CN105125210A - Brain wave evoking method and device - Google Patents
Brain wave evoking method and device Download PDFInfo
- Publication number
- CN105125210A CN105125210A CN201510570569.6A CN201510570569A CN105125210A CN 105125210 A CN105125210 A CN 105125210A CN 201510570569 A CN201510570569 A CN 201510570569A CN 105125210 A CN105125210 A CN 105125210A
- Authority
- CN
- China
- Prior art keywords
- brain wave
- sample
- brain
- training sample
- similarity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses a brain wave evoking method and device. The brain of a trainee is stimulated through a preset training sample, and sample brain waves corresponding to the training sample are collected; target brain waves are preset, and feature vectors of the target brain waves are extracted; feature vectors of the sample brain waves are extracted, and the first similarity between the feature vectors of the sample brain waves and the feature vectors of the target brain waves is calculated; a brain stimulating sample for making the trainee produce the target brain waves is selected according to the first similarity, the technical problem that existing brain waves produced on the basis of evoked potentials are not stable or standard enough is solved, the brain stimulating sample corresponding to the target brain waves is obtained according to the target brain waves, and therefore brain wave signals consistent with those of the preset target brain waves can be obtained; the brain stimulating sample for making the trainee produce the target brain waves is selected according to the first similarity, stability and reliability can be achieved, the standard brain waves are obtained, and therefore guarantees are provided for a subsequent brain machine interface system with brain wave signals as control signals.
Description
Technical field
The present invention relates to brain-computer interface technical field, what be specifically related to a kind of brain wave brings out method and device.
Background technology
Evoked ptential (evokedpotential, EP) is the combined reaction of the electrical activity that brain stimulates to external world.Evoked ptential is significant in multiple subjects such as clinical medicine, electrophysiology, psychology and Cognitive Science.Current Evoked ptential is mainly divided into three kinds of mode according to stimulation mode difference: auditory evoked potential (AuditoryEvokedPotential, AEP), visual evoked potential (VisualEvokedPotential, and somatosensory evoked potential (SomatosensoryEvokedPotential, SEP) VEP).
But the brain wave produced based on above-mentioned three kinds of main Evoked ptential is stablized and specification not, such as same set of inducing device may produce different brain waves for different tested objects, and it is also unstable to bring out the brain wave of generation for same tested object, thus causing the brain-computer interface stability of control system producing brain wave based on Evoked ptential poor, degree of accuracy is not high.
Summary of the invention
What the invention provides a kind of brain wave brings out method and device, stablizes the technical problem with specification not to solve the existing brain wave produced based on Evoked ptential.
According to an aspect of the present invention, what provide a kind of brain wave brings out method, comprising:
Adopt the training sample preset to carry out brain stimulation to trainer, and gather the sample brain wave corresponding with training sample;
Goal-selling brain wave, and the characteristic vector extracting target brain wave;
Extract the characteristic vector of sample brain wave, and calculate the first similarity between the characteristic vector of sample brain wave and the characteristic vector of target brain wave;
The brain stimulation sample producing target brain wave for tester is chosen according to the first similarity.
Further, adopt the training sample preset to carry out brain stimulation to trainer, and the collection sample brain wave corresponding with training sample comprises:
Electrode for encephalograms is arranged on the position of the scalp supraoccipital bone knuckle of multiple trainer;
For each trainer, screen shows multiple training sample successively according to each self-corresponding stimulation displaying time of multiple training sample, and the default lounge interval is set between multiple training sample display;
According to the lounge interval, the sample brain wave that each trainer produces for multiple training sample is carried out segmentation intercepting, and the meansigma methods of each training sample being carried out the sample brain wave of brain stimulation generation to different trainer is as the average sample brain wave corresponding with each training sample;
By frequency spectrum analysis method, obtain the major frequency components of the average sample brain wave corresponding with each training sample, and using major frequency components as the sample brain wave corresponding with each training sample.
Further, comprise before obtaining the major frequency components of the average sample brain wave corresponding with each training sample:
Carry out pretreatment to the average sample brain wave corresponding with each training sample, pretreatment comprises one or more operations in denoising, filtering, processing and amplifying.
Further, choose according to the first similarity the brain stimulation sample producing target brain wave for tester to comprise:
Choose the training sample that sample brain wave corresponding to the first maximum similarity is corresponding, as the brain stimulation sample producing target brain wave for tester.
Further, choose according to the first similarity the brain stimulation sample producing target brain wave for tester to comprise:
The training sample choosing the sample brain wave corresponding with it corresponding by the first similarity order from large to small carries out brain stimulation to tester, and calculates the second similarity of the characteristic vector of brain wave and the characteristic vector of target brain wave produced according to training sample simultaneously;
Choose the second similarity and be greater than training sample corresponding to default similarity threshold, as the brain stimulation sample producing target brain wave for tester.
Further, the type of training sample is any one or the multiple combination in text, picture, expression, animation, audio frequency and video type.
According to a further aspect in the invention, provide a kind of inducing device of brain wave, comprising:
Training devices, for adopting the training sample preset to carry out brain stimulation to trainer, and gathers the sample brain wave corresponding with training sample;
Target preinstall apparatus, for goal-selling brain wave, and extracts the characteristic vector of target brain wave;
First similarity accountant, for extracting the characteristic vector of sample brain wave, and calculates the first similarity between the characteristic vector of sample brain wave and the characteristic vector of target brain wave;
Brain stimulation sample selecting device, for choosing the brain stimulation sample producing target brain wave for tester according to the first similarity.
Further, device comprises:
Electrode installing device, for being arranged on the position of the scalp supraoccipital bone knuckle of multiple trainer by electrode for encephalograms;
Training sample setting device, for for each trainer, screen shows multiple training sample successively according to each self-corresponding stimulation displaying time of multiple training sample, and arranges the default lounge interval between multiple training sample display;
Average sample brain wave acquisition device, for according to the lounge interval, the sample brain wave that each trainer produces for multiple training sample is carried out segmentation intercepting, and the meansigma methods of each training sample being carried out the sample brain wave of brain stimulation generation to different trainer is as the average sample brain wave corresponding with each training sample;
Major frequency components acquisition device, for obtaining the major frequency components of the average sample brain wave corresponding with each training sample, and using major frequency components as the sample brain wave corresponding with each training sample.
Further, major frequency components acquisition device also comprises:
Pretreatment unit, for carrying out pretreatment to the average sample brain wave corresponding with each training sample, pretreatment comprises one or more operations in denoising, filtering, processing and amplifying.
Further, brain stimulation sample selecting device comprises:
Maximum similarity selecting device, for choosing training sample corresponding to sample brain wave corresponding to the first maximum similarity, as the brain stimulation sample producing target brain wave for tester.
The present invention has following beneficial effect:
What the invention provides a kind of brain wave brings out method and device, by adopting the training sample preset to carry out brain stimulation to trainer, and gathers the sample brain wave corresponding with training sample, goal-selling brain wave, and the characteristic vector extracting target brain wave, extract the characteristic vector of sample brain wave, and calculate the first similarity between the characteristic vector of sample brain wave and the characteristic vector of target brain wave, the brain stimulation sample producing target brain wave for tester is chosen according to the first similarity, solve the existing brain wave produced based on Evoked ptential and stablize the technical problem with specification not, achieve and obtain the brain stimulation sample corresponding with target brain wave according to target brain wave, thus the eeg signal consistent with the target brain wave preset can be obtained, and the brain stimulation sample producing target brain wave for tester is chosen by the first similarity, can produce reliable and stable, and the brain wave of specification, thus provide guarantee as the brain-computer interface control system of control signal for this eeg signal of follow-up employing.
Except object described above, feature and advantage, the present invention also has other object, feature and advantage.Below with reference to figure, the present invention is further detailed explanation.
Accompanying drawing explanation
The accompanying drawing forming a application's part is used to provide a further understanding of the present invention, and schematic description and description of the present invention, for explaining the present invention, does not form inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 be the brain wave of the preferred embodiment of the present invention bring out method flow diagram;
What Fig. 2 was the preferred embodiment of the present invention for the brain wave of a specific embodiment brings out method flow diagram;
To be the preferred embodiment of the present invention adopt the training sample preset to carry out brain stimulation to trainer for the bringing out in method of brain wave of a specific embodiment to Fig. 3, and gather the method flow diagram of the sample brain wave corresponding with training sample;
Fig. 4 is the inducing device structural representation of the brain wave of the preferred embodiment of the present invention.
Description of reference numerals:
10, training devices; 20, target preinstall apparatus; 30, the first similarity accountant; 40, brain stimulation sample selecting device.
Detailed description of the invention
Below in conjunction with accompanying drawing, embodiments of the invention are described in detail, but the multitude of different ways that the present invention can be defined by the claims and cover is implemented.
With reference to Fig. 1, what the preferred embodiments of the present invention provided a kind of brain wave brings out method, comprising:
Step S101, adopts the training sample preset to carry out brain stimulation to trainer, and gathers the sample brain wave corresponding with training sample;
Step S102, goal-selling brain wave, and the characteristic vector extracting target brain wave;
Step S103, extracts the characteristic vector of sample brain wave, and calculates the first similarity between the characteristic vector of sample brain wave and the characteristic vector of target brain wave;
Step S104, chooses the brain stimulation sample producing target brain wave for tester according to the first similarity.
Brain wave provided by the invention bring out method, by adopting the training sample that presets to carry out brain stimulation to trainer, and gather the sample brain wave corresponding with training sample, goal-selling brain wave, and the characteristic vector extracting target brain wave, extract the characteristic vector of sample brain wave, and calculate the first similarity between the characteristic vector of sample brain wave and the characteristic vector of target brain wave, the brain stimulation sample producing target brain wave for tester is chosen according to the first similarity, solve the existing brain wave produced based on Evoked ptential and stablize the technical problem with specification not, achieve and obtain the brain stimulation sample corresponding with target brain wave according to target brain wave, thus the eeg signal consistent with the target brain wave preset can be obtained, and the brain stimulation sample producing target brain wave for tester is chosen by the first similarity, can produce reliable and stable, and the brain wave of specification, thus provide guarantee as the brain-computer interface control system of control signal for this eeg signal of follow-up employing.
The method of bringing out of the brain wave of the present embodiment is by presetting target brain wave, and find according to the characteristic vector of target brain wave the brain stimulation sample producing this target brain wave for tester, thus realize can producing the brain wave consistent with target brain wave for most of tester according to the brain stimulation sample obtained bring out method, and brain stimulation is carried out to tester can obtain relatively stable specification and controlled brain wave according to the brain stimulation sample obtained, solve and adopt same set of inducing device may produce different brain waves for different tested objects, and it is also unstable to bring out the brain wave of generation for same tested object, thus cause the brain-computer interface stability of control system producing brain wave based on Evoked ptential poor, degree of accuracy is not high, substantially increase the validity and reliability of Evoked ptential.
In addition, goal-selling brain wave in the present embodiment is the brain wave of user-defined arbitrary form, by goal-selling brain wave, the present embodiment can brain wave to provide various forms, thus for needing the brain machine control system of multi-way contral signal to provide convenience.In the present embodiment, target brain wave can be set by the wave function of brain wave, also can be set by the characteristic item of brain wave, such as periodic quantity, frequency values, crest value, peak frequencies value, phase signal value etc.
The first similarity in the present embodiment refers to the similarity degree between sample brain wave and target brain wave, is represented by the distance between sample brain wave and target brain wave characteristic of correspondence vector.The distance of two characteristic vectors is shorter, illustrates that similarity corresponding to these two characteristic vectors is larger.The mode of conventional description vectors distance has Euclidean distance, cosine angle etc.Represent that the computing formula of the similarity of two characteristic vectors is by the included angle cosine value calculating two characteristic vectors:
wherein, X, Y represent the feature vector, X of sample brain wave and the characteristic vector Y of target brain wave respectively.The span of cos θ is [0,1], cos θ more close to 1 time, then represent that the similarity between sample brain wave and target brain wave is higher, otherwise cos θ value is more close to 0, then the similarity between expression sample brain wave and target brain wave is lower.The similarity of sample brain wave and target brain wave can be reflected comparatively intuitively by the included angle cosine value calculating two characteristic vectors.
Alternatively, adopt the training sample preset to carry out brain stimulation to trainer, and the collection sample brain wave corresponding with training sample comprises:
Electrode for encephalograms is arranged on the position of the scalp supraoccipital bone knuckle of multiple trainer;
For each trainer, screen shows multiple training sample successively according to each self-corresponding stimulation displaying time of multiple training sample, and the default lounge interval is set between multiple training sample display;
According to the lounge interval, the sample brain wave that each trainer produces for multiple training sample is carried out segmentation intercepting, and the meansigma methods of each training sample being carried out the sample brain wave of brain stimulation generation to different trainer is as the average sample brain wave corresponding with each training sample;
By frequency spectrum analysis method, obtain the major frequency components of the average sample brain wave corresponding with each training sample, and using major frequency components as the sample brain wave corresponding with each training sample.
Because different trainers might not be identical to the sample brain wave that the brain stimulation of same training sample produces, therefore the present embodiment carries out brain stimulation for each training sample in training sample sequence to multiple trainer, and the major frequency components multiple trainer being carried out the meansigma methods of the sample brain wave of brain stimulation generation is as the sample brain wave corresponding with each training sample.By same training sample is carried out brain stimulation to multiple trainer, and get multiple trainer and carry out the major frequency components of the meansigma methods of the sample brain wave of brain stimulation generation as the sample brain wave corresponding with training sample, improve the extraction accuracy of sample brain wave, the sample brain wave extracted is made to have versatility, and reliability.
In concrete implementation process, the present embodiment can calculate the stimulation displaying time corresponding with training sample flexibly according to the type of the training sample chosen, such as can calculate reading time needed for the training sample reading text type according to the character length of the training sample of text type, and using the training sample stimulation displaying time on computers of this reading time as text type, and for the training sample of video type, then can using the training sample stimulation displaying time on computers of the reproduction time needed for the training sample playing this video type as this video type.
In addition, the present embodiment can adopt Fourier transformation or fast Fourier transform to carry out spectrum analysis to average sample brain wave, thus obtains the major frequency components of average sample brain wave.Sample brain wave corresponding with each training sample in the present embodiment is a corresponding control signal passage respectively, or an action command, or direct control signal.By adopting frequency spectrum analysis method, obtain the major frequency components of average sample brain wave, and using major frequency components as the sample brain wave corresponding with training sample, spontaneous brain electricity ripple or outside noise can be avoided the impact of the sample brain wave that training sample produces, thus obtain truer and effective sample brain wave.
It should be noted that, the present embodiment adopts the training sample preset to carry out brain stimulation to trainer, and gathers corresponding with training sample sample brain wave employing offline mode and train.In concrete implementation process, the present embodiment also comprises and stores in systems in which, the good training sample of these off-line trainings and the sample brain wave corresponding with training sample so that the follow-up training sample mated with it according to target brain wave fast search.
Alternatively, comprise before obtaining the major frequency components of the average sample brain wave corresponding with each training sample: carry out pretreatment to the average sample brain wave corresponding with each training sample, pretreatment comprises one or more operations in denoising, filtering, processing and amplifying.
By carrying out pretreatment to average sample brain wave before the major frequency components obtaining average sample brain wave, noise can be removed, filter clutter, thus obtaining more steady and audible major frequency components.
Alternatively, choose according to the first similarity the brain stimulation sample producing target brain wave for tester to comprise:
Choose the training sample that sample brain wave corresponding to the first maximum similarity is corresponding, as the brain stimulation sample producing target brain wave for tester.
The present embodiment can obtain the sample brain wave corresponding with each training sample by Induced by Stimulation, therefore when needing to produce the brain wave corresponding with the target brain wave preset, can obtain according to the characteristic vector of target brain wave and the similarity of the characteristic vector of sample brain wave the corresponding brain stimulation sample producing target brain wave.Again because the sample brain wave of the present embodiment carries out the major frequency components acquisition of the average sample brain wave of brain stimulation generation by multiple trainer, therefore for each the sample brain wave obtained, it has all been through trains to multiple training objects the representative brain wave obtained.Therefore the present embodiment directly can choose training sample corresponding to sample brain wave corresponding to the first maximum similarity, as the brain stimulation sample producing target brain wave for tester.The training sample that sample brain wave corresponding to maximum the first similarity of direct employing is corresponding, as the brain stimulation sample producing target brain wave for tester, representative, and the brain wave maximum with target brain wave similarity can be obtained according to the brain stimulation sample selected.
Alternatively, choose according to the first similarity the brain stimulation sample producing target brain wave for tester to comprise:
The training sample choosing the sample brain wave corresponding with it corresponding by the first similarity order from large to small carries out brain stimulation to tester, and calculates the second similarity of the characteristic vector of brain wave and the characteristic vector of target brain wave produced according to training sample simultaneously;
Choose the second similarity and be greater than training sample corresponding to default similarity threshold, as the brain stimulation sample producing target brain wave for tester.
Due to direct by training sample corresponding for sample brain wave corresponding for the first maximum similarity, differ as the brain stimulation sample producing target brain wave for tester and obtain and the immediate real EEG electric wave of target brain wave surely, the real EEG electric wave in the present embodiment refers to and adopts the brain stimulation sample obtained according to the first similarity to bring out the real EEG electric wave obtained.For this problem, the training sample that the present embodiment chooses the sample brain wave corresponding with it corresponding by the first similarity order from large to small carries out brain stimulation to tester, and calculate the second similarity of the characteristic vector of brain wave and the characteristic vector of target brain wave produced according to training sample simultaneously, and choose the second similarity and be greater than training sample corresponding to default similarity threshold, as the brain stimulation sample producing target brain wave for tester.In addition, why the present embodiment chooses the second similarity is greater than training sample corresponding to default similarity threshold, as producing the brain stimulation sample of target brain wave for tester mainly in order to reduce amount of calculation, particularly, the present embodiment is when choosing training sample corresponding to the sample brain wave corresponding with it and carrying out brain stimulation to tester by the first similarity order from large to small, as long as when detecting that the characteristic vector of brain wave and the second similarity of the characteristic vector of target brain wave produced according to training sample is greater than training sample corresponding to default similarity threshold and the replacing of deconditioning sample, and directly the second similarity is greater than training sample corresponding to default similarity threshold, as the brain stimulation sample producing target brain wave for tester.
The present embodiment is by calculating the second similarity of the characteristic vector of brain wave and the characteristic vector of target brain wave produced according to training sample, and choose the second similarity and be greater than training sample corresponding to default similarity threshold, as the brain stimulation sample producing target brain wave for tester.Can not only obtain with target brain wave closer to brain wave, and amount of calculation is than by calculating the characteristic vector of brain wave that all training samples produce and the brain stimulation sample that the second similarity of the characteristic vector of target brain wave is chosen as producing target brain wave for tester will reduce greatly, thus improve computational speed, accelerate the brain stimulation sample for obtaining closest to target brain wave.
Alternatively, the type of training sample is any one or the multiple combination in text, picture, expression, animation, audio frequency and video type.
The type of the training sample of the present embodiment can be any one or multiple combination in text, picture, expression, animation, audio frequency and video type, polytype brain wave is obtained by adopting training sample that is dissimilar or its combination to bring out, thus enriched the kind of goal-selling brain wave, for later use target brain wave provides multiple selection as the control signal of brain-computer interface control system.
For a specific embodiment, the process of bringing out method of the brain wave of the embodiment of the present invention and principle are specifically described below.With reference to figure 2, the method for bringing out realizing brain wave for this specific embodiment comprises:
Step S201, adopts the training sample preset to carry out brain stimulation to trainer, and gathers the sample brain wave corresponding with training sample.The present embodiment adopts the training sample preset to carry out brain stimulation to trainer, gathers the sample brain wave corresponding with training sample and adopts offline mode to train.And the present embodiment adopts the training sample of text type to carry out brain stimulation to trainer.Suppose that the training sample number of the present embodiment is 100, the number of trainer is 10.Particularly, the present embodiment adopts the training sample preset to carry out brain stimulation to trainer, and the collection sample brain wave corresponding with training sample comprises the following steps:
Step S2011, is arranged on the position of the scalp supraoccipital bone knuckle of multiple trainer by electrode for encephalograms.Because the present embodiment adopts the training sample of text type to carry out brain stimulation to trainer, its brain wave produced belongs to the brain wave of vision induced type, therefore position electrode for encephalograms being arranged on the scalp supraoccipital bone knuckle of trainer can obtain accurate and reliable eeg signal.
Step S2012, for each trainer, screen shows multiple training sample successively according to each self-corresponding stimulation displaying time of multiple training sample, and arranges the default lounge interval between multiple training sample display.In concrete implementation process, the stimulation displaying time corresponding with each training sample sets according to the length of the training sample of text type, also namely the stimulation displaying time corresponding with it can be calculated by the training sample length-flexible detecting text type, such as the training sample of the text type of one section of 100 character, system can according to average reading level 40 character per second of adult, and calculating the stimulation displaying time corresponding with this training sample is 2.5 seconds.
It is conveniently follow-up for same training sample that the present embodiment arranges the default lounge interval between each training sample stimulation display, calculates it carries out the brain wave of brain stimulation generation meansigma methods to different trainer.
Step S2013, according to the lounge interval, the sample brain wave that each trainer produces for multiple training sample is carried out segmentation intercepting, and the meansigma methods of each training sample being carried out the sample brain wave of brain stimulation generation to different trainer is as the average sample brain wave corresponding with each training sample.Particularly, the intercepting time that the present embodiment can arrange the brain wave that each trainer produces for different training sample is the midrange of lounge interval, so not only effectively distinguish the brain wave that different training sample produces, and take into full account the response time that the training sample of trainer's read text type exists.
Step S2014, by frequency spectrum analysis method, obtains the major frequency components of the average sample brain wave corresponding with each training sample, and using major frequency components as the sample brain wave corresponding with each training sample.
In addition, the training sample that the present embodiment chooses text type carries out brain stimulation to trainer, wherein, text type can comprise polytype text, such as according to the text type that humorous degree divides, the text type according to different Partition for Interest Points and the text type according to different industries, different field division etc.
Step S202, goal-selling brain wave, and the characteristic vector extracting target brain wave.Conventional feature extracting method has autoregression AR model, wavelet transformation, independent component analysis and cospace pattern etc.The present embodiment adopts the method for wavelet transformation to extract the characteristic vector of target brain wave.Wavelet transformation is a kind of signal procesing in time domain method.Compared with Fourier transformation, wavelet transformation is the localization analysis of temporal frequency, it progressively carries out multi-scale refinement by flexible shift operations to signal, finally reach high frequency treatment time subdivision, the frequency segmentation of low frequency place, automatically can adapt to the requirement that time frequency signal is analyzed, thus any details of signal can be focused on, solve the difficult problem of Fourier transformation.Particularly under the condition lacking priori, wavelet transformation can to detect in EEG signals in short-term effectively, low-energy transient pulse, and its maximum advantage is the different frequency composition adopting variable time-frequency window to remove analytic signal.
Suppose that the target brain wave M of the present embodiment represents, and the characteristic vector of hypothesis M is expressed as m=(m
1m
2.m
i.m
r), wherein, r represents the dimension of the characteristic vector of M.The dimension of the follow-up characteristic vector for sample brain wave of the present embodiment is identical with the dimension of the characteristic vector of target brain wave.
Step S203, extracts the characteristic vector of sample brain wave, and calculates the first similarity between the characteristic vector of sample brain wave and the characteristic vector of target brain wave.Particularly, supposing that 100 training samples provided according to step S201 obtain the sample brain wave sequence corresponding with these 100 training samples is S=(S
1s
2.S
i.S
100), wherein S
irepresent the sample brain wave that i-th training sample produces, and be s for the characteristic vector that each sample brain wave extracts
i=(s
i1s
i2.s
ii.s
ir), wherein s
irepresent the characteristic vector of the sample brain wave that i-th training sample produces.
Target brain wave due to the present embodiment is known quantity, therefore can obtain the characteristic vector m=(m of target brain wave by wavelet transformation
1m
2.m
i.m
r), and the characteristic vector of sample brain wave also can obtain by carrying out wavelet transformation to sample brain wave, is not difficult to calculate the first similarity N=(N of the characteristic vector of target brain wave and the characteristic vector of sample brain wave like this
1n
2.N
i.N
100), wherein N
irepresent the characteristic vector s of the sample brain wave that i-th training sample produces
iand the similarity between the characteristic vector m of target brain wave.Particularly, the present embodiment asks for similarity by the included angle cosine value of calculating two characteristic vectors.
Step S204, the training sample choosing the sample brain wave corresponding with it corresponding by the first similarity order from large to small carries out brain stimulation to tester, and calculates the second similarity of the characteristic vector of brain wave and the characteristic vector of target brain wave produced according to training sample simultaneously.
Step S205, chooses the second similarity and is greater than training sample corresponding to default similarity threshold, as the brain stimulation sample producing target brain wave for tester.
Particularly, the training sample that first the present embodiment chooses the first maximum similarity corresponding carries out brain stimulation to tester, and calculates the second similarity of the characteristic vector of the brain wave produced according to this training sample and the characteristic vector of target brain wave simultaneously.Then choose the second similarity and be greater than training sample corresponding to default similarity threshold, as the brain stimulation sample producing target brain wave for tester, it is 0.9 (similarity threshold scope is 0-1) that the present embodiment pre-defines similarity threshold.Like this, when the second similarity of the characteristic vector of the brain wave produced according to training sample and the characteristic vector of target brain wave is greater than 0.9, directly using now corresponding training sample as the brain stimulation sample being used for tester and producing target brain wave.As can be seen here, the present embodiment produces the brain wave high with target brain wave similarity by finally choosing according to the calculating of the second similarity for different testers.Thus avoid same set of inducing device and may produce different brain waves for different tested objects, and bring out the also unstable problem of the brain wave of generation for same tested object.It should be noted that, the method that the present embodiment calculates the second similarity is identical with the method calculating the first similarity.
Alternatively, the present embodiment also can choose training sample corresponding to sample brain wave corresponding to the first maximum similarity, as the brain stimulation sample producing target brain wave for tester.Particularly, the present embodiment also can choose the first similarity N=(N
1n
2.N
i.N
100) in training sample corresponding to sample brain wave corresponding to maximum value, as the brain stimulation sample producing target brain wave for tester.
The brain wave that the embodiment of the present invention provides bring out method, by adopting the training sample that presets to carry out brain stimulation to trainer, and gather the sample brain wave corresponding with training sample, goal-selling brain wave, and the characteristic vector extracting target brain wave, extract the characteristic vector of sample brain wave, and calculate the first similarity between the characteristic vector of sample brain wave and the characteristic vector of target brain wave, the brain stimulation sample producing target brain wave for tester is chosen according to the first similarity, solve the existing brain wave produced based on Evoked ptential and stablize the technical problem with specification not, achieve and obtain the brain stimulation sample corresponding with target brain wave according to target brain wave, thus the eeg signal consistent with the target brain wave preset can be obtained, and the brain stimulation sample producing target brain wave for tester is chosen by the first similarity, can produce reliable and stable, and the brain wave of specification, thus provide guarantee as the brain-computer interface control system of control signal for this eeg signal of follow-up employing.
With reference to Fig. 4, the inducing device of the brain wave that the preferred embodiments of the present invention provide, comprising:
Training devices, for adopting the training sample preset to carry out brain stimulation to trainer, and gathers the sample brain wave corresponding with described training sample;
Target preinstall apparatus, for goal-selling brain wave, and extracts the characteristic vector of described target brain wave;
First similarity accountant, for extracting the characteristic vector of described sample brain wave, and calculates the first similarity between the characteristic vector of described sample brain wave and the characteristic vector of described target brain wave;
Brain stimulation sample selecting device, for choosing the brain stimulation sample producing described target brain wave for tester according to described first similarity.
Alternatively, training devices comprises:
Electrode installing device, for being arranged on the position of the scalp supraoccipital bone knuckle of multiple trainer by electrode for encephalograms;
Training sample setting device, for for each trainer, screen shows multiple training sample successively according to each self-corresponding stimulation displaying time of multiple training sample, and arranges the default lounge interval between multiple training sample display;
Average sample brain wave acquisition device, for according to the lounge interval, the sample brain wave that each trainer produces for multiple training sample is carried out segmentation intercepting, and the meansigma methods of each training sample being carried out the sample brain wave of brain stimulation generation to different trainer is as the average sample brain wave corresponding with each training sample;
Major frequency components acquisition device, for obtaining the major frequency components of the average sample brain wave corresponding with each training sample, and using major frequency components as the sample brain wave corresponding with each training sample.
Alternatively, major frequency components acquisition device also comprises:
Pretreatment unit, for carrying out pretreatment to the average sample brain wave corresponding with each training sample, pretreatment comprises one or more operations in denoising, filtering, processing and amplifying.
Alternatively, brain stimulation sample selecting device comprises:
Maximum similarity selecting device, for choosing training sample corresponding to sample brain wave corresponding to the first maximum similarity, as the brain stimulation sample producing target brain wave for tester.
Screen in the embodiment of the present invention can be the screen of mobile terminal device, also can be the screen of computer, wherein mobile terminal device can be desktop computer, panel computer, personal digital assistant, mobile phone, television set, vehicle-mounted computer, wearable communication equipment etc.The specific works process of the inducing device of the brain wave of the present embodiment and operation principle can refer to the work process of bringing out method and the operation principle of the brain wave in the present embodiment.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. brain wave bring out a method, it is characterized in that, comprising:
Adopt the training sample preset to carry out brain stimulation to trainer, and gather the sample brain wave corresponding with described training sample;
Goal-selling brain wave, and the characteristic vector extracting described target brain wave;
Extract the characteristic vector of described sample brain wave, and calculate the first similarity between the characteristic vector of described sample brain wave and the characteristic vector of described target brain wave;
The brain stimulation sample producing described target brain wave for tester is chosen according to described first similarity.
2. brain wave according to claim 1 bring out method, it is characterized in that, adopt the training sample that presets to carry out brain stimulation to trainer, and gather the sample brain wave corresponding with described training sample and comprise:
Electrode for encephalograms is arranged on the position of the scalp supraoccipital bone knuckle of multiple trainer;
For trainer described in each, screen shows multiple described training sample successively according to each self-corresponding stimulation displaying time of multiple training sample, and the default lounge interval is set between multiple described training sample display;
According to the described lounge interval, the sample brain wave that trainer described in each produces for multiple described training sample is carried out segmentation intercepting, and the meansigma methods of training sample described in each being carried out the sample brain wave of brain stimulation generation to different trainer is as the average sample brain wave corresponding with training sample described in each;
By frequency spectrum analysis method, obtain the major frequency components of the average sample brain wave corresponding with training sample described in each, and using described major frequency components as the sample brain wave corresponding with training sample described in each.
3. brain wave according to claim 2 bring out method, it is characterized in that, comprise before obtaining the major frequency components of the average sample brain wave corresponding with training sample described in each:
Carry out pretreatment to the average sample brain wave corresponding with training sample described in each, described pretreatment comprises one or more operations in denoising, filtering, processing and amplifying.
4. bring out method according to the arbitrary described brain wave of claim 1-3, it is characterized in that, choose according to described first similarity the brain stimulation sample producing described target brain wave for tester and comprise:
Choose the training sample that sample brain wave corresponding to maximum described first similarity is corresponding, as the brain stimulation sample producing described target brain wave for tester.
5. bring out method according to the arbitrary described brain wave of claim 1-3, it is characterized in that, choose according to described first similarity the brain stimulation sample producing described target brain wave for tester and comprise:
The training sample choosing the sample brain wave corresponding with it corresponding by described first similarity order from large to small carries out brain stimulation to tester, and calculates the second similarity of the characteristic vector of the brain wave produced according to described training sample and the characteristic vector of described target brain wave simultaneously;
Choose described second similarity and be greater than training sample corresponding to default similarity threshold, as the brain stimulation sample producing described target brain wave for described tester.
6. brain wave according to claim 5 bring out method, it is characterized in that, the type of described training sample is any one or multiple combination in text, picture, expression, animation, audio frequency and video type.
7. an inducing device for brain wave, is characterized in that, comprising:
Training devices, for adopting the training sample preset to carry out brain stimulation to trainer, and gathers the sample brain wave corresponding with described training sample;
Target preinstall apparatus, for goal-selling brain wave, and extracts the characteristic vector of described target brain wave;
First similarity accountant, for extracting the characteristic vector of described sample brain wave, and calculates the first similarity between the characteristic vector of described sample brain wave and the characteristic vector of described target brain wave;
Brain stimulation sample selecting device, for choosing the brain stimulation sample producing described target brain wave for tester according to described first similarity.
8. the inducing device of brain wave according to claim 7, is characterized in that, described training devices comprises:
Electrode installing device, for being arranged on the position of the scalp supraoccipital bone knuckle of multiple trainer by electrode for encephalograms;
Training sample setting device, for for trainer described in each, screen shows multiple described training sample successively according to each self-corresponding stimulation displaying time of multiple training sample, and the default lounge interval is set between multiple described training sample display;
Average sample brain wave acquisition device, for according to the described lounge interval, the sample brain wave that trainer described in each produces for multiple described training sample is carried out segmentation intercepting, and the meansigma methods of training sample described in each being carried out the sample brain wave of brain stimulation generation to different trainer is as the average sample brain wave corresponding with training sample described in each;
Major frequency components acquisition device, for obtaining the major frequency components of the average sample brain wave corresponding with training sample described in each, and using described major frequency components as the sample brain wave corresponding with training sample described in each.
9. the inducing device of brain wave according to claim 8, is characterized in that, described major frequency components acquisition device also comprises:
Pretreatment unit, for carrying out pretreatment to the average sample brain wave corresponding with training sample described in each, described pretreatment comprises one or more operations in denoising, filtering, processing and amplifying.
10. the inducing device of brain wave according to claim 9, is characterized in that, described brain stimulation sample selecting device comprises:
Maximum similarity selecting device, for choosing training sample corresponding to sample brain wave corresponding to maximum described first similarity, as the brain stimulation sample producing described target brain wave for tester.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510570569.6A CN105125210A (en) | 2015-09-09 | 2015-09-09 | Brain wave evoking method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510570569.6A CN105125210A (en) | 2015-09-09 | 2015-09-09 | Brain wave evoking method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105125210A true CN105125210A (en) | 2015-12-09 |
Family
ID=54711061
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510570569.6A Pending CN105125210A (en) | 2015-09-09 | 2015-09-09 | Brain wave evoking method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105125210A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107997738A (en) * | 2017-10-31 | 2018-05-08 | 努比亚技术有限公司 | A kind of physiological single processing method, apparatus and computer-readable recording medium |
| CN111031057A (en) * | 2019-12-21 | 2020-04-17 | 北京理工大学 | Information transmission method for inducing brain waves based on stimulation signals |
| CN114707132A (en) * | 2021-05-20 | 2022-07-05 | 广州番禺职业技术学院 | Brain wave encryption and decryption method and system based on emotional voice |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6385486B1 (en) * | 1997-08-07 | 2002-05-07 | New York University | Brain function scan system |
| JP2007000280A (en) * | 2005-06-22 | 2007-01-11 | Toyota Motor Corp | Arousal lowering determination device |
| CN103258215A (en) * | 2013-05-10 | 2013-08-21 | 杭州电子科技大学 | Multi-lead correlation analysis electroencephalo-graph (EEG) feature extraction method |
| CN103699230A (en) * | 2014-01-14 | 2014-04-02 | 东南大学 | Digital interface interaction method on basis of icon electrocerebral control |
-
2015
- 2015-09-09 CN CN201510570569.6A patent/CN105125210A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6385486B1 (en) * | 1997-08-07 | 2002-05-07 | New York University | Brain function scan system |
| JP2007000280A (en) * | 2005-06-22 | 2007-01-11 | Toyota Motor Corp | Arousal lowering determination device |
| CN103258215A (en) * | 2013-05-10 | 2013-08-21 | 杭州电子科技大学 | Multi-lead correlation analysis electroencephalo-graph (EEG) feature extraction method |
| CN103699230A (en) * | 2014-01-14 | 2014-04-02 | 东南大学 | Digital interface interaction method on basis of icon electrocerebral control |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107997738A (en) * | 2017-10-31 | 2018-05-08 | 努比亚技术有限公司 | A kind of physiological single processing method, apparatus and computer-readable recording medium |
| CN111031057A (en) * | 2019-12-21 | 2020-04-17 | 北京理工大学 | Information transmission method for inducing brain waves based on stimulation signals |
| CN114707132A (en) * | 2021-05-20 | 2022-07-05 | 广州番禺职业技术学院 | Brain wave encryption and decryption method and system based on emotional voice |
| CN114707132B (en) * | 2021-05-20 | 2023-04-18 | 广州番禺职业技术学院 | Brain wave encryption and decryption method and system based on emotional voice |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Babadi et al. | A review of multitaper spectral analysis | |
| CN104257379A (en) | Electroencephalogram processing apparatus and method and sleep monitoring worn device | |
| CN109271020A (en) | A kind of stable state vision inducting brain-machine interface method of evaluating performance based on eye movement tracking | |
| CN102940490B (en) | Method for extracting motor imagery electroencephalogram signal feature based on non-linear dynamics | |
| US20200333378A1 (en) | Method and apparatus for processing oscilloscope signal and oscilloscope | |
| US20090030628A1 (en) | Signal analyser | |
| CN103070682B (en) | Extraction method and extraction device for direction of eye movement based on electro-oculogram signals | |
| CN104503593A (en) | Control information determination method and device | |
| CN110059564B (en) | Feature extraction method based on power spectral density and cross-correlation entropy spectral density fusion | |
| CN105125210A (en) | Brain wave evoking method and device | |
| CN105307095B (en) | A kind of high definition audio frequency measurement method based on FFT | |
| CN105096938A (en) | Method and device for obtaining user characteristic information of user | |
| CN109946512A (en) | A kind of dynamic power analysis method for improving frequency domain interpolation | |
| CN111671420B (en) | Method for extracting features from resting state electroencephalogram data and terminal equipment | |
| CN104783928B (en) | A kind of assessment system testing cochlear implant speech strategy performance | |
| CN114756121A (en) | Virtual reality interactive emotion detection and regulation system based on brain-computer interface | |
| CN103720471A (en) | Factor analysis based ocular artifact removal method | |
| CN104267380A (en) | Associated display method for full-pulse signal multidimensional parameters | |
| CN108260012B (en) | Electronic device, video playing control method and related product | |
| CN203965503U (en) | A kind of hyperchannel nV level alternating voltage checkout equipment | |
| Ali et al. | Time-frequency coherence of multichannel EEG signals: Synchrosqueezing transform based analysis | |
| CN111317467A (en) | Electroencephalogram signal analysis method and device, terminal device and storage medium | |
| CN114169366B (en) | Neural feedback training system and method | |
| Bian et al. | Research on brain-computer interface technology based on steady state visual evoked potentials | |
| CN107590788A (en) | A kind of remote sensing image processing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20151209 |
|
| WD01 | Invention patent application deemed withdrawn after publication |