequence, identification can also be carried out, so, exclusive stimulus sequence adds the pattern of the double certifications of SSVEP signals, improves and sets The grade of standby secrecy, is less susceptible to be broken into.
Above in conjunction with Fig. 1, the side of the identification according to embodiments of the present invention based on EEG signals is described in detail Method, below in conjunction with Fig. 2 to Fig. 3, the device of the identification according to embodiments of the present invention based on EEG signals is described.
As shown in Fig. 2 the device 200 of the identification based on EEG signals according to embodiments of the present invention includes：
Determining unit 210, for determining goal stimulus frequency sequence [f1,f2,f3,……,fn], n is positive integer；
Display unit 220, for showing the goal stimulus frequency sequence [f for user to be detected1,f2,f3,……,fn] right The n section stimulus signals answered, the display frequency of i-th section of stimulus signal is the goal stimulus frequency sequence in the n section stimulus signals [f1,f2,f3,……,fn] in i-th of frequency fi, i takes 1,2,3 ... n；
Acquiring unit 230, for obtaining the user to be detected because n section stable state visions caused by the n section stimulus signals lure Generating position SSVEP signals；
Processing unit 240, for being more than or equal to when the similarity of the n section SSVEP signals and the default SSVEP signals of n sections During threshold value, determine that the user identity to be detected is correct；When the n section SSVEP signals and the n sections preset the similarity of SSVEP signals During less than the threshold value, the user identity mistake to be detected is determined.
Therefore, the device of the identification based on EEG signals of the embodiment of the present invention, according to goal stimulus frequency sequence Stimulus signal is shown for user to be detected, so as to gather SSVEP signals caused by user to be detected, by the SSVEP signals and in advance If SSVEP signals are contrasted, the identification of user to be detected is carried out.The characteristics of due to SSVEP signal high s/n ratios, can be with So that signal is more readily detected, stimulation duration can be shortened compared to existing brain electricity personal identification method；SSVEP signals mainly collect In in the brain occipital region of people, it is only necessary to less electrode, such as an electrode can collect the signal of abundant information amount, make With conveniently；The signal that SSVEP signals induce as a kind of primary visual cortex, it is not necessary to the higher cognitive activity of people participates in, because This its influenceed by the state of mind of people smaller, signal characteristic is relatively more stable；The abundant amplitude-frequency response of SSVEP signals and Phase-frequency response can be directly corresponding with the system physiological property of people's primary visual cortex so that the identity based on this structure is known Other system more maintains secrecy, and is not easy to be replicated forgery.
Alternatively, the determining unit 210 is specifically used for：Obtain the frequency of stimulation sequence to be detected of user's input to be detected [f1′,f2′,f3′,……,fn′]；As the frequency of stimulation sequence [f to be detected1′,f2′,f3′,……,fn'] and the goal stimulus Frequency sequence [f1,f2,f3,……,fn] it is identical when, obtain the goal stimulus frequency sequence [f1,f2,f3,……,fn] corresponding to The n section stimulus signals；As the frequency of stimulation sequence [f to be detected1′,f2′,f3′,……,fn'] and the goal stimulus frequency sequence [f1,f2,f3,……,fn] it is different when, determine the user identity mistake to be detected.
Alternatively, the processing unit 240 is specifically used for：According to the n section SSVEP signals at least one domain with it is corresponding extremely The n sections in a few domain preset SSVEP signals, determine that the n section SSVEP signals preset the coefficient correlation of SSVEP signals with the n sections VectorCoefficient correlation vectorIn each element representation n section SSVEP signals and the n sections preset the phases of SSVEP signals Relation number；WillIt is defined as the similarity that the n section SSVEP signals preset SSVEP signals with the n sections, wherein,Represent power Weight parameter vector, weight parameter vectorIn each element representation coefficient correlation vectorThe weighted value of middle corresponding element.
Alternatively, the processing unit 240 is specifically used for：Determine i-th section of SSVEP signals X in the n section SSVEP signalsi(fi, Ti) with the n sections preset the X in SSVEP signalsi′(fi,Ti) time domain coefficient correlation be rX(fi), i takes 1,2,3 ... n；According to The time domain correlation coefficient rX(fi), determine coefficient correlation vector
Alternatively, the processing unit 240 is specifically used for：By i-th section of SSVEP signals X in the n section SSVEP signalsi(fi,Ti) Enter line translation and obtain Zi(fi,Yi)；The n sections are preset into the X in SSVEP signalsi′(fi,Ti) enter line translation and obtain Zi′(fi,Yi)； Determine Zi(fi,Yi) and Zi′(fi,Yi) transform domain coefficient correlation be rY(fi), i takes 1,2,3 ... n；It is related according to the transform domain Coefficient rY(fi), determine coefficient correlation vector
Alternatively, the processing unit 240 is specifically used for：By i-th section of SSVEP signals X in the n section SSVEP signalsi(fi,Ti) Carry out Fourier transformation and obtain Zi(fi,Fi)；The n sections are preset into the X in SSVEP signalsi′(fi,Ti) carry out Fourier transformation obtain To Zi′(fi,Fi)；Determine Zi(fi,Fi) and Zi′(fi,Fi) frequency domain correlation coefficient be rF(fi), i takes 1,2,3 ... n；According to this Frequency domain correlation coefficient rF(fi), determine coefficient correlation vector
It should be understood that the device 200 of the identification based on EEG signals according to embodiments of the present invention may correspond to perform Method 100 in the embodiment of the present invention, and above and other operation and/or function difference of the modules in device 200 In order to realize the corresponding flow of each method in Fig. 1, for sake of simplicity, will not be repeated here.
Therefore, the device of the identification based on EEG signals of the embodiment of the present invention, according to goal stimulus frequency sequence Stimulus signal is shown for user to be detected, so as to gather SSVEP signals caused by user to be detected, by the SSVEP signals and in advance If SSVEP signals are contrasted, the identification of user to be detected is carried out.The characteristics of due to SSVEP signal high s/n ratios, can be with So that signal is more readily detected, stimulation duration can be shortened compared to existing brain electricity personal identification method；SSVEP signals mainly collect In in the brain occipital region of people, it is only necessary to less electrode, such as an electrode can collect the signal of abundant information amount, make With conveniently；The signal that SSVEP signals induce as a kind of primary visual cortex, it is not necessary to the higher cognitive activity of people participates in, because This its influenceed by the state of mind of people smaller, signal characteristic is relatively more stable；The abundant amplitude-frequency response of SSVEP signals and Phase-frequency response can be directly corresponding with the system physiological property of people's primary visual cortex so that the identity based on this structure is known Other system more maintains secrecy, and is not easy to be replicated forgery.In addition, for the goal stimulus frequency sequence of setting, body can also be carried out Part identification, so, exclusive stimulus sequence add the pattern of the double certifications of SSVEP signals, improve the grade of equipment secrecy, are less susceptible to It is broken into.
As shown in figure 3, the embodiment of the present invention additionally provides a kind of device 300 of the identification based on EEG signals, bag Processor 310 and memory 320 are included, bus system 330 can also be included.Wherein, processor 310 and memory 320 are by total Linear system system 330 is connected, and the memory 320 is used for store instruction, and the processor 310 is used for the finger for performing the memory 320 storage Order.The store program codes of memory 320, and processor 310 program code that is stored in memory 320 can be called to perform with Lower operation：Determine goal stimulus frequency sequence [f1,f2,f3,……,fn], n is positive integer；The target is shown for user to be detected Frequency of stimulation sequence [f1,f2,f3,……,fn] corresponding to n section stimulus signals, i-th section of stimulus signal in the n section stimulus signals Display frequency is the goal stimulus frequency sequence [f1,f2,f3,……,fn] in i-th of frequency fi, i takes 1,2,3 ... n；Obtain The user to be detected is due to n sections Steady State Visual Evoked Potential SSVEP signals caused by the n section stimulus signals；As n sections SSVEP When the similarity of signal and the default SSVEP signals of n sections is more than or equal to threshold value, determine that the user identity to be detected is correct；As the n When the similarity of section SSVEP signals and the default SSVEP signals of the n sections is less than the threshold value, the user identity mistake to be detected is determined.
Therefore, the device of the identification based on EEG signals of the embodiment of the present invention, according to goal stimulus frequency sequence Stimulus signal is shown for user to be detected, so as to gather SSVEP signals caused by user to be detected, by the SSVEP signals and in advance If SSVEP signals are contrasted, the identification of user to be detected is carried out.The characteristics of due to SSVEP signal high s/n ratios, can be with So that signal is more readily detected, stimulation duration can be shortened compared to existing brain electricity personal identification method；SSVEP signals mainly collect In in the brain occipital region of people, it is only necessary to less electrode, such as an electrode can collect the signal of abundant information amount, make With conveniently；The signal that SSVEP signals induce as a kind of primary visual cortex, it is not necessary to the higher cognitive activity of people participates in, because This its influenceed by the state of mind of people smaller, signal characteristic is relatively more stable；The abundant amplitude-frequency response of SSVEP signals and Phase-frequency response can be directly corresponding with the system physiological property of people's primary visual cortex so that the identity based on this structure is known Other system more maintains secrecy, and is not easy to be replicated forgery.
It should be understood that in embodiments of the present invention, the processor 310 can be CPU (Central Processing Unit, referred to as " CPU "), the processor 310 can also be other general processors, digital signal processor (DSP), application specific integrated circuit (ASIC), ready-made programmable gate array (FPGA) or other PLDs, discrete gate Or transistor logic, discrete hardware components etc..General processor can be that microprocessor or the processor can also It is any conventional processor etc..
The memory 320 can include read-only storage and random access memory, and to processor 310 provide instruction and Data.The a part of of memory 320 can also include nonvolatile RAM.For example, memory 320 can also be deposited Store up the information of device type.
The bus system 330 can also include power bus, controlling bus and status signal in addition to including data/address bus Bus etc..But for the sake of clear explanation, various buses are all designated as bus system 330 in figure.
In implementation process, each step of the above method can pass through the integrated logic circuit of the hardware in processor 310 Or the instruction of software form is completed.The step of method with reference to disclosed in the embodiment of the present invention, can be embodied directly at hardware Reason device performs completion, or performs completion with the hardware in processor and software module combination.Software module can be located at random Memory, flash memory, read-only storage, the ability such as programmable read only memory or electrically erasable programmable memory, register In the ripe storage medium in domain.The storage medium is located at memory 320, and processor 310 reads the information in memory 320, knot Close the step of its hardware completes the above method.To avoid repeating, it is not detailed herein.
Alternatively, the processor 310 is used for：Obtain the frequency of stimulation sequence [f to be detected of user's input to be detected1′, f2′,f3′,……,fn′]；As the frequency of stimulation sequence [f to be detected1′,f2′,f3′,……,fn'] and the goal stimulus frequency Sequence [f1,f2,f3,……,fn] it is identical when, obtain the goal stimulus frequency sequence [f1,f2,f3,……,fn] corresponding to the n sections Stimulus signal；As the frequency of stimulation sequence [f to be detected1′,f2′,f3′,……,fn'] and the goal stimulus frequency sequence [f1, f2,f3,……,fn] it is different when, determine the user identity mistake to be detected.
Alternatively, the processor 310 is used for：According to the n section SSVEP signals at least one domain with it is corresponding at least one The n sections in domain preset SSVEP signals, determine that the n section SSVEP signals preset the coefficient correlation vector of SSVEP signals with the n sectionsCoefficient correlation vectorIn each element representation n section SSVEP signals and the n sections preset the phase relations of SSVEP signals Number；WillIt is defined as the similarity that the n section SSVEP signals preset SSVEP signals with the n sections, wherein,Represent weight ginseng Number vector, weight parameter vectorIn each element representation coefficient correlation vectorThe weighted value of middle corresponding element.
Alternatively, the processor 310 is used for：Determine i-th section of SSVEP signals X in the n section SSVEP signalsi(fi,Ti) with being somebody's turn to do N sections preset the X in SSVEP signalsi′(fi,Ti) time domain coefficient correlation be rX(fi), i takes 1,2,3 ... n；According to the time domain phase Relation number rX(fi), determine coefficient correlation vector
Alternatively, the processor 310 is used for：By i-th section of SSVEP signals X in the n section SSVEP signalsi(fi,Ti) become Get Z in returni(fi,Yi)；The n sections are preset into the X in SSVEP signalsi′(fi,Ti) enter line translation and obtain Zi′(fi,Yi)；Determine Zi (fi,Yi) and Zi′(fi,Yi) transform domain coefficient correlation be rY(fi), i takes 1,2,3 ... n；According to the transform domain correlation coefficient rY (fi), determine coefficient correlation vector
Alternatively, the processor 310 is used for：By i-th section of SSVEP signals X in the n section SSVEP signalsi(fi,Ti) carry out Fu In leaf transformation obtain Zi(fi,Fi)；The n sections are preset into the X in SSVEP signalsi′(fi,Ti) carry out Fourier transformation obtain Zi′ (fi,Fi)；Determine Zi(fi,Fi) and Zi′(fi,Fi) frequency domain correlation coefficient be rF(fi), i takes 1,2,3 ... n；According to the frequency domain Correlation coefficient rF(fi), determine coefficient correlation vector
It should be understood that the device 300 of the identification based on EEG signals according to embodiments of the present invention may correspond to this hair The device 200 of the identification based on EEG signals in bright embodiment, and can correspond to perform according to embodiments of the present invention Method 100 in corresponding main body, and above and other operation and/or function of the modules in device 300 is respectively The corresponding flow of each method in Fig. 1 is realized, for sake of simplicity, will not be repeated here.
Therefore, the device of the identification based on EEG signals of the embodiment of the present invention, according to goal stimulus frequency sequence Stimulus signal is shown for user to be detected, so as to gather SSVEP signals caused by user to be detected, by the SSVEP signals and in advance If SSVEP signals are contrasted, the identification of user to be detected is carried out.The characteristics of due to SSVEP signal high s/n ratios, can be with So that signal is more readily detected, stimulation duration can be shortened compared to existing brain electricity personal identification method；SSVEP signals mainly collect In in the brain occipital region of people, it is only necessary to less electrode, such as an electrode can collect the signal of abundant information amount, make With conveniently；The signal that SSVEP signals induce as a kind of primary visual cortex, it is not necessary to the higher cognitive activity of people participates in, because This its influenceed by the state of mind of people smaller, signal characteristic is relatively more stable；The abundant amplitude-frequency response of SSVEP signals and Phase-frequency response can be directly corresponding with the system physiological property of people's primary visual cortex so that the identity based on this structure is known Other system more maintains secrecy, and is not easy to be replicated forgery.In addition, for the goal stimulus frequency sequence of setting, body can also be carried out Part identification, so, exclusive stimulus sequence add the pattern of the double certifications of SSVEP signals, improve the grade of equipment secrecy, are less susceptible to It is broken into.
Those of ordinary skill in the art are it is to be appreciated that the list of each example described with reference to the embodiments described herein Member and algorithm steps, it can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually Performed with hardware or software mode, application-specific and design constraint depending on technical scheme.Professional and technical personnel Described function can be realized using distinct methods to each specific application, but this realization is it is not considered that exceed The scope of the present invention.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description, The specific work process of device and unit, the corresponding process in preceding method embodiment is may be referred to, will not be repeated here.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, can be with Realize by another way.For example, device embodiment described above is only schematical, for example, the unit Division, only a kind of division of logic function, can there is other dividing mode, such as multiple units or component when actually realizing Another system can be combined or be desirably integrated into, or some features can be ignored, or do not perform.It is another, it is shown or The mutual coupling discussed or direct-coupling or communication connection can be the indirect couplings by some interfaces, device or unit Close or communicate to connect, can be electrical, mechanical or other forms.
The unit illustrated as separating component can be or may not be physically separate, show as unit The part shown can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple On NE.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs 's.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, can also That unit is individually physically present, can also two or more units it is integrated in a unit.
If the function is realized in the form of SFU software functional unit and is used as independent production marketing or in use, can be with It is stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially in other words The part to be contributed to prior art or the part of the technical scheme can be embodied in the form of software product, the meter Calculation machine software product is stored in a storage medium, including some instructions are causing a computer equipment (can be People's computer, server, or network equipment etc.) perform all or part of step of each embodiment methods described of the present invention. And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited Reservoir (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with the medium of store program codes.
The foregoing is only a specific embodiment of the invention, but protection scope of the present invention is not limited thereto, any Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all be contained Cover within protection scope of the present invention.Therefore, protection scope of the present invention described should be defined by scope of the claims.

Claims (12)

  1. A kind of 1. method of the identification based on EEG signals, it is characterised in that including：
    Determine goal stimulus frequency sequence [f1,f2,f3,……,fn], n is positive integer；
    Goal stimulus frequency sequence [the f is shown for user to be detected1,f2,f3,……,fn] corresponding to n section stimulus signals, institute The display frequency for stating i-th section of stimulus signal in n section stimulus signals is the goal stimulus frequency sequence [f1,f2,f3,……,fn] In i-th of frequency fi, i takes 1,2,3 ... n；
    The user to be detected is obtained due to n sections Steady State Visual Evoked Potential SSVEP signals caused by the n sections stimulus signal；
    When the similarity that the n sections SSVEP signals and n sections preset SSVEP signals is more than or equal to threshold value, determine described to be checked It is correct to survey user identity；
    When the similarity that the n sections SSVEP signals and the n sections preset SSVEP signals is less than the threshold value, it is determined that described treat Detect user identity mistake.
  2. 2. according to the method for claim 1, it is characterised in that the determination goal stimulus frequency sequence [f1,f2, f3,……,fn], including：
    Obtain the frequency of stimulation sequence [f ' to be detected of user's input to be detected1,f′2,f′3,……,f′n]；
    As the frequency of stimulation sequence [f ' to be detected1,f′2,f′3,……,f′n] and the goal stimulus frequency sequence [f1,f2, f3,……,fn] it is identical when, obtain the goal stimulus frequency sequence [f1,f2,f3,……,fn] corresponding to the n sections stimulate letter Number；
    As the frequency of stimulation sequence [f ' to be detected1,f′2,f′3,……,f′n] and the goal stimulus frequency sequence [f1,f2, f3,……,fn] it is different when, determine the user identity mistake to be detected.
  3. 3. method according to claim 1 or 2, it is characterised in that methods described also includes：
    SSVEP signals are preset with the n sections in corresponding at least one domain according to the n sections SSVEP signals at least one domain, Determine that the n sections SSVEP signals preset the coefficient correlation vector of SSVEP signals with the n sectionsThe coefficient correlation vector In each element representation described in n section SSVEP signals and the n sections preset the coefficient correlations of SSVEP signals；
    WillIt is defined as the similarity that the n sections SSVEP signals preset SSVEP signals with the n sections, wherein,Represent power Weight parameter vector, the weight parameter vectorIn each element representation described in coefficient correlation vectorThe power of middle corresponding element Weight values.
  4. 4. according to the method for claim 3, it is characterised in that the determination n sections SSVEP signals and the n sections are pre- If the coefficient correlation vector of SSVEP signalsIncluding：
    Determine i-th section of SSVEP signals X in the n sections SSVEP signalsi(fi,Ti) with the n sections preset the X ' in SSVEP signalsi (fi,Ti) time domain coefficient correlation be rX(fi), i takes 1,2,3 ... n；
    According to the time domain correlation coefficient rX(fi), determine the coefficient correlation vector
  5. 5. the method according to claim 3 or 4, it is characterised in that described to determine the n sections SSVEP signals and the n sections The coefficient correlation vector of default SSVEP signalsIncluding：
    By i-th section of SSVEP signals X in the n sections SSVEP signalsi(fi,Ti) enter line translation and obtain Zi(fi,Yi)；
    The n sections are preset into the X ' in SSVEP signalsi(fi,Ti) enter line translation and obtain Z 'i(fi,Yi)；
    Determine Zi(fi,Yi) and Z 'i(fi,Yi) transform domain coefficient correlation be rY(fi), i takes 1,2,3 ... n；
    According to the transform domain correlation coefficient rY(fi), determine the coefficient correlation vector
  6. 6. the method according to claim 3 or 4, it is characterised in that described to determine the n sections SSVEP signals and the n sections The coefficient correlation vector of default SSVEP signalsIncluding：
    By i-th section of SSVEP signals X in the n sections SSVEP signalsi(fi,Ti) carry out Fourier transformation obtain Zi(fi,Fi)；
    The n sections are preset into the X ' in SSVEP signalsi(fi,Ti) carry out Fourier transformation obtain Z 'i(fi,Fi)；
    Determine Zi(fi,Fi) and Z 'i(fi,Fi) frequency domain correlation coefficient be rF(fi), i takes 1,2,3 ... n；
    According to the frequency domain correlation coefficient rF(fi), determine the coefficient correlation vector
  7. A kind of 7. device of the identification based on EEG signals, it is characterised in that including：
    Determining unit, for determining goal stimulus frequency sequence [f1,f2,f3,……,fn], n is positive integer；
    Display unit, for showing the goal stimulus frequency sequence [f for user to be detected1,f2,f3,……,fn] corresponding to n Section stimulus signal, the display frequency of i-th section of stimulus signal is the goal stimulus frequency sequence [f in the n sections stimulus signal1, f2,f3,……,fn] in i-th of frequency fi, i takes 1,2,3 ... n；
    Acquiring unit, for obtaining the user to be detected due to n sections stable state vision inducting electricity caused by the n sections stimulus signal Position SSVEP signals；
    Processing unit, for being more than or equal to threshold value when the similarity of the n sections SSVEP signals and the default SSVEP signals of n sections When, determine that the user identity to be detected is correct；When the n sections SSVEP signals preset the similar of SSVEP signals to the n sections When degree is less than the threshold value, the user identity mistake to be detected is determined.
  8. 8. device according to claim 7, it is characterised in that the determining unit is specifically used for：
    Obtain the frequency of stimulation sequence [f ' to be detected of user's input to be detected1,f′2,f′3,……,f′n]；
    As the frequency of stimulation sequence [f ' to be detected1,f′2,f′3,……,f′n] and the goal stimulus frequency sequence [f1,f2, f3,……,fn] it is identical when, obtain the goal stimulus frequency sequence [f1,f2,f3,……,fn] corresponding to the n sections stimulate letter Number；
    As the frequency of stimulation sequence [f ' to be detected1,f′2,f′3,……,f′n] and the goal stimulus frequency sequence [f1,f2, f3,……,fn] it is different when, determine the user identity mistake to be detected.
  9. 9. the device according to claim 7 or 8, it is characterised in that the processing unit is specifically used for：
    SSVEP signals are preset with the n sections in corresponding at least one domain according to the n sections SSVEP signals at least one domain, Determine that the n sections SSVEP signals preset the coefficient correlation vector of SSVEP signals with the n sectionsThe coefficient correlation vector In each element representation described in n section SSVEP signals and the n sections preset the coefficient correlations of SSVEP signals；
    WillIt is defined as the similarity that the n sections SSVEP signals preset SSVEP signals with the n sections, wherein,Represent power Weight parameter vector, the weight parameter vectorIn each element representation described in coefficient correlation vectorThe power of middle corresponding element Weight values.
  10. 10. device according to claim 9, it is characterised in that the processing unit is specifically used for：
    Determine i-th section of SSVEP signals X in the n sections SSVEP signalsi(fi,Ti) with the n sections preset the X ' in SSVEP signalsi (fi,Ti) time domain coefficient correlation be rX(fi), i takes 1,2,3 ... n；
    According to the time domain correlation coefficient rX(fi), determine the coefficient correlation vector
  11. 11. the device according to claim 9 or 10, it is characterised in that the processing unit is specifically used for：
    By i-th section of SSVEP signals X in the n sections SSVEP signalsi(fi,Ti) enter line translation and obtain Zi(fi,Yi)；
    The n sections are preset into the X ' in SSVEP signalsi(fi,Ti) enter line translation and obtain Z 'i(fi,Yi)；
    Determine Zi(fi,Yi) and Z 'i(fi,Yi) transform domain coefficient correlation be rY(fi), i takes 1,2,3 ... n；
    According to the transform domain correlation coefficient rY(fi), determine the coefficient correlation vector
  12. 12. the device according to claim 9 or 10, it is characterised in that the processing unit is specifically used for：
    By i-th section of SSVEP signals X in the n sections SSVEP signalsi(fi,Ti) carry out Fourier transformation obtain Zi(fi,Fi)；
    The n sections are preset into the X ' in SSVEP signalsi(fi,Ti) carry out Fourier transformation obtain Z 'i(fi,Fi)；
    Determine Zi(fi,Fi) and Z 'i(fi,Fi) frequency domain correlation coefficient be rF(fi), i takes 1,2,3 ... n；
    According to the frequency domain correlation coefficient rF(fi), determine the coefficient correlation vector
CN201610361544.XA 2016-05-26 2016-05-26 The method and apparatus of identification based on EEG signals Pending CN107437011A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201610361544.XA CN107437011A (en) 2016-05-26 2016-05-26 The method and apparatus of identification based on EEG signals
PCT/CN2016/104445 WO2017201972A1 (en) 2016-05-26 2016-11-03 Identity recognition method and device based on electroencephalogram signal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610361544.XA CN107437011A (en) 2016-05-26 2016-05-26 The method and apparatus of identification based on EEG signals

Publications (1)

Publication Number Publication Date
CN107437011A true CN107437011A (en) 2017-12-05

Family

ID=60411041

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610361544.XA Pending CN107437011A (en) 2016-05-26 2016-05-26 The method and apparatus of identification based on EEG signals

Country Status (2)

Country Link
CN (1) CN107437011A (en)
WO (1) WO2017201972A1 (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108153414A (en) * 2017-12-19 2018-06-12 同济大学 Recognition methods, system and the device of Steady State Visual Evoked Potential signal
CN108985029A (en) * 2018-06-05 2018-12-11 中国科学院半导体研究所 The brain electricity personal identification method of view-based access control model stimulation
CN109101807A (en) * 2018-09-10 2018-12-28 清华大学 A kind of brain electricity identity authority control system and method
CN109766751A (en) * 2018-11-28 2019-05-17 西安电子科技大学 Stable state vision inducting brain electricity personal identification method and system based on Frequency Domain Coding
CN109948700A (en) * 2019-03-19 2019-06-28 北京字节跳动网络技术有限公司 Method and apparatus for generating characteristic pattern
CN111329474A (en) * 2020-03-04 2020-06-26 西安电子科技大学 Electroencephalogram identity recognition method and system based on deep learning and information updating method
CN111714120A (en) * 2020-05-08 2020-09-29 广东食品药品职业学院 Brain-computer interface system capable of evaluating visual positioning ability and application thereof
CN113360876A (en) * 2021-06-30 2021-09-07 杭州回车电子科技有限公司 SSVEP-based identity recognition method and device, electronic device and storage medium
CN113935376A (en) * 2021-10-13 2022-01-14 中国科学技术大学 Brain function subregion partitioning method based on joint constraint canonical correlation analysis
CN114424945A (en) * 2021-12-08 2022-05-03 中国科学院深圳先进技术研究院 Brain wave biological feature recognition system and method based on random graphic image flash

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109816406B (en) * 2019-02-26 2021-01-22 北京理工大学 Article marking method, device, equipment and 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
CN114246594B (en) * 2021-12-17 2024-04-09 天津大学 Electroencephalogram signal processing method, background electroencephalogram prediction model training method and device

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102866775A (en) * 2012-09-04 2013-01-09 同济大学 System and method for controlling brain computer interface (BCI) based on multimode fusion
US8483816B1 (en) * 2010-02-03 2013-07-09 Hrl Laboratories, Llc Systems, methods, and apparatus for neuro-robotic tracking point selection
CN105389004A (en) * 2015-10-22 2016-03-09 上海斐讯数据通信技术有限公司 Brain wave unlocking terminal screen system and method

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101828921A (en) * 2010-06-13 2010-09-15 天津大学 Identity identification method based on visual evoked potential (VEP)
CN102509103A (en) * 2011-09-29 2012-06-20 天津大学 Identifying method based on visual evoked P3 potential
CN102755162B (en) * 2012-06-14 2014-04-30 天津大学 Audio-visual cognitive event-related electroencephalogram-based identification method
CN103729586A (en) * 2013-12-20 2014-04-16 北京握奇数据***有限公司 Method and system for authenticating passwords on basis of brain wave signals
CN105389494A (en) * 2015-11-04 2016-03-09 上海大学 Biometric identification password verification system and method based on brain wave

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8483816B1 (en) * 2010-02-03 2013-07-09 Hrl Laboratories, Llc Systems, methods, and apparatus for neuro-robotic tracking point selection
CN102866775A (en) * 2012-09-04 2013-01-09 同济大学 System and method for controlling brain computer interface (BCI) based on multimode fusion
CN105389004A (en) * 2015-10-22 2016-03-09 上海斐讯数据通信技术有限公司 Brain wave unlocking terminal screen system and method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
MONTRI PHOTHISONOTHAI: "An Investigation of using SSVEP for EEG-based User Authentication System", 《PROCEEDINGS OF APSIPA ANNUAL SUMMIT AND CONFERENCE 2015》 *

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108153414A (en) * 2017-12-19 2018-06-12 同济大学 Recognition methods, system and the device of Steady State Visual Evoked Potential signal
CN108153414B (en) * 2017-12-19 2021-06-15 同济大学 Method, system and device for identifying steady-state visual evoked potential signals
CN108985029A (en) * 2018-06-05 2018-12-11 中国科学院半导体研究所 The brain electricity personal identification method of view-based access control model stimulation
CN109101807A (en) * 2018-09-10 2018-12-28 清华大学 A kind of brain electricity identity authority control system and method
CN109766751A (en) * 2018-11-28 2019-05-17 西安电子科技大学 Stable state vision inducting brain electricity personal identification method and system based on Frequency Domain Coding
CN109766751B (en) * 2018-11-28 2022-02-01 西安电子科技大学 Steady-state vision-evoked electroencephalogram identity recognition method and system based on frequency domain coding
CN109948700B (en) * 2019-03-19 2020-07-24 北京字节跳动网络技术有限公司 Method and device for generating feature map
CN109948700A (en) * 2019-03-19 2019-06-28 北京字节跳动网络技术有限公司 Method and apparatus for generating characteristic pattern
CN111329474A (en) * 2020-03-04 2020-06-26 西安电子科技大学 Electroencephalogram identity recognition method and system based on deep learning and information updating method
CN111714120A (en) * 2020-05-08 2020-09-29 广东食品药品职业学院 Brain-computer interface system capable of evaluating visual positioning ability and application thereof
CN113360876A (en) * 2021-06-30 2021-09-07 杭州回车电子科技有限公司 SSVEP-based identity recognition method and device, electronic device and storage medium
CN113935376A (en) * 2021-10-13 2022-01-14 中国科学技术大学 Brain function subregion partitioning method based on joint constraint canonical correlation analysis
CN113935376B (en) * 2021-10-13 2023-03-10 中国科学技术大学 Brain function subregion partitioning method based on joint constraint canonical correlation analysis
CN114424945A (en) * 2021-12-08 2022-05-03 中国科学院深圳先进技术研究院 Brain wave biological feature recognition system and method based on random graphic image flash
CN114424945B (en) * 2021-12-08 2024-05-31 中国科学院深圳先进技术研究院 Brain wave biological feature recognition system and method based on random graphic image flash

Also Published As

Publication number Publication date
WO2017201972A1 (en) 2017-11-30

Similar Documents

Publication Publication Date Title
CN107437011A (en) The method and apparatus of identification based on EEG signals
Wang et al. Convolutional neural networks using dynamic functional connectivity for EEG-based person identification in diverse human states
Jayarathne et al. BrainID: Development of an EEG-based biometric authentication system
Singh et al. Bioelectrical signals as emerging biometrics: Issues and challenges
CA2835460C (en) System and method for enabling continuous or instantaneous identity recognition based on physiological biometric signals
Lemm et al. Spatio-spectral filters for improving the classification of single trial EEG
US11132430B2 (en) Login method and apparatus and electronic device
El-Fiqi et al. Convolution neural networks for person identification and verification using steady state visual evoked potential
Chamatidis et al. Using deep learning neural networks for ECG based authentication
CN104536573B (en) A kind of brain-machine interface method stimulated based on high frequency flicker emotion
CN104571504B (en) A kind of online brain-machine interface method based on Imaginary Movement
Singhal et al. Person identification using evoked potentials and peak matching
CN109770900A (en) Brain-computer interface based on convolutional neural networks instructs delivery method, system, device
CN101843491A (en) Resting electroencephalogram identification method based on bilinear model
Wang et al. Stable EEG Biometrics Using Convolutional Neural Networks and Functional Connectivity.
CN107957780A (en) A kind of brain machine interface system based on Steady State Visual Evoked Potential physiological property
ŠŤASTNÝ et al. Overcoming Inter-Subject Variability In BCI Using EEG-Based Identification.
Demir et al. Bio-inspired filter banks for SSVEP-based brain-computer interfaces
Frank et al. Personalizable neurological user authentication framework
Seha et al. A new training approach for deep learning in EEG biometrics using triplet loss and EMG-driven additive data augmentation
Thomas et al. EEG-based biometrie authentication using self-referential visual stimuli
Pham EEG-based person authentication for security systems
CN109784027A (en) Auth method based on dual EEG signals feature
Wang et al. Never lost keys: A novel key generation scheme based on motor imagery EEG in end-edge-cloud system
Dutta et al. Functional connectivity methods for EEG-based biometrics on a large, heterogeneous dataset

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

Application publication date: 20171205

RJ01 Rejection of invention patent application after publication