CN103995582B - Brain-computer interface character input method and system based on steady-state visual evoked potential (SSVEP) - Google Patents
Brain-computer interface character input method and system based on steady-state visual evoked potential (SSVEP) Download PDFInfo
- Publication number
- CN103995582B CN103995582B CN201410171291.0A CN201410171291A CN103995582B CN 103995582 B CN103995582 B CN 103995582B CN 201410171291 A CN201410171291 A CN 201410171291A CN 103995582 B CN103995582 B CN 103995582B
- Authority
- CN
- China
- Prior art keywords
- character
- input
- brain
- led
- character input
- 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.)
- Expired - Fee Related
Links
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention provides a brain-computer interface character input method and system based on steady-state visual evoked potential (SSVEP). Input characters are divided into N1 character sets and a character and function mixture set, wherein characters arranged in the same order in the N1 character sets are selected through the same key. The system comprises an LED visual stimulator, an electroencephalography acquisition platform, a PC real-time processing system and a character input interface. The LED visual stimulator is used for generating visual stimulation at different frequencies. The electroencephalography acquisition platform is used for collecting electroencephalography signals in real time, and then the electroencephalography signals are subjected to amplifying, filtering and analog-digital conversion and then input into a PC. The PC real-time processing system is used for processing the received electroencephalography signals in real time, detecting SSVEP signal frequencies, identifying LEDs corresponding to the signal frequencies and transmitting instructions represented by the signal frequencies to the character input interface. The character input interface is used for conducting corresponding processing according to the types of the instructions and displaying characters to be input in a display area of a computer screen. The brain-computer interface character input method and system based on SSVEP has the advantages of being high in input speed, high in accuracy, good in robustness and free of user training.
Description
Technical field
The present invention relates to brain science technology and computer technology, it is more particularly related to brain-computer interface technology with
Character-input techniques.
Background technology
Human brain passes through neuromuscular channel communication with the outside world or controls external environment condition, and numerous disease may destroy these god
Through muscle passage.For example, amyotrophic lateral sclerosis, brain stem apoplexy, spinal cord injury, brain paralysis, op parkinson's, multiple hardening illness
And other numerous diseases can damage the neural channel controlling muscle or infringement muscle itself.Those are subject to the serious shadow of these diseases
The people ringing may lose autonomous muscle and control, thus produces dyskinesia, so that it cannot being communicated with the external world
Or control external equipment.
Brain-computer interface (Brain-Computer Interface, BCI) monitors the cerebration of user, understands the meaning of user
Figure, and the intention of user is converted to external command.As a kind of new, non-muscle communication port, BCI can make one direct
Expressed thoughts or commanding apparatus by brain, without by language or limb action.Suffer from for serious motion is disabled
Person, their intention can be sent to external device (ED), such as computer, household electrical appliance, care appliances and nerve prosthesis by BCI
Deng thus improving their quality of life.
Traditional characters input method, is either based on keyboard or mouse, is also based on hand-written or gesture, is required for perfecting
Finger just enable.Dyskinesia is suffered from for those so that the patient of double upper limb disability, these methods are all incompetent
For power.In order to liberate both hands, realize not relying on the character input of staff, people have invented and have been based on voice technology, eye tracking
Technology and the characters input method of brain-computer interface technology.The limitation of voice technology is easily to be affected by ambient noise, and sight line
The shortcoming of tracking technique is that sight line spot placement accuracy is low.In addition, gaze tracking system to having a mind to during user operation and cannot be not intended to
Target selection activity effectively distinguished, limit the practical application of this kind of system.
Different brain electricity (Electroencephalography, EEG) component of signals, for example slow cortical potential, motion are thought
As the mu/beta rhythm and pace of moving things, event correlation P300 current potential and VEP (the Visual Evoked that produce
Potential, VEP), can be used as the characteristic signal of BCI.Wherein, the BCI based on P300 current potential can provide enough
Control command, realizes a control command and inputs a character, thus most of existing brain spells device (Mental
Speller P300 current potential) is used as the characteristic signal of BCI.However, there are two intrinsic lacking in the BCI based on P300 current potential
Point:One is that the time required for one character of input is longer when the number of characters of need input is more, leads to low character input speed
Degree;Two are a lack of time upper robustness and user between, lead to low character input accuracy.Wherein first shortcoming can be led to
Cross other technologies to be made up, and second shortcoming is difficult to overcome, make to be limited in actual applications based on the BCI of P300
System.In these BCI implementation methods, the BCI based on VEP is due to having following four advantages, thus obtains increasing
Note and pay attention to:1) high rate of information transmission;2) train with little need for user;3) low user's rate of change;4) easy to use.
VEP reflects the Vision information processing mechanism of brain, is one kind response to flash stimulation for the human eye.According to repeat pierce
Sharp frequency is different, and VEP can be divided into transient state VEP and stable state VEP.When the repetition rate of visual stimulus is relatively low, continuously pierce twice
Swash the response causing and will not produce crossover, stimulate the response producing independent of former stimulation every time, this response is referred to as transient state
VEP(Transient VEP,TVEP);When the repetition rate of visual stimulus is higher than 6Hz, the continuously response of Induced by Stimulation several times
Meeting superposition together, makes Cortical Neurons granting synchronous with frequency of stimulation, and the rhythm and pace of moving things in occipital region and top area electrical activity of brain is bright
Aobvious enhancing, forms a kind of stable response, referred to as stable state VEP (Steady-State VEP, SSVEP).SSVEP has and vision
Frequency of stimulation identical fundamental frequency and its harmonic wave, its fundamental frequency can use specific signal processing algorithm to be detected.
The frequency of the amplitude of SSVEP response and phase reappearance visual pattern, intensity and structure, the repetition of this visual pattern
Both can have been produced by computer LCD display it is also possible to be produced by the light-emitting diode display independent of computer.Due to by screen
The restriction of refresh rate harmony wave frequency rate, computer LCD display at most can only produce 5 frequency of stimulation based on frequency coding
(i.e. control command).In order to increase the number that can input character, the character entry apparatus based on LCD require an adaptive plan
Slightly create a graphical user interface (Graphic User Interface, GUI), combine several orders according to this interface defeated
Enter a character, lead to low character input speed.
Comparatively speaking, an independent LED visual stimulator does not limit in structure, and is provided that and compares LCD display
Higher visual stimulus.In theory, LED visual stimulator can provide enough stimulation target, realizes an order input one
Individual character;In fact, in LED panel limited by an area, too many stimulation target (i.e. light-emitting diode display) leads to adjacent thorn
Interval between sharp target diminishes, and causes interfering between stimulating light source, has a strong impact on the identification of frequency of stimulation, thus dropping
The accuracy of low character input.Therefore, in LED panel, the number of stimulation target should think over, and the number of stimulation target should be
Balance is obtained in input speed and precision.
Presently, there are using the major technique that BCI realizes character input be BCI technology based on P300, based on sight line with
Track and the technology of P300, and the mixing BCI technology based on SSVEP and P300.For example, Chinese patent is " based on P300 brain electricity electricity
Input in Chinese BCI system (200710164418.6) of position ", Chinese character is split as 5 basic strokes by the Five-stroke Method, by inspection
Survey P300 current potential and input stroke needed for each Chinese character, realize Chinese text input;Chinese patent " a kind of based on eye tracking and
The character entry apparatus (200910080852.5) of P300 brain electric potential ", are limited centered on sight line point and need to input character place
Region, realize the input of character in limited area by detecting P300 current potential;Chinese patent is " dual using brain electricity time-frequency component
The quick character input method (201210013087.7) of positioning normal form ", the character that can input is divided into 4 regions by function,
Select the subregion that character is located by detecting SSVEP current potential, defeated by detecting that P300 current potential enters line character in selected region
Enter.These technology overcome first shortcoming of the BCI based on P300, i.e. the problem of character input speed, without overcoming it
Second shortcoming, i.e. sane sex chromosome mosaicism and user between in time.
Content of the invention
It is an object of the invention to provide a kind of characters input method based on Steady State Visual Evoked Potential brain-computer interface and be
System, is characterized in that high input speed, correct rate for input are high, system robustness is good, does not need user to train.
The present invention is achieved by the following technical solutions.
Characters input method based on Steady State Visual Evoked Potential brain-computer interface of the present invention is it is characterised in that adopt
With the decision tree character input rule of two steps, to realize more multicharacter input using a small amount of LED button and defeated
Enter a character or command function at most needs two selections.
Two described step decision tree input rules refer to, the K character that can input and command function are divided into K1 character
Group and a character mix group with function;Each character group includes K2 character Command, and mixing group include K3 character with
Command function;K1 character group constitutes a main menu, and wherein each character group can be selected with an entree single-button;Often
K2 character in individual character group constitutes a submenu, the identical that puts in order in K1 submenu character, can use same suitable
Sequence number button is selected;Character is mixed group and constitutes a separate menu, each character in separate menu or function with function
The available button of order is selected.So, K=K1 × K2+K3 character and function can be carried out with K1+K2+K3 button
Select it is achieved that with the less more multicharacter purpose of key-press input.
Described characters input method is:
To input character and previous input character not in same submenu, then user first have to select should
The entree single-button of character place character group, according still further to the order selecting sequence button of this character place character group;
The character of the character to input and previous input need not select this character in same submenu, then user
The entree single-button of place character group, directly according to the order selecting sequence button of this character place character group;
To the character in input separate menu and command function, directly inputted with a button respectively.
Character input system based on Steady State Visual Evoked Potential brain-computer interface of the present invention, regards including a LED
Feel stimulator, a brain wave acquisition platform, a PC real time processing system and a character input interface.Wherein, visual stimulus
Device is used for producing the visual stimulus of different frequency;Brain wave acquisition platform be used for Real-time Collection EEG signal, through amplification, filtering with
After analog-to-digital conversion, PC computer is inputted by data wire;PC real time processing system is used for the EEG signal receiving is located in real time
Reason, the frequency of detection SSVEP signal, identify the corresponding LED of this frequency, and the order being represented is sent to character input and connects
Mouthful;Character input interface is used for doing corresponding process according to the type of order, and the character being intended to input shows in computer screen
Show that region shows.
1st, described LED visual stimulator is made up of stimulator panel and stimulator control circuit, for producing different frequencies
The visual stimulus of rate.
Wherein, described stimulator panel, for simulating a character input keyboard being made up of N number of button.Stimulator
Panel is made up of the light-emitting diode display of N number of coloured light that turns white, and is arranged in the stimulation matrix of N1 × N2.Each LED is the square of 2cm × 2cm
Shape square.Interfere in order to avoid light source produces, between adjacent LED, should have enough distances.Two adjacent LED of the present invention
Horizontal and vertical distance is 2.5cm.6 light emitting diodes are comprised, composition is by 3 light-emitting diodes inside each light-emitting diode display
The double column structure of pipe series winding.
Wherein, described stimulator control circuit, produces circuit and LED current drive circuit including stimulus signal;Stimulate
Signal generating circuit is used for producing the square-wave signal of N road different frequency, and LED current drive circuit is used for each light-emitting diode display
Apply enough driving current, to ensure that LED has enough luminous intensities.
Stimulus signal produces circuit and is made up of CPLD chip and its peripheral circuit, and N road square-wave signal is by CPLD chip I/O end
Mouth output.Their mode of operation to be set by 4 toggle switch of CPLD periphery.The flashing rate of each LED can be passed through
Program setting and change, frequency range is 6Hz~20Hz, and the minimum interval between frequency is 0.2Hz.
By the N road square-wave signal of CPLD chip I/O port output, it is delivered to N paths of LEDs current driving circuit, applies electric current
Square-wave signal after driving is used for controlling light-emitting diode display to light.N road square-wave signal and current driving circuit are separate, mutual
Do not disturb.
2nd, described brain wave acquisition platform includes electrode cap and eeg amplifier, and electrode cap is used for gathering EEG signals, brain electricity
Amplifier is used for the EEG signals of collection are filtered, amplify and analog-to-digital conversion.Electrode cap includes forming EEG by 7 electrodes
Signal record passage, rests the head on domain positioned at brain.Place standard according to international 10/20 system, 7 signal electrodes are located at O1, O2 respectively,
Pz,P3,P4,P7,P8.Ground electrode is located at Fz, and reference electrode is located at left ear-lobe;Eeg amplifier multiplication factor is 20000, filtering
Frequency band is 0.01~30Hz, and the sampling rate of analog-to-digital conversion is 1024Hz.
3rd, described PC real time processing system execution system initialization, visual stimulator startup, data acquisition control, data connect
Receive and the operations such as preservation, Digital Signal Processing and control command output.PC real time processing system has two major functions:
One is that the working condition to whole system is controlled, and two is that eeg data is carried out with real-time processing, the operation life of identifying user
Order.System controlling software is based on C++ platform development.DSP program is write by Matlab language, by application program
Interface (Application Program Interface, API) is called by system control program.Eeg amplifier and computer
Between data transfer follow ICP/IP protocol.
Digital Signal Processing refers to, using special algorithm, EEG signals are carried out with real-time processing, detection SSVEP signal
Frequency, identifies the corresponding user command of this frequency.The character entry apparatus that the present invention provides use canonical correlation analysis
(Canonical Correlation Analysis, CCA) detects to the frequency of SSVEP in EEG signals.CCA is a kind of
, there is certain implicit related situation for two variables in Multivariable Statistical Methods.In this character entry apparatus, CCA method
For describing the correlation between stimulus signal and the EEG signals of record.
4th, the order that described character input interface PC real time processing system sends, is carried out accordingly according to the type of order
Process, including following three kinds of situations:
1) if receive is main menu commands, etc. submenu order to be received;
2) if receive is submenu order, it is combined with previous main menu commands, determines that user thinks
Character to be inputted, and this character is shown in computer screen viewing area;
3) if receive is separate menu order, judgement is needed to be character keys or function key.If character keys, then
Its corresponding character is shown in computer screen viewing area;If function key, then it is which kind of function key firstly the need of differentiation, so
After execute corresponding feature operation.
The basic thought of the present invention is, under conditions of using medium-scale stimulation target, using the judgement of two steps
Tree character input constructing tactics character input interface, so that one character of input at most needs two orders.Due to SSVEP current potential
Good stability, this character input system, on the premise of obtaining high input speed and accuracy, has high robustness,
And do not need user to train, make this character input system can serve vast motion as real application systems
Disabled patient.
Compared with prior art, the present invention has following beneficial effect:
1) character input system that the present invention provides, using SSVEP signal as the input signal of BCI system, has high word
Symbol input accuracy and the in time high robustness and user between;
2) character input system that the present invention provides inputs a character at most needs two orders, character input speed
Hurry up;
3) the character input system working stability that the present invention provides, user does not need training just can use.
Brief description
Fig. 1 is the character input system theory diagram based on Steady State Visual Evoked Potential brain-computer interface for the present invention.
Fig. 2 is the present invention two step decision tree character input rule schematic diagram.
Fig. 3 is LED visual stimulator panel schematic diagram of the present invention.
Fig. 4 is that LED visual stimulator control signal of the present invention produces circuit theory diagrams.
Fig. 5 is LED visual stimulator current driving circuit schematic diagram of the present invention.
Fig. 6 is character input system EEG signals recording electrode location map of the present invention.
Fig. 7 is character input system PC real time processing system program flow diagram of the present invention.
Fig. 8 is that character input system of the present invention uses the method that CCA detects SSVEP frequency.X is multichannel brain electric signal,For frequency of stimulation fkReference signal.
Fig. 9 is character input system character input interface routine flow chart of the present invention.
Specific embodiment
For making the objects, technical solutions and advantages of the present invention clearer, clear and definite, the embodiment that develops simultaneously referring to the drawings is entered
One step describes in detail.It should be appreciated that the application of the present invention is not limited to following citings, one of skill in the art is come
Say, can be improved according to the above description or convert, all these modifications and variations broadly fall into claims of the present invention
Protection domain.
As shown in figure 1, a LED vision thorn is included based on the character input system of Steady State Visual Evoked Potential brain-computer interface
Sharp device, a brain wave acquisition platform, a PC real time processing system and a character input interface.Wherein, visual stimulator by
16 light-emitting diode display compositions, they are lighted with different frequencies simultaneously.When user watches certain LED attentively, in brain occipitalia region
SSVEP will be produced, its fundamental frequency is identical with the glow frequency of this LED;Brain wave acquisition platform Real-time Collection EEG signal, warp
After crossing amplification, filtering and analog-to-digital conversion, PC computer is inputted by data wire;PC real time processing system is to the EEG signal receiving
Carry out real-time processing, the frequency of detection SSVEP signal, identify the corresponding LED of this frequency, and the order being represented is sent to word
Symbol input interface.Character input interface does corresponding process according to the type of order, and is intended to the character inputting in computer screen
Curtain viewing area shows.
As shown in Fig. 2 the character that can input and function are chosen as 35, including 26 English alphabet A-Z and 1 space word
Symbol5 conventional english punctuation marks:Comma (), fullstop (.), question mark (?), quotation marks (") with colon (:), and 3 work(
Can key:← (backspace),(carriage return) and CL (caps lock).In order to reduce the number of LED button on visual stimulator, character is defeated
Enter the decision tree rule using two steps.This 35 selections are divided into 5 character group and 1 character mixes group with function.Often
One character group includes 6 characters and selects, and mixing group is included 5 characters and selected with function.This 5 character group use label respectively
A-F、G-L、M-R、S-X、Y-:Represent, they constitute a main menu, are arranged in figure the first row.Each character group can be with one
Individual LED button is selected, and has selected on-off action to character therein;6 characters in each character group constitute a son
Menu, put in order in 5 submenus identical character, can use same button to be selected;Character mixes group structure with function
A separate menu, each character or function is become to be selected using a button.Therefore, this 35 characters and command function can
Selected with 5+6+5=16 button, thus greatly reducing the number of stimulator face onboard led.
As shown in figure 3, each rectangular block represents a light-emitting diode display, 16 LED are distributed in visual stimulator panel
On, constitute the stimulation matrix of 4 × 4.Character in each square is the label of this LED button.Label be A-F, G-L,
M-R、S-X、Y-:5 LED buttons based on menu key.Each entree single-button is shown below 6 of its corresponding submenu
Character, to facilitate user to select;Put in order in 5 submenus identical character, can use same button to be selected.Cause
This, 6 in each submenu character according to put in order can respectively using label be 1,2,3,4,5,6 LED button carries out
Select;Label is.,←、The LED button of CL is respectively used to select 5 characters in Fig. 2 separate menu and function life
Order.
According to the decision tree input rule of this two steps, the character in a submenu to be selected, typically will first select it
Corresponding entree singly-bound.If the character of to be selected or continuously several characters and previous choosing belongs to same sub- dish
Single, then can directly select this character or the several characters of continuous selection in submenu, need not move through entree singly-bound and select.Example
As, when user wants to input word ' bad ', because three letters in this word are all in first submenu of Fig. 2,
He only needs to select entree single-button ' A-F ' once before input alphabetical ' b ', then continuously watch attentively label be 2,1,4 LED
Button.Therefore, the decision tree input method of this two steps, reduces input step under many circumstances, shortens character defeated
The time entering.5 buttons corresponding with separate menu are used for prompt operation, and an available button is directly selected.Therefore, originally
Invention one character of input at most needs two orders, has input speed faster.
As shown in Figure 4, Figure 5, stimulator control circuit is by LED modulated signal producing circuit and LED current drive circuit group
Become.Fig. 4 LED modulated signal producing circuit is made up of CPLD chip EMP1270T144 and its peripheral circuit, produces 16 tunnel modulation
Depth is 1/2 square-wave signal, by I/O port S1~S16 output.Port CLK0 (pin 18) and crystal oscillator LTC6905-
80 are connected, the LED modulated signal for producing the square-wave signal of 48MHz, required for frequency dividing can produce.The flash of light frequency of LED
Rate can pass through program setting and change, and frequency range is 6Hz~20Hz, and the minimum interval between frequency is 0.2Hz.Dial-up is opened
Close SW1~SW4 to can be used for setting the mode of operation of 16 road square-wave signals.Square-wave modulation signal S1~S16 exports respectively to a road
Current driving circuit.Fig. 5 gives a road current driving circuit schematic diagram.Square-wave signal after driving through electric current, controls 16
LED is glistened with different frequencies and different time series patterns, and guarantees there is enough luminous intensities.16 road current driving circuits
Separate, do not interfere with each other.
As shown in fig. 6, electrode cap is used for gathering EEG signals, 7 electrodes positioned at occipitalia region are remembered as EEG signals
Record passage, electrode is placed according to international 10/20 system, and 7 signal electrode positions are respectively O1, O2, Pz, P3, P4, P7, P8, ground
Electrode is located at Fz, and reference electrode is located at left ear-lobe.Contact with scalp well for guarantee electrode, inject conductive special in electrode jack
The good conducting resinl of property, resistance is below 5 kilo-ohms.
Experimenter expresses operation intention by watching specific LED attentively, and corresponding EEG signals gather through electrode cap, then warp
After eeg amplifier amplification, filtering and analog-to-digital conversion, it is transferred to PC real time processing system.The multiplication factor of eeg amplifier is
20000 times, filtered band is 0.01~30Hz, and the sampling rate of analog-to-digital conversion is 1024Hz.
As shown in fig. 7, PC real time processing system is realized by software in a computer, its process step is followed successively by system
Initialization, visual stimulator startup, data acquisition control, data receiver and preservation, Digital Signal Processing and control command are defeated
Go out.PC real time processing system has two major functions:One is that the working condition to whole system is controlled, and two is to brain electricity
Data carries out real-time processing, the operational order of identifying user.System controlling software is based on C++ platform development.Digital Signal Processing
Program is write by Matlab language, is called by system control program by application programming interfaces (API).Eeg amplifier and calculating
Data transfer between machine follows ICP/IP protocol.
Digital Signal Processing refers to using canonical correlation analysis (Canonical Correlation Analysis, CCA)
Digitized EEG signals are carried out with real-time processing, the frequency of detection SSVEP signal, identifies the corresponding light-emitting diode display of this frequency
And its user command representing.
Eeg data length for real-time processing is 2 seconds, and moving step length is 0.5 second.In order to Further aim identification can
By property, only double same frequency of stimulation is detected when, SSVEP frequency identification is just effective.
As shown in figure 8, identify comprising the following steps that of the frequency of SSVEP signal using CCA algorithm:
1) determine reference signal:It is assumed that there is frequency of stimulation to be respectively f1,f2,…,fKK target.X and YfRepresent two
The stochastic variable of individual multidimensional, wherein X are NtSecond long multichannel brain electric signal;YfRepresent and X length identical reference signal.Should
Reference signal is the column vector that a sine by frequency of stimulation f and its harmonic wave and cosine form
Yf=(sin (2 π ft), cos (2 π ft) ..., sin (2 π Nhft),cos(2πNhft))T(1)
N in formulahIt is the number of harmonic wave, N in the present embodimenth=3.
2) CCA coefficient is calculated to all frequency of stimulation:One of multichannel brain electric signal X and reference signalCalculate as CCA
The input of method, calculates CCA coefficient to each frequency of stimulation of this LED visual stimulator.
Consider a pair linear combination x=XTWxWith y=YTWy.The effect of CCA is to find weight vector WxWith Wy, make between x and y
Related maximization.In other words, optimization problem below (2) can solve under two constraintss (3) with (4)
3) determining user command:With WxAnd WyCorresponding maximum ρ is maximum canonical correlation.X and Y is respectively in WxAnd WyOn
Projection, i.e. x and y, be referred to as canonical variable.The canonical correlation ρ of outputkCan be used for SSVEP frequency identification.There is greatest coefficient
The frequency of ρ is judged as the order C of user, and available formula is expressed as follows
ρ in formulakIt is EEG signals in frequency of stimulation fkCCA coefficient, K is the number of stimulation target.
As shown in figure 9, character input interface is realized by software in a computer, send for receiving PC real time processing system
Order, and according to order type processed accordingly, including following three situation:
1) if receive is main menu commands, etc. submenu order to be received;
2) if receive is submenu order, determine that user wants the character inputting in conjunction with main menu commands, and will
This character shows in computer screen viewing area;
3) if receive is separate menu order, judgement is needed to be character keys or function key.If character keys, then
Its corresponding character is shown in computer screen viewing area;If function key, then it is which kind of function key firstly the need of differentiation, so
After execute one of following three kinds of feature operations:
If 1. backspace key (←), then the character manipulation that recently enters is deleted in execution;
If 2. enter keyThe then line feed operation of one new paragraph of execution input;
If 3. caps lock key (CL), then execution is transformed into the uppercase operation of input.
Character input accuracy rate and character input speed are to evaluate two the key technical indexes of a character input system.
In order to evaluate the performance of the character input system based on Steady State Visual Evoked Potential brain-computer interface for the present invention, seven experimenters participate in
One experiment based on the input English character, word and sentence of the present embodiment.This seven experimenters have successfully completed
This test experiments, their average character input accuracys rate and character input speed is respectively 95.8% and 7.0 characters/point
Clock.This test result indicate that, this system has high performance compared with the homogeneous system based on brain-computer interface.
Claims (8)
1. a kind of characters input method based on Steady State Visual Evoked Potential brain-computer interface, is characterized in that the decision tree using two steps
Character input rule completes character input;
Two described step decision tree input rules are that the K character that can input and command function are divided into K1 character group and one
Character mixes group with function;Each character group includes K2 character Command, and mixing group includes K3 character and command function;
K1 character group constitutes a main menu, and wherein each character group can be selected with an entree single-button;Each character group
In K2 character constitute a submenu, the identical that puts in order in K1 submenu character, available same serial number button
Selected;Character is mixed group and constitutes a separate menu with function, and each character in separate menu or command function can use
One button is selected;
The process of character input is:
If not in same submenu, user selects this word to the character of the character that will input and previous input first
The entree single-button of symbol place character group, according still further to the order selecting sequence button of this character place character group;
If in same submenu, user need not select this character to the character of the character that will input and previous input
The entree single-button of place character group, directly according to the order selecting sequence button of this character place character group;
Character in separate menu and command function, are directly inputted with a button respectively as shortcut command.
2. described in a kind of claim 1 based on what the characters input method of Steady State Visual Evoked Potential brain-computer interface realized it is
System, is characterized in that, including a LED visual stimulator, a brain wave acquisition platform, a PC real time processing system and a word
Symbol input interface;Wherein, described LED visual stimulator, for producing the visual stimulus of different frequency;Described brain wave acquisition
Platform, for Real-time Collection EEG signal, after amplification, filtering and analog-to-digital conversion, inputs PC computer by data wire;Institute
The PC real time processing system stated, for the EEG signal receiving is carried out with real-time processing, the frequency of detection SSVEP signal, identification should
The corresponding LED of frequency, and the order being represented is sent to character input interface;Described character input interface, for basis
The type of order does corresponding process, and the character being intended to input shows in computer screen viewing area.
3. a kind of character input system based on Steady State Visual Evoked Potential brain-computer interface according to claim 2, it is special
Levy the LED visual stimulator being described and include stimulator panel and stimulator control circuit;
Wherein, described stimulator panel, for simulating a character input keyboard being made up of N number of button;
Wherein, described stimulator control circuit, for producing N road different frequency, modulation depth be 1/2 square-wave signal, control
Make N number of LED to glisten with different frequencies and different sequential.
4. a kind of character input system based on Steady State Visual Evoked Potential brain-computer interface according to claim 2, it is special
Levy the brain wave acquisition platform being described and include electrode cap and eeg amplifier;
Wherein, described electrode cap, for gathering scalp EEG signals;Form EEG signal recording channel including by 7 electrodes,
Positioned at brain occipitalia region;Place standard according to international 10/20 system, 7 signal electrodes are located at O1, O2, Pz, P3 respectively,
P4, P7, P8, one ground electrode is located at Fz, and a reference electrode is located at left ear-lobe;
Wherein, described eeg amplifier, for being filtered, amplifying and analog-to-digital conversion to the EEG signals gathering.
5. a kind of character input system based on Steady State Visual Evoked Potential brain-computer interface according to claim 2, it is special
Levy the PC real time processing system being described to include working state of system is carried out with real-time control and eeg data is located in real time
Reason;
Wherein, described real-time control is carried out to working state of system, start for execution system initialization, visual stimulator,
Data acquisition control, data receiver and preservation and control command output function;
Wherein, described real-time processing is carried out to eeg data, for detecting the frequency of SSVEP signal, identify that this frequency corresponds to
Light-emitting diode display and its representative user command.
6. a kind of character input system based on Steady State Visual Evoked Potential brain-computer interface according to claim 2, it is special
Levying is described character input interface, for receiving the order of PC real time processing system transmission, carries out phase according to the type of order
The process answered, including following three situation:
If receive is main menu commands, etc. submenu order to be received;
If receive is submenu order, determine that user wants the character inputting in conjunction with main menu commands, and by this character
Show in computer screen viewing area;
If receive is separate menu order, judgement is needed to be character keys or function key;If character keys, then it is right
The character answered shows in computer screen viewing area;If function key, then execute corresponding feature operation.
7. a kind of character input system based on Steady State Visual Evoked Potential brain-computer interface according to claim 3, it is special
Levying is described stimulator panel, byThe light-emitting diode display composition of the individual coloured light that turns white, is arranged in the stimulation matrix of N1 × N2;Often
Individual LED is the rectangular block of 2cm × 2cm, plays the effect of a button;The horizontal and vertical distance of two adjacent LED is
2.5cm;6 light emitting diodes are comprised, composition is by the double column structure of 3 light emitting diode series windings inside each light-emitting diode display.
8. a kind of character input system based on Steady State Visual Evoked Potential brain-computer interface according to claim 3, it is special
Levying is described stimulator control circuit, produces circuit and LED current drive circuit including stimulus signal;
Wherein, described stimulus signal produces circuit, for producing the square-wave signal of N road different frequency;This circuit is by CPLD core
Piece and its peripheral circuit are constituted, and N road square-wave signal is by CPLD chip I/O port output;Their mode of operation is by CPLD periphery
4 toggle switch setting;The flashing rate of each LED can pass through program setting and change, frequency range be 6Hz~
20Hz, the minimum interval between frequency is 0.2Hz;
Wherein, described LED current drive circuit, for applying enough driving currents to each light-emitting diode display, to ensure
LED has enough luminous intensities;By the N road square-wave signal of CPLD chip I/O port output, it is delivered to N paths of LEDs electric current and drives
Circuit, applies the square-wave signal after electric current drives and is used for controlling light-emitting diode display to light;N road square-wave signal and electric current drive electricity
Road is separate, do not interfere with each other.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410171291.0A CN103995582B (en) | 2014-04-25 | 2014-04-25 | Brain-computer interface character input method and system based on steady-state visual evoked potential (SSVEP) |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410171291.0A CN103995582B (en) | 2014-04-25 | 2014-04-25 | Brain-computer interface character input method and system based on steady-state visual evoked potential (SSVEP) |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103995582A CN103995582A (en) | 2014-08-20 |
CN103995582B true CN103995582B (en) | 2017-02-15 |
Family
ID=51309770
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410171291.0A Expired - Fee Related CN103995582B (en) | 2014-04-25 | 2014-04-25 | Brain-computer interface character input method and system based on steady-state visual evoked potential (SSVEP) |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103995582B (en) |
Families Citing this family (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2627075C1 (en) * | 2016-10-28 | 2017-08-03 | Ассоциация "Некоммерческое партнерство "Центр развития делового и культурного сотрудничества "Эксперт" | Neuro computer system for selecting commands based on brain activity registration |
CN106774843A (en) * | 2016-11-23 | 2017-05-31 | 同济大学 | A kind of cipher input equipment and method |
CN106681494B (en) * | 2016-12-07 | 2020-08-11 | 华南脑控(广东)智能科技有限公司 | Environment control method based on brain-computer interface |
CN106802723A (en) * | 2017-01-18 | 2017-06-06 | 西安电子科技大学 | A kind of Two bors d's oeuveres Chinese input system based on Steady State Visual Evoked Potential |
CN107037889B (en) * | 2017-03-06 | 2020-04-03 | 南昌大学 | Natural writing character input method and system based on visual brain-computer interface |
CN107212883B (en) * | 2017-05-24 | 2019-10-18 | 天津理工大学 | A kind of mechanical arm writing device and control method based on brain electric control |
CN107296586A (en) * | 2017-06-20 | 2017-10-27 | 黄涌 | Collimation error detection device/method and writing system/method based on the equipment |
CN107291239B (en) * | 2017-06-29 | 2020-03-31 | 清华大学 | Visual stimulation system, visual stimulation generation method and BCI system |
CN107390869B (en) * | 2017-07-17 | 2019-07-02 | 西安交通大学 | Efficient brain control Chinese character input method based on movement vision Evoked ptential |
US11266849B2 (en) * | 2017-12-12 | 2022-03-08 | Eb Neuro S.P.A. | Control device and a machine for interactive cerebral and bodily navigation with real-time anatomical display and control functions |
CN107967064B (en) * | 2018-01-22 | 2020-02-28 | 南京邮电大学 | Intelligent home control system based on brain-computer interface and laser-assisted aiming |
CN110262658B (en) * | 2019-06-14 | 2023-02-03 | 长春理工大学 | Brain-computer interface character input system based on enhanced attention and implementation method |
CN110379376B (en) * | 2019-07-04 | 2020-11-24 | 北京航空航天大学 | Liquid crystal display for SSVEP and stimulation pattern display method thereof |
CN113934300B (en) * | 2021-10-19 | 2022-08-23 | 北京烽火万家科技有限公司 | Equipment control method and device based on brain-computer interface, electronic equipment and medium |
CN114081510B (en) * | 2021-11-22 | 2024-07-23 | 西安交通大学 | Motor imagery brain-computer interface method based on stress motor-vision multisource stimulation |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102609090A (en) * | 2012-01-16 | 2012-07-25 | 中国人民解放军国防科学技术大学 | Electrocerebral time-frequency component dual positioning normal form quick character input method |
CN103607519A (en) * | 2013-10-17 | 2014-02-26 | 南昌大学 | Brain-computer interface-based telephone system for double-upper limb disabled people |
CN103699217A (en) * | 2013-11-18 | 2014-04-02 | 南昌大学 | Two-dimensional cursor motion control system and method based on motor imagery and steady-state visual evoked potential |
-
2014
- 2014-04-25 CN CN201410171291.0A patent/CN103995582B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102609090A (en) * | 2012-01-16 | 2012-07-25 | 中国人民解放军国防科学技术大学 | Electrocerebral time-frequency component dual positioning normal form quick character input method |
CN103607519A (en) * | 2013-10-17 | 2014-02-26 | 南昌大学 | Brain-computer interface-based telephone system for double-upper limb disabled people |
CN103699217A (en) * | 2013-11-18 | 2014-04-02 | 南昌大学 | Two-dimensional cursor motion control system and method based on motor imagery and steady-state visual evoked potential |
Also Published As
Publication number | Publication date |
---|---|
CN103995582A (en) | 2014-08-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103995582B (en) | Brain-computer interface character input method and system based on steady-state visual evoked potential (SSVEP) | |
CN107037889B (en) | Natural writing character input method and system based on visual brain-computer interface | |
CN102609090B (en) | Electrocerebral time-frequency component dual positioning normal form quick character input method | |
Gao et al. | A BCI-based environmental controller for the motion-disabled | |
CN103793058B (en) | A kind of active brain-computer interactive system Mental imagery classification of task method and device | |
CN104799984A (en) | Assistance system for disabled people based on brain control mobile eye and control method for assistance system | |
CN103995583B (en) | Realize the device and method thereof at the brain-computer interface interface for P300 compositions | |
CN103699226A (en) | Tri-modal serial brain-computer interface method based on multi-information fusion | |
CN102654793B (en) | Electrocerebral-drive high-reliability control system based on dual-mode check mechanism | |
CN107390869B (en) | Efficient brain control Chinese character input method based on movement vision Evoked ptential | |
CN105824418A (en) | Brain-computer interface communication system based on asymmetric visual evoked potential | |
CN101515199A (en) | Character input device based on eye tracking and P300 electrical potential of the brain electricity | |
CN102799267B (en) | Multi-brain-computer interface method for three characteristics of SSVEP (Steady State Visual Evoked Potential), blocking and P300 | |
CN101464728A (en) | Human-machine interaction method with vision movement related neural signal as carrier | |
CN109582131A (en) | The asynchronous mixing brain-machine interface method of one kind and system | |
Cao et al. | A high rate online SSVEP based brain-computer interface speller | |
CN103607519A (en) | Brain-computer interface-based telephone system for double-upper limb disabled people | |
CN108509040A (en) | Mixing brain machine interface system based on multidimensional processiug and adaptive learning | |
KR101768106B1 (en) | Apparatus and method for speller based on brain-computer interface | |
CN110688013A (en) | English keyboard spelling system and method based on SSVEP | |
Edlinger et al. | A hybrid brain-computer interface for improving the usability of a smart home control | |
CN101339413A (en) | Switching control method based on brain electric activity human face recognition specific wave | |
Yu et al. | A P300-based brain–computer interface for Chinese character input | |
CN107272880A (en) | A kind of cursor positioning, cursor control method and device | |
Jia | Neural network in the application of EEG signal classification 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 | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20200619 Address after: No.369 Shiqiao Road, Changdong Industrial Zone, Nanchang City, Jiangxi Province Patentee after: Jiangxi Chiba Color Printing Co.,Ltd. Address before: 999 No. 330031 Jiangxi province Nanchang Honggutan University Avenue Patentee before: Lattice Power (Jiangxi) Corp. |
|
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170215 Termination date: 20210425 |