CN110174938A - A kind of method and system of data input - Google Patents
A kind of method and system of data input Download PDFInfo
- Publication number
- CN110174938A CN110174938A CN201910304787.3A CN201910304787A CN110174938A CN 110174938 A CN110174938 A CN 110174938A CN 201910304787 A CN201910304787 A CN 201910304787A CN 110174938 A CN110174938 A CN 110174938A
- Authority
- CN
- China
- Prior art keywords
- sensing
- signal acquisition
- acquisition module
- input
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/017—Gesture based interaction, e.g. based on a set of recognized hand gestures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
- G06F3/023—Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
- G06F3/0233—Character input methods
Abstract
The present invention is suitable for technical field of data processing, provides a kind of method and system of data input, comprising: the stimulus to the sense organ module generates according to preset block partitioning algorithm and exports the numeric keypad comprising multiple input blocks；The body-sensing signal acquisition module obtains the body-sensing data of user, and the input block of the body-sensing data correlation is identified as target block；The input instruction that the stimulus to the sense organ module includes based on the target block, generates instruction selection interface；The electroencephalogramsignal signal acquisition module obtains the EEG signals that the user is fed back based on described instruction selection interface；The data processing module parses the EEG signals, determines target instruction target word selected by the user.The present invention ensure that input instruction is multifarious simultaneously, can be by the corresponding target block of body-sensing data activation, and screening a large amount of invalid instructions to improve the accuracy of following target instructions selection ensure that the handling capacity of brain machine signal.
Description
Technical field
The invention belongs to the method and apparatus that technical field of data processing more particularly to a kind of data input.
Background technique
With the continuous development of human-computer interaction technology, other than common body feeling interaction means, technology today is had been able to
Realize that brain-machine interaction, brain-machine interaction mainly pass through brain-computer interface, obtain the EEG signals of user, to carry out to EEG signals
Parsing, determines the corresponding interactive instruction of EEG signals, realizes and directly controlled by human brain to external equipment.However due to brain
Electrical signal intensity is smaller, obtains that difficulty is larger, and that there are handling capacities is small for existing brain electricity input technology, inputs the lower feelings of precision
Condition.
Existing brain-machine interaction technology can improve the accuracy rate of brain-machine interaction, by eye-tracking technology in order to obtain
The eye movement for getting user needs user additionally to wear the biggish eye-tracking equipment of volume, can cause the discomfort of user,
It can be seen that two aspects of input accuracy and interactive experience can not be balanced simultaneously in the prior art, brain-machine interaction is affected
The popularization of technology.
Summary of the invention
In view of this, the embodiment of the invention provides a kind of method and apparatus of data input, to solve existing brain machine
Interaction technique can improve the accuracy rate of brain-machine interaction by eye-tracking technology, in order to get the eye movement of user,
It needs user additionally to wear the biggish eye-tracking equipment of volume, the discomfort of user can be caused, it can be seen that, in the prior art
The problem of two aspects of input accuracy and interactive experience can not be balanced simultaneously, affect the popularization of brain-machine interaction technology.
The first aspect of the embodiment of the present invention provides a kind of method of data input, applied to the system of data input,
The data entry system includes at least: electroencephalogramsignal signal acquisition module, body-sensing signal acquisition module, stimulus to the sense organ module and number
According to processing module；
The method of data input includes:
The stimulus to the sense organ module generates according to preset block partitioning algorithm and exports the number comprising multiple input blocks
Word keyboard；Each input block includes at least one input instruction；
The body-sensing signal acquisition module obtains the body-sensing data of user, and by the input block of the body-sensing data correlation
It is identified as target block；
The input instruction that the stimulus to the sense organ module includes based on the target block, generates instruction selection interface；
The electroencephalogramsignal signal acquisition module obtains the EEG signals that the user is fed back based on described instruction selection interface；
The data processing module parses the EEG signals, determines target instruction target word selected by the user.
The second aspect of the embodiment of the present invention provides a kind of system of data input, comprising: electroencephalogramsignal signal acquisition module,
Body-sensing signal acquisition module, stimulus to the sense organ module and data processing module；
The stimulus to the sense organ module, for generating and exporting comprising multiple input areas according to preset block partitioning algorithm
The numeric keypad of block；Each input block includes at least one input instruction；
The body-sensing signal acquisition module, for obtaining the body-sensing data of user, and by the defeated of the body-sensing data correlation
Enter block and is identified as target block；
The stimulus to the sense organ module, the input instruction for including based on the target block, generates instruction selection
Interface；
The electroencephalogramsignal signal acquisition module, the brain telecommunications fed back for obtaining the user based on described instruction selection interface
Number；
The data processing module determines target instruction target word selected by the user for parsing the EEG signals.
The method and system for implementing a kind of data input provided in an embodiment of the present invention have the advantages that
The embodiment of the present invention passes through stimulus to the sense organ module by the way that numeric keypad to be divided into multiple and different input blocks
Numeric keypad after the division is exported, user can be chosen from multiple input blocks by body-sensing signal acquisition module
Corresponding target block, then stimulus to the sense organ module can generate instruction selection circle according to the input instruction that target block includes
Face is guaranteeing that input instruction is multifarious while can be improved the accuracy rate of input operation, is then passing through eeg signal acquisition mould
Block obtains the EEG signals of user, and is parsed by data processing module to the brain electric information, and identification obtains selected by user
Target instruction target word, realize and external equipment directly controlled by EEG signals.Compared with existing brain-machine interaction technology, this
Invention wears ophthalmogyric device without user, to reduce the volume of entire data entry system, since the acquisition of body-sensing data is set
Standby equipment volume is smaller, and the body-sensing data of user can be obtained by all kinds of portable wearable devices such as gloves, wrist, thus
The operating experience of user is improved, at the same time, numeric keypad can be divided into multiple input blocks by the system of data input,
It ensure that input instruction is multifarious simultaneously, can be by the corresponding target block of body-sensing data activation, it is a large amount of invalid to screen
Instruction, thus improve following target instructions selection accuracy, ensure that the handling capacity of brain machine signal.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is a kind of interaction diagrams of the method for data input that first embodiment of the invention provides；
Fig. 2 is the schematic diagram for the numeric keypad that one embodiment of the invention provides；
Fig. 3 is a kind of schematic diagram for instruction selection interface that one embodiment of the invention provides；
Fig. 4 be one embodiment of the invention provide electroencephalogramsignal signal acquisition module in include electrode configuration figure；
Fig. 5 is a kind of single pass signal amplification circuit that one embodiment of the invention provides；
Fig. 6 is a kind of method S102 specific implementation flow chart for data input that second embodiment of the invention provides；
Fig. 7 is a kind of method S104 specific implementation flow chart for data input that third embodiment of the invention provides；
Fig. 8 is a kind of method S104 specific implementation flow chart for data input that fourth embodiment of the invention provides；
Fig. 9 is a kind of method specific implementation flow chart for data input that fifth embodiment of the invention provides；
Figure 10 is a kind of method S105 specific implementation flow chart for data input that sixth embodiment of the invention provides；
Figure 11 is a kind of method specific implementation flow chart for data input that seventh embodiment of the invention provides；
Figure 12 is a kind of structural block diagram of the system for data input that one embodiment of the invention provides.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
The embodiment of the present invention passes through stimulus to the sense organ module by the way that numeric keypad to be divided into multiple and different input blocks
Numeric keypad after the division is exported, user can be chosen from multiple input blocks by body-sensing signal acquisition module
Corresponding target block, then stimulus to the sense organ module can generate instruction selection circle according to the input instruction that target block includes
Face is guaranteeing that input instruction is multifarious while can be improved the accuracy rate of input operation, is then passing through eeg signal acquisition mould
Block obtains the EEG signals of user, and is parsed by data processing module to the brain electric information, and identification obtains selected by user
Target instruction target word, realize and external equipment directly controlled by EEG signals, solve existing brain-machine interaction technology, can
To improve the accuracy rate of brain-machine interaction by eye-tracking technology, in order to get the eye movement of user, user's volume is needed
The outer biggish eye-tracking equipment of wearing volume, can cause the discomfort of user, it can be seen that, it can not put down simultaneously in the prior art
The problem of weigh two aspects of input accuracy and interactive experience, affects the popularization of brain-machine interaction technology.
In embodiments of the present invention, the executing subject of process is the system of data input, and the system of data input is at least
It include: electroencephalogramsignal signal acquisition module, body-sensing signal acquisition module, stimulus to the sense organ module and data processing module.At the data
Reason module includes but is not limited to: server, computer, smart phone and tablet computer etc. have setting for data-handling capacity
It is standby.Fig. 1 shows the interaction diagrams of the method for the data input of first embodiment of the invention offer, and details are as follows:
In S101, the stimulus to the sense organ module is generated and is exported comprising multiple defeated according to preset block partitioning algorithm
Enter the numeric keypad of block；Each input block includes at least one input instruction.
In the present embodiment, stimulus to the sense organ module according to the demand of user, can be defeated by different stimulus to the sense organ modes
Numeric keypad out.For example, if stimulus to the sense organ module be a kind of visual stimulus module, can by the devices such as VR/AR glasses come
Visual stimulus is carried out to user；If stimulus to the sense organ module is a kind of auditory stimulation module, the external speakers such as loudspeaker can be passed through
To carry out auditory stimulation to user.The type of the stimulus to the sense organ includes but is not limited to: visual stimulus, auditory stimulation, olfactory stimulation,
The stimulus types such as haptic stimulus.User can correspond to the stimulus to the sense organ module of stimulus type by wearing, to select suitably to feel
Official's stimulus type.The system of data input can obtain the module id of the stimulus to the sense organ module first on startup, thus really
The fixed corresponding stimulus to the sense organ type of the module id, so that it is determined that the division mode and output mode of numeric keypad, generate defeated
Script out, then according to the relevant operation of output script execution S101.
In the present embodiment, comprising optional input instruction in the numeric keypad, therefore stimulus to the sense organ module is to the number
Keyboard is divided, and after obtaining multiple input blocks, each input block can also determine the input area according to institute overlay area
The corresponding input instruction of block, each input block include at least an input instruction.It should be noted that input instruction can be with
For data input instruction, each input instruction can correspond to single character, the character string that can also be constituted for multiple character combinations；
Input instruction can also be the operational order that is controlled the component of external equipment, and external equipment refers to receiving the operation
After order, specified operation, such as starting device, pass hull closure etc. can be completed.
Optionally, in the present embodiment, stimulus to the sense organ module can carry out numeric keypad according to the quantity of input block
Equal division, the mode of specific equal division can be with are as follows: obtain the region area of numeric keypad, and based on the region area and
The quantity of block is inputted, the block area of each input block is calculated, and numeric keypad is divided based on the block area.
The total number of instructions that numeric keypad includes can also be obtained, and according to total number of instructions and the quantity of input block, is calculated each
It inputs the corresponding block of block and instructs number, and instruct the input of number corresponding number to refer to the block for each input block arrangement
It enables.
Optionally, in the present embodiment, due in S102, the system of data input can be determined according to body-sensing data with
Corresponding target block, i.e. each body-sensing type can be associated with an input block.Therefore, it is drawn to improve input block
The accuracy rate and subsequent selection efficiency divided, the standard body-sensing that the available body-sensing databases of stimulus to the sense organ module contain
Template number, and standard body-sensing template number is identified as to input the number of block, and the number logarithm based on the input block
Word keyboard is divided.
Optionally, in the present embodiment, stimulus to the sense organ module can determine the motion profile of each standard body-sensing template, and
According to the complexity of each motion profile, the probability that uses of each standard body-sensing template is determined, and should according to using probability to be
The input instruction of the associated input block distribution corresponding number of standard body-sensing template.If this is bigger using probability, corresponding defeated
It is more to enter instruction；And using probability smaller, then corresponding input instruction is fewer, so as to improve input efficiency.
Illustratively, Fig. 2 is the schematic diagram for the numeric keypad that one embodiment of the invention provides.It is shown in Figure 2, sense organ thorn
Module is swashed when exporting the numeric keypad, is divided into five input blocks, and each input block includes that the input of corresponding number refers to
It enables, the corresponding standard body-sensing template of each input block.
In S102, the body-sensing signal acquisition module obtains the body-sensing data of user, and by the body-sensing data correlation
Input block be identified as target block.
In the present embodiment, body-sensing signal acquisition module can be worn on the body of user, and pass through built-in sensing
The body-sensing data of device acquisition user.For example, can be used for obtaining user hand if the body-sensing signal acquisition module is body-sensing gloves
Refer to motion profile, and determines the gesture of user by finger motion locus；If the body-sensing signal acquisition module is body-sensing clothes,
The posture information that can be used for obtaining user generates a plurality of types of input instructions by posture changing.Certainly, body-sensing signal
Acquisition module may include multiple, different body-sensing signal acquisition modules for acquiring the body-sensing signal of user's different parts, body
Sense signal acquisition module can merge the body-sensing signal that all body-sensing signal acquisition modules acquire, and generate body-sensing number
According to, and the target block being associated according to the body-sensing data acquisition after merging；Certainly, body-sensing signal acquisition module can be based on
Itself collected body-sensing data, matches corresponding target block, to realize different body-sensing signal acquisition modules
The purpose of parallel input data improves the efficiency of data input.
In the present embodiment, the system of data input can be associated at least one standard body in advance for each input block
Feel type, body-sensing signal pickup assembly can acquire body of the user after numeric keypad output after detecting numeric keypad output
Feel data, it is preferable that body-sensing signal pickup assembly can be set data acquisition duration, i.e., to export numeric keypad at the time of be
Start time is obtained Client-initiated body-sensing signal in data acquisition duration, and is converted to based on the body-sensing signal acquired
Body-sensing data, it is of course also possible to be directly body-sensing data by body-sensing signal identification without conversion.Then data input is
System is matched according to the body-sensing data with each associated standard body-sensing data of input block, determines target according to matching result
Block.
Optionally, in the present embodiment, body-sensing signal acquisition module can calculate the body-sensing data and each preset mark
Similarity between quasi- body-sensing template chooses similarity and is greater than the standard body-sensing template of preset similar threshold value as target body-sensing
Template, and the associated input block of target body-sensing template is identified as target block；Certainly, body-sensing signal acquisition module may be used also
To choose the highest standard body-sensing template of matching degree as target body-sensing template, and by the defeated of the target body-sensing target association
Enter block as target block.
Illustratively, body-sensing signal acquisition module is body-sensing gloves, and detects user and press which finger, each finger
A corresponding input block, body-sensing signal identify the hand that user presses in this operation according to the body-sensing data collected
Refer to, and the corresponding input block of the finger pressed is identified as target block.It should be noted that user can press multiple hands
Refer to, in this case, body-sensing signal acquisition module can identify that the corresponding input block of multiple fingers is target block.
In S103, the input instruction that the stimulus to the sense organ module includes based on the target block generates instruction
Selection interface.
In the present embodiment, the system of input data can be according to the target block selected by user, will be in numeric keypad
Invalid input instruction is rejected, so as to improve the accuracy of subsequent input instruction.Based on this, stimulus to the sense organ module meeting
The input instruction that each target block includes is extracted, and an instruction selection interface is generated based on all inputs instruction.Optionally,
The input block of non-targeted block is hidden in the numeric keypad that stimulus to the sense organ module can export in S101, is only retained display and is used
The target block of family selection.
Optionally, in the present embodiment, stimulus to the sense organ module can be based on the display priority of each input instruction, to mesh
Each input instruction that mark block includes is rearranged, and instruction selection interface is generated.Wherein, display priority can basis
The history selection number of each input instruction is determined that can also be instructed according to each input being determined using probability, for example,
Application scenarios are text input scene, and include " s " character word quantity more than the word quantity comprising " z " character, then table
Show that the probability of occurrence of " s " can be greater than the probability of occurrence of " z ", in this case, then " s " display priority can be higher than the display of " z "
Priority.
Illustratively, Fig. 3 is a kind of schematic diagram for instruction selection interface that one embodiment of the invention provides.Referring to Fig. 3 institute
Show, which includes multiple input instructions, and each input instruction is arranged according to preset character order, used
The input instruction that family is keyed in needed for being selected based on the instruction selection interface.Optionally, which includes 36
Character, lines up 6 rows 6 column, and stimulus to the sense organ module can be shown in such a way that row/column flashes.User needs to pay close attention to be selected
The target item selected, and write from memory in target item flashing and count the number of its flashing or carry out other related psychological tasks.In this way, target item
Flashing (referred to as vision target stimulation) will be induced in EEG signals specificity ERP ingredient.By to these events correlation
Current potential is identified, determines row and column where target item, and finally determines target instruction target word selected by user.Induced event phase
The amplitude of powered-down position is related with stimulus intervals (ISI), therefore, in order to ensure that signal amplitude, it is proposed that ISI takes 120ms or more, and stimulates
The reinforcement time (time that brightness enhances in Fig. 3) is subject to and visually can clearly perceive, and can use 50~80ms.Same stimulation can be existed
It in primary output repeats to present repeatedly, and gained sample superposed average is improved signal quality.Stimulation number of repetition desirable 1~
It 10 times, is adjusted according to user's actual conditions.
In S104, the electroencephalogramsignal signal acquisition module obtains the brain that the user is fed back based on described instruction selection interface
Electric signal.
In the present embodiment, for stimulus to the sense organ module after outputing instruction selection interface, user can be in required selection
When target instruction target word is shown, the EEG signals of specific waveforms are induced, in this case, the EEG signals of the specific waveforms can pass through
What electroencephalogramsignal signal acquisition module acquired, data processing module can parse EEG signals, so that it is determined that user institute
The target instruction target word of selection.Optionally, stimulus to the sense organ module can export each input to be selected by preset output mode and refer to
It enables, such as is shown by ranks flashing or single flashing lamp mode.
In the present embodiment, institute is not limited to scalp EEG signals (EEG) using EEG signals, it is possible to use electrocorticogram
(ECoG), other reflection brains such as magneticencephalogram (EMG), functional near infrared spectrum (fNIRS), functional magnetic resonance (fMRI) are living
Dynamic signal type.
Optionally, in the present embodiment, the electroencephalogramsignal signal acquisition module is by dry electrode, amplifier, A/D conversion module group
At.Dry electrode uses high impedance Ag/AgCl electrode, can be placed on the electrode cap by highly elastic fiber braiding.The latter makes
Used time is worn on the head of user, to guarantee effective contact of electrode and scalp.Compared with traditional Ag/AgCl wet electrode, electricity is done
Pole is more advantageous to user and wears and use at any time without smearing conductive paste.Preferably, the present embodiment configuration comprising Fz, Cz, Pz,
8 electrodes including the position Oz, O1, O2, T5, T6；These electrodes are according to international 10/20 system standard arrangement；It is accurate to improve
Rate, it is expansible to use more multi-electrode, such as increase C3, C4, P3, P4 electrode near middle line, or in above-mentioned 8 primary electrode weeks
It encloses and increases electrode to improve density.Excellent signal quality is most important to the normal operation of apparatus of the present invention.EEG signals width
Degree is fainter usually from a few μ V to several hundred μ V, therefore to obtain higher signal quality, the present invention needs low noise, Gao Zeng
The beneficial instrument amplifier with high common mode inhibition measures, for example, input reference noiseInput current noiseFront end input impedance >=1012Ω, input current≤30fA, input capacitance≤1.5pF.Amplifier can be used point
It builds from component or is realized using existing integrated chip.The analog signal of acquisition is converted to digital letter by A/D conversion module
Number, common A/D chip in the market can be used and realize.The present invention needs the A/D conversion accuracy of at least 16bit, and sample rate should be
128Hz or more.
Optionally, the electrode configuration figure for including in the electroencephalogramsignal signal acquisition module that Fig. 4 provides for one embodiment of the invention.Ginseng
As shown in Figure 4, simulation ground electrode (GND) is placed on forehead, and reference electrode (A1, A2) is placed on left and right ear-lobe or mastoid location.
Preferably, electroencephalogramsignal signal acquisition module can to direct collected brain wave patterns by amplifying circuit and bandpass filter into
Row Shape correction.Fig. 5 is a kind of single pass signal amplification circuit that one embodiment of the invention provides.As shown in figure 5, by it is preceding,
Dual-stage amplifier and a bandpass filter composition afterwards.Prime instrument amplifier includes the differential amplifier circuit that IC1, IC2 are constituted,
The subtracter constituted with IC3.Signal measurement electrode in Vin map interlinking 4, Vref connect reference electrode (A1 or A2), meet GND to simulation
Electrode.Due to " empty short " effect of operational amplifier, the electric current for flowing through Rg is (Vin-Vref)/Rg；Simultaneously as " void is disconnected " effect
It answers, the electric current for flowing through Rg all reaches the output end of IC1 and IC2 by negative-feedback circuit, to obtain (V1-V2)=(Vin-
Vref)(1+2R/Rg).Similarly, it is known that the output voltage Vp=(V2-V1) of IC3 composition subtracter.Therefore, pre-amplifier
Overall magnification is A=(1+2R/Rg).To avoid operational amplifier current from being saturated, the amplification of preamplifying circuit of the present invention
Multiple takes 5~20, it is proposed that takes 10.It is fertile hereby or Chebyshev filter, free transmission range suggestion are that Bart can be used in bandpass filter
0.05Hz~40Hz.Post-amplifier (IC4) amplification factor takes 50~200, it is proposed that takes 100.The present invention is needed to multiple EEG electricity
Pole Channel Synchronous acquires signal, therefore according to number of electrodes (8 EEG electrodes as proposed) used, EEG acquisition module should be wrapped
Multiple single-channel amplifiers containing respective number.Amplified output signal Vo is converted to digital signal by A/D conversion module.
The analog signal of acquisition is converted to digital signal by A/D conversion module, and common A/D chip in the market can be used and realize.This hair
The bright A/D conversion accuracy for needing at least 16bit, sample rate should be in 128Hz or more.
In S105, the data processing module parses the EEG signals, determines target instruction target word selected by the user.
In the present embodiment, data processing module can solve EEG signals by preset brain electricity analytical algorithm
Analysis, determines target instruction target word associated by the EEG signals.Specifically, data processing model can recorde each input instruction
Standard brain wave patterns, the EEG signals that data processing module collects this are matched with each standard brain wave patterns,
The highest input instruction of matching degree is selected as target instruction target word；Or data processing module records two type of target/non-target
Other brain electrical feature waveform, training two-value linear classification model；This collected EEG signals is sent by data processing module
Trained two-value linear classification model is identified, determines target instruction target word.
In the prior art, the additional wearing biggish eye-tracking equipment of volume is needed in conjunction with the method for eye-tracking technology,
It can cause the discomfort of user；The method that brain-computer interface is combined with intelligent control is right by the auxiliary of intelligent control technology
The intermediate process of brain-computer interface manipulation has carried out automatic processing, however does not improve the information throughput of brain-computer interface itself,
Therefore suitable for application in the scene for fully relying on brain-computer interface and being manipulated, such as complicated panel manipulated, text output；For
For motion sensing control technology, the complicated panel that more buttons are carried out under virtual/augmented reality environment manipulates (such as simulated flight device control
Panel processed), text output when, it is higher to the required precision of body-sensing action recognition, while when lacking touch feedback, carry out
Good interactive experience is relatively inaccessible to when these operations.The embodiment of the present invention provides index using body-sensing information for brain-computer interface and refers to
It leads, does not need high-precision somatosensory recognition, in virtual/augmented reality environment, in combination with the somatosensory recognition of equipment itself offer
Function wears inertial sensor (such as Intelligent glove), improves the interactive experience under the scenes such as complicated panel manipulation, text output；
In clinical application, the situation of limb motion is unable to accurately control for patient's (such as Parkinson's disease patients), by wearing inertia
Sensor (such as Intelligent glove) uses other somatosensory recognition methods, it can be made using limited locomitivity, improve brain machine
The control performance of interface improves the quality of life of patient.
Above as can be seen that a kind of method of data input provided in an embodiment of the present invention is by the way that numeric keypad to be divided into
Multiple and different input blocks, and the numeric keypad after the division is exported by stimulus to the sense organ module, user can lead to
It crosses body-sensing signal acquisition module and chooses corresponding target block from multiple input blocks, then stimulus to the sense organ module can basis
The input instruction that target block includes generates instruction selection interface, is guaranteeing that input instruction is multifarious while can be improved input
The accuracy rate of operation is then obtained the EEG signals of user by electroencephalogramsignal signal acquisition module, and passes through data processing module pair
The brain electric information is parsed, and identification obtains target instruction target word selected by user, is realized through EEG signals to external equipment
It directly controls.Compared with existing brain-machine interaction technology, the present invention wears ophthalmogyric device without user, to reduce entire number
It, can be all kinds of just by gloves, wrist etc. since the equipment volume of body-sensing data acquisition equipment is smaller according to the constitution of input system
The wearable device taken obtains the body-sensing data of user, to improve the operating experience of user, at the same time, data input is
Numeric keypad can be divided into multiple input blocks by system, ensure that input instruction is multifarious simultaneously, can passed through body-sensing
The corresponding target block of data activation, a large amount of invalid instructions of screening, so that the accuracy of following target instructions selection is improved,
It ensure that the handling capacity of brain machine signal.
Fig. 6 shows the specific implementation flow of the method S102 of data input of second embodiment of the invention offer a kind of
Figure.The executing subject of the embodiment of the present invention is response server, and referring to Fig. 6, relative to embodiment described in Fig. 1, the present embodiment is mentioned
A kind of method S102 of data input supplied includes: S1021~S1022, and specific details are as follows:
Further, the body-sensing signal acquisition module obtains the body-sensing data of user, and by the body-sensing data correlation
Input block be identified as target block, comprising:
In S1021, the body-sensing signal acquisition module calculates the body-sensing data and the associated mark of each input block
Matching degree between quasi- data.
In the present embodiment, input block can be set before using the system formally inputted using data in user
Number is divided, and is the corresponding gesture of each input block arrangement or trigger action.Body-sensing signal acquisition module can obtain user
Gesture and/or the corresponding body-sensing data of trigger action in the provisioning process, and the body-sensing data acquired in the provisioning process
It is identified as normal data, and is stored in database, and establishes the incidence relation between input block and each normal data.
In the present embodiment, body-sensing signal acquisition module can extract the associated normal data of each input block, and respectively
Calculate the matching degree between each normal data and this body-sensing data collected.Optionally, if the body-sensing data are fortune
Dynamic rail mark, body-sensing signal acquisition module can generate the motion profile and normal data pair of body-sensing data in preset reference axis
The standard trajectory answered, and the registration between two tracks is calculated, which is identified as matching degree between the two.
In S1022, the matching degree is greater than the mark of preset matching threshold by the body-sensing signal acquisition model
The corresponding input block of quasi- data is identified as the target block.
In the present embodiment, since user can choose multiple input blocks, in this case, body-sensing in once-through operation
A matching threshold has can be set in signal acquisition module, and body-sensing signal acquisition model can choose matching degree greater than the matching threshold
The corresponding input block of normal data is as target block, since the body-sensing data and the normal data matching degree are higher, then table
Show and trigger the corresponding movement of the normal data, therefore the associated input block of the normal data can be identified as to user's selection
Target block.
In embodiments of the present invention, by calculating separately between the body-sensing data and each normal data that this is collected
Matching degree, so that it is determined that user selection target block, so as to improve target block identification accuracy, to mention
The high usage experience of the data input based on brain-machine interaction.
Fig. 7 shows the specific implementation flow of the method S104 of data input of third embodiment of the invention offer a kind of
Figure.Referring to Fig. 7, relative to Fig. 1 the embodiment described, a kind of method S104 of data input provided in this embodiment includes:
S1041~S1042, specific details are as follows:
Further, the electroencephalogramsignal signal acquisition module includes multiple brain wave acquisition interfaces, and each brain wave acquisition connects
The corresponding input block of mouth；It is anti-based on described instruction selection interface that the electroencephalogramsignal signal acquisition module obtains the user
The EEG signals of feedback, comprising:
In S1041, the electroencephalogramsignal signal acquisition module activates the corresponding brain wave acquisition interface of the target block.
In the present embodiment, which configures multiple brain wave acquisition interfaces, it is preferable that the brain wave acquisition
The input number of block that the number of interface can include with numeric keypad is identical, and different brain wave acquisition interfaces is closed for receiving
In the EEG signals of all inputs instruction in the input block, thus realize the parallel EEG signals for receiving user and generating, from
And the load of each brain wave acquisition interface can be reduced, the quality of signal acquisition is improved, target instruction target word is further improved
Identify accuracy.
In the present embodiment, the target block that electroencephalogramsignal signal acquisition module can be chosen according to user, determines the target area
The associated brain wave acquisition interface of block, each target block at least correspond to a brain wave acquisition interface, can also correspond to certainly multiple
Brain wave acquisition interface.Optionally, a brain wave acquisition interface can correspond to multiple input blocks.Electroencephalogramsignal signal acquisition module can be with
According to the corresponding relationship between input block and brain wave acquisition interface, the associated brain wave acquisition interface of target block, and base are determined
In with the associated brain wave acquisition interface of target block.
In S1042, the electroencephalogramsignal signal acquisition module obtains user's base by the brain wave acquisition interface simultaneously
In N number of EEG signals of described instruction selection interface feedback；The N is the number of the target block.
In the present embodiment, electroencephalogramsignal signal acquisition module can obtain corresponding region by multiple brain wave acquisition interface concurrents
The EEG signals received, and multiple EEG signals received are sent to data processing module, wherein the brain acquired
The quantity of electric signal is identical as the quantity of brain wave acquisition interface of activation, and of the quantity of brain wave acquisition interface and target block
Number is identical, therefore the number of the EEG signals fed back is identical as the number of target block.
In embodiments of the present invention, by configuring multiple brain wave acquisition interfaces in electroencephalogramsignal signal acquisition module, so as to
The purpose for enough realizing parallel acquisition EEG signals, reduces the load of each brain wave acquisition interface, improves target instruction target word identification
Accuracy.
Fig. 8 shows the specific implementation flow of the method S104 of data input of fourth embodiment of the invention offer a kind of
Figure.Referring to Fig. 8, relative to embodiment described in Fig. 1, S104 includes: in a kind of method of data input provided in this embodiment
S1043~S1046, specific details are as follows:
Further, the electroencephalogramsignal signal acquisition module obtains the brain that the user is fed back based on described instruction selection interface
Electric signal, comprising:
In S1043, the electroencephalogramsignal signal acquisition module is filtered original signal by preset bandpass filter
Processing, obtains filtering signal.
In the present embodiment, electroencephalogramsignal signal acquisition module can pre-process the original signal directly acquired,
So as to improve the accuracy of subsequent operation.Based on this, it can be configured with bandpass filter in electroencephalogramsignal signal acquisition module, lead to
Bandpass filter is crossed to the original signal directly collected, i.e., without passing through any processing, is fed back from brain wave acquisition interface
Original EEG signals, are filtered, and so as to be filtered to the noise for including in EEG signals, realize to original letter
Number carry out the purpose of shaping, and the filtering signal after being filtered.Specifically, which is 0.5
~6Hz, passband ripple are less than 3dB, and stopband attenuation is greater than 40dB, the less iir filter of parameter can be used.
In S1044, the electroencephalogramsignal signal acquisition module is instructed according to input each in described instruction selection interface and is corresponded to
Energizing cycle, the filtering signal is divided into multiple subsignal sections.
In the present embodiment, stimulus to the sense organ module, can be according to preset Energizing cycle in output order selection interface
By flashing or the modes such as highlighted show specific input instruction, i.e., each input, which instructs, is corresponding with an Energizing cycle, at this
The content of input instruction can be highlighted in Energizing cycle, user can judge whether in the Energizing cycle of each input instruction
Choose input instruction.If judgement selects input instruction, the brain wave of the first kind can be generated；If judgement does not select the input
Instruction, then can be generated the brain wave of Second Type, can then be divided to brain wave according to Energizing cycle, judge to be segmented
Subsignal section corresponding to brain wave type, whether identification user selected the corresponding input of the Energizing cycle to instruct.It is based on
This, eeg signal acquisition device can divide filtering signal according to preset Energizing cycle, and obtain multiple subsignals
Section, the input instruction in each subsignal section corresponding instruction selection interface.
In S1045, the electroencephalogramsignal signal acquisition module determines sampling frequency according to the frequency acquisition of the original signal
Rate, and it is down-sampled to each subsignal section progress respectively based on the sampling frequency, it is right to generate each subsignal section
The sampled sequence answered.
In the present embodiment, in order to reduce the data volumes of EEG signals, electroencephalogramsignal signal acquisition module can be to each segmentation
Subsignal section progress afterwards is down-sampled, so that continuous subsignal section is converted to a sampled sequence, to balance data
Integrity degree while, can also improve the data volume of delivery efficiency and EEG signals.Based on this, electroencephalogramsignal signal acquisition module meeting
The frequency acquisition of original signal is obtained, and is based on preset down-sampled ratio, determines sampling frequency.For example, original signal
Frequency acquisition is 120Hz, and down-sampled ratio is 1/6th, then corresponding down-sampled frequency is then 20Hz.
In the present embodiment, electroencephalogramsignal signal acquisition module respectively carries out each subsignal section according to sampling frequency down-sampled
Operation, to obtain the corresponding sampled sequence of each subsignal section.Preferably, if the Energizing cycle of each subsignal section is consistent,
The element number that the sampled sequence then generated includes is identical.
In S1046, the electroencephalogramsignal signal acquisition module is by the sampled sequence according to the corresponding subsignal section
Segment number successively merges, and generates the EEG signals.
In the present embodiment, electroencephalogramsignal signal acquisition module can go out occurrence according to each subsignal section in original signal
Sequence then successively merges all sampled sequences according to the segment number as the segment number of the subsignal section, generates brain electricity
Signal, and EEG signals are sent to data processing module and are parsed.
In embodiments of the present invention, primary signal is pre-processed by filtering, segmentation, down-sampled etc., so as to
The accuracy of output EEG signals is improved, so as to improve the accuracy of following target instructions identification.
Fig. 9 shows a kind of specific implementation flow chart of the method for data input of fifth embodiment of the invention offer.Ginseng
See Fig. 9, relative to embodiment described in Fig. 8, a kind of method of data input provided in this embodiment is in the data processing module
The EEG signals are parsed, before determining target instruction target word selected by the user, further includes: S901~S903, it is specific to be described in detail such as
Under:
Further, the EEG signals are parsed in the data processing module, determines that target selected by the user refers to
Before order, further includes:
In S901, the data processing module generates multiple training instructions, and controls the stimulus to the sense organ module and generate
Training selection interface comprising the training instruction.
In the present embodiment, since the brain wave patterns of each user have differences, in order to improve the identification of EEG signals
Accuracy, data input system before the use, need to carry out model training, so as to the currently used user of determination
EEG signals characteristic.Optionally, the system of data input on startup, can obtain the user identifier of active user, judge number
According in library whether include the associated two-value linear classification model of the user identifier, if comprising, then it represents that the user carry out
Training operation；Conversely, executing the operation of S901 to S903 if not including.
In the present embodiment, data processing model will create multiple training instructions, and prompt generating training selection interface
When, user needs to select multiple training instructions of above-mentioned creation, and acquires brain telecommunications of the user when confirmation selects training quality
Number, the brain wave patterns of " selection " and the brain wave patterns of expression " not selecting " are indicated so as to tell the user.
In the present embodiment, which includes the training instruction of above-mentioned creation, also comprising selecting without user
The illegal command selected.Optionally, in order to improve the accuracy of two-value linear classification model, the system of data input can repeat more
The operation of secondary S901 to S903, the training signal obtained based on multi collect are trained study to two-value linear classification model.
In S902, the electroencephalogramsignal signal acquisition module obtains instruction of the user based on the trained selection interface feedback
Practice signal.
In the present embodiment, the available user of brain wave acquisition module is fed back based on the training selection interface of above-mentioned output
Training signal.The training signal is specially that user selects the EEG signals induced when training instruction.Optionally, stimulus to the sense organ module
Each instruction in trained selection interface can be successively lighted by preset Energizing cycle, thus comprising related in training signal
It in the subsignal section of selection training instruction, and include about the subsignal section for not selecting illegal command.Based on this, at data
Reason module can be identified for indicating brain wave patterns characteristic when selection, and brain wave patterns characteristic of expression when not selecting.
In S903, the data processing module is believed by training described in preset gradually linear discriminant analysis arithmetic analysis
Number, and the two-value linear classification model about the user is established based on analysis result；The two-value linear classification model is used for
Parse the EEG signals of the user.
In the present embodiment, data processing model can construct the two-value of the user according to the training signal collected
Linear classification model, to can be determined by the two-value linear classification model during parsing of subsequent EEG signals
The selected aimed quality of user.
In the present embodiment, the mode for specifically generating two-value linear classification model can be with are as follows: by the corresponding spy of training instruction
The label of sign vector is set as 1, sets -1 for the label of the corresponding feature vector of illegal command, based on the training letter acquired
Number and preset Energizing cycle, training signal is divided into multiple subsignal sections, based on corresponding to each subsignal section swash
Periodicity is encouraged, determines the associated instruction type of subsignal section, so as to divide about the corresponding sub- letter of training instruction
Number section, and about the corresponding subsignal section of illegal command, obtained by gradually Fisher face and above-mentioned division
Subsignal section is trained two-value linear classification model, wherein the output category value of the corresponding subsignal section of training instruction is
1, and the output category value of the corresponding subsignal section of illegal command is -1.
In embodiments of the present invention, pre- by being carried out before the system inputted using data to two-value linear classification model
Training, can acquire brain wave patterns characteristic related to user, and improve the identification accuracy of subsequent target instruction target word.
Figure 10 shows the specific implementation flow of the method S105 of data input of sixth embodiment of the invention offer a kind of
Figure.Referring to Figure 10, relative to embodiment described in Fig. 9, a kind of method S105 of data input provided in this embodiment includes:
S1051~S1052, specific details are as follows:
Further, the data processing module parses the EEG signals, determines target instruction target word selected by the user,
Include:
In S1051, the data processing module calculates separately the brain telecommunications by the two-value linear classification model
The classification value of each sampled sequence in number.
In the present embodiment, EEG signals can be divided into multiple by data processing model after getting EEG signals
Sampled sequence, and each sampled sequence is imported into the two-value linear classification model, it calculates associated about sampled sequence
Classification value.Due to the corresponding input instruction of each sampled sequence, data processing model can be according to the classification of the sampled sequence
Value judges whether user has selected the associated input instruction of the sampled sequence.
In S1052, the data processing module determines the target instruction target word based on all classification values.
In the present embodiment, since the corresponding input of each sampled sequence instructs, data processing module determine it is each
After the classification value of sampled sequence, sampled sequence that classification value is the first place value can be chosen as object sample sequence, such as the
One place value is 1, and is target instruction target word by the corresponding input instruction identification of the sampled sequence.Optionally, data processing module may be used also
To instruct corresponding classification value, in conjunction with code used method, finally determine target item according to each input, if stimulus to the sense organ module uses
Be ranks flashing instruction selection interface, then the associated all score values difference of row/column flashing where being instructed with each input
It adds up, obtains the cumulative score value of each input instruction, finally output possesses the character of maximum score value.
In embodiments of the present invention, the classification of each sampled sequence is calculated by the two-value linear classification model after training
Value improves the selection efficiency of target instruction target word to identify the target instruction target word of user's selection.
Figure 11 shows a kind of specific implementation flow chart of the method for data input of seventh embodiment of the invention offer.Ginseng
See Figure 11, relative to any embodiment of Fig. 1-10, a kind of method of data input provided in this embodiment is in the body-sensing
Signal acquisition module obtain user body-sensing data, and by the input block of the body-sensing data correlation be identified as target block it
Before, further includes: S1101~S1102, specific details are as follows:
In S1101, the body-sensing acquisition position that the body-sensing signal acquisition module obtains the user is in non-athletic shape
The body-sensing a reference value of state.
In the present embodiment, body-sensing signal acquisition module can believe the body-sensing before the body-sensing data for obtaining user
Number acquisition module is calibrated, i.e. acquisition user is under non-athletic state, i.e., body-sensing a reference value under relaxation state, so as to
It is recognized accurately whether user carries out body-sensing triggering, and records the body-sensing data in body-sensing triggering period.
Optionally, in the present embodiment, which can be equipped with 10 inertial sensor nodes, can
The 3-axis acceleration data of each finger, palm, arm are obtained in real time.The acquisition methods of body-sensing data specifically: 1. sensor is logical
Cross USB cable connection data processing module；2. using and connecting Intelligent glove with body-sensing signal acquisition module software kit, body is acquired
Feel data；3. electroencephalogramsignal signal acquisition module obtains Intelligent glove body-sensing data by the TCP port of preset opening in real time.
In S1102, the body-sensing signal acquisition module constructs the body about the user based on the body-sensing a reference value
Thoughts and feelings hair decision function；The body-sensing decision function is for generating the body-sensing data；The body-sensing decision function specifically:
Wherein, Action is the body-sensing decision function；For the body-sensing a reference value；Th0For preset firing level
Value；AcczFor the body-sensing data to be obtained；Sign is the body-sensing transfer function of the body-sensing signal acquisition module.
In the present embodiment, the body-sensing a reference value that body-sensing signal acquisition module can will acquire imported into preset body-sensing
It triggers in decision model, triggers decision function about the body-sensing of user to generate, if the output of body-sensing triggering decision function
Value is greater than preset activation threshold value, such as 0, then identifies that user is passing through body-sensing signal acquisition module input body-sensing data, instead
It, then do not acquire the body-sensing data of user.Wherein, Th0For detection threshold value, it can use 3~10, wherein the smaller detection sensitivity of value
It is higher.
In embodiments of the present invention, by obtaining the body-sensing a reference value of user before using body-sensing signal acquisition model,
So as to improve the accuracy of body-sensing data acquisition.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
Figure 12 shows a kind of structural block diagram of the system of data input of one embodiment of the invention offer, data input
System include response server and at least one service request terminal, response server and service request terminal are for holding
Each step in the corresponding embodiment of row Fig. 1.Referring specifically to the associated description in embodiment corresponding to Fig. 1 and Fig. 1.In order to
Convenient for explanation, only the parts related to this embodiment are shown.
Referring to Figure 12, the system of data input includes electroencephalogramsignal signal acquisition module 121, body-sensing signal acquisition module 122, sense
Official's stimulating module 123 and data processing module 124；
The stimulus to the sense organ module 123, for generating and exporting comprising multiple inputs according to preset block partitioning algorithm
The numeric keypad of block；Each input block includes at least one input instruction；
The body-sensing signal acquisition module 122, for obtaining the body-sensing data of user, and by the body-sensing data correlation
Input block is identified as target block；
The stimulus to the sense organ module 123, the input instruction for including based on the target block, generates instruction choosing
Select interface；
The electroencephalogramsignal signal acquisition module 121, the brain fed back for obtaining the user based on described instruction selection interface
Electric signal；
The data processing module 124 determines target instruction target word selected by the user for parsing the EEG signals.
Optionally, the body-sensing signal acquisition module 122 is also used to:
The body-sensing signal acquisition module 122, for calculating the body-sensing data and the associated standard of each input block
Matching degree between data；
The body-sensing signal acquisition module 122, for by the corresponding input of the highest normal data of the matching degree
Block is identified as the target block.
Optionally, the electroencephalogramsignal signal acquisition module 121 includes multiple brain wave acquisition interfaces, and each brain wave acquisition connects
The corresponding input block of mouth；
The electroencephalogramsignal signal acquisition module 121 is also used to:
The electroencephalogramsignal signal acquisition module 121, for activating the corresponding brain wave acquisition interface of the target block；
The electroencephalogramsignal signal acquisition module 121 is based on for obtaining the user simultaneously by the brain wave acquisition interface
N number of EEG signals of described instruction selection interface feedback；The N is the number of the target block.
Optionally, the electroencephalogramsignal signal acquisition module 121 is also used to:
The electroencephalogramsignal signal acquisition module 121, for being filtered place to original signal by preset bandpass filter
Reason, obtains filtering signal；
The electroencephalogramsignal signal acquisition module 121, it is corresponding for being instructed according to input each in described instruction selection interface
The filtering signal is divided into multiple subsignal sections by Energizing cycle；
The electroencephalogramsignal signal acquisition module 121 determines sampling frequency for the frequency acquisition according to the original signal,
And it is down-sampled to each subsignal section progress respectively based on the sampling frequency, it is corresponding to generate each subsignal section
Sampled sequence；
The electroencephalogramsignal signal acquisition module 121, for the section by the sampled sequence according to the corresponding subsignal section
Number successively merges, and generates the EEG signals.
Optionally, the data processing module 124 and the electroencephalogramsignal signal acquisition module 121 are also used to:
The data processing module 124 for generating multiple training instructions, and controls the stimulus to the sense organ module and generates packet
Training selection interface containing the training instruction；
The electroencephalogramsignal signal acquisition module 121, for obtaining instruction of the user based on the trained selection interface feedback
Practice signal；
The data processing module 124, for being believed by training described in preset gradually linear discriminant analysis arithmetic analysis
Number, and the two-value linear classification model about the user is established based on analysis result；The two-value linear classification model is used for
Parse the EEG signals of the user.
Optionally, the data processing module 124 is also used to:
The data processing module 124, for calculating separately the EEG signals by the two-value linear classification model
In each sampled sequence classification value；
The data processing module 124, for determining the target instruction target word based on all classification values.
Optionally, the body-sensing signal acquisition module 122 is also used to:
The body-sensing signal acquisition module 122, the body-sensing for obtaining the user acquire position and are in non-athletic state
Body-sensing a reference value；
The body-sensing signal acquisition module 122, for constructing the body-sensing about the user based on the body-sensing a reference value
Trigger decision function；The body-sensing decision function is for generating the body-sensing data；The body-sensing decision function specifically:
Wherein, Action is the body-sensing decision function；For the body-sensing a reference value；Th0For preset firing level
Value；AcczFor the body-sensing data to be obtained；Sign is the body-sensing transfer function of the body-sensing signal acquisition module.
Therefore, in the system of data input provided in an embodiment of the present invention, ophthalmogyric device is worn without user, to reduce
The constitution of entire data entry system since the equipment volume of body-sensing data acquisition equipment is smaller can pass through gloves, wrist
The body-sensing data of user are obtained etc. all kinds of portable wearable devices, so that the operating experience of user is improved, at the same time, data
Numeric keypad can be divided into multiple input blocks by the system of input, ensure that input instruction is multifarious simultaneously, can be with
By the corresponding target block of body-sensing data activation, a large amount of invalid instructions of screening, to improve following target instructions selection
Accuracy, ensure that the handling capacity of brain machine signal.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of method of data input, is applied to data entry system, which is characterized in that the data entry system at least wraps
It includes: electroencephalogramsignal signal acquisition module, body-sensing signal acquisition module, stimulus to the sense organ module and data processing module；
The method of data input includes:
The stimulus to the sense organ module generates according to preset block partitioning algorithm and exports the number key comprising multiple input blocks
Disk；Each input block includes at least one input instruction；
The body-sensing signal acquisition module obtains the body-sensing data of user, and the input block of the body-sensing data correlation is identified
For target block；
The input instruction that the stimulus to the sense organ module includes based on the target block, generates instruction selection interface；
The electroencephalogramsignal signal acquisition module obtains the EEG signals that the user is fed back based on described instruction selection interface；
The data processing module parses the EEG signals, determines target instruction target word selected by the user.
2. the method according to claim 1, wherein the body-sensing signal acquisition module obtains the body-sensing number of user
According to, and the input block of the body-sensing data correlation is identified as target block, comprising:
The body-sensing signal acquisition module calculates between the body-sensing data and each input associated normal data of block
With degree；
The normal data that the matching degree is greater than preset matching threshold by the body-sensing signal acquisition model is corresponding defeated
Enter block and is identified as the target block.
3. the method according to claim 1, wherein the electroencephalogramsignal signal acquisition module includes multiple brain wave acquisitions
Interface, each corresponding input block of the brain wave acquisition interface；
The electroencephalogramsignal signal acquisition module obtains the EEG signals that the user is fed back based on described instruction selection interface, comprising:
The electroencephalogramsignal signal acquisition module activates the corresponding brain wave acquisition interface of the target block；
The electroencephalogramsignal signal acquisition module is obtained the user by the brain wave acquisition interface simultaneously and is selected based on described instruction
N number of EEG signals of interface feedback；The N is the number of the target block.
4. being based on the method according to claim 1, wherein the electroencephalogramsignal signal acquisition module obtains the user
The EEG signals of described instruction selection interface feedback, comprising:
The electroencephalogramsignal signal acquisition module is filtered original signal by preset bandpass filter, obtains filtering letter
Number；
The electroencephalogramsignal signal acquisition module instructs corresponding Energizing cycle according to input each in described instruction selection interface, by institute
It states filtering signal and is divided into multiple subsignal sections；
The electroencephalogramsignal signal acquisition module determines sampling frequency, and be based on the pumping according to the frequency acquisition of the original signal
Sample frequency is down-sampled to each subsignal section progress respectively, generates the corresponding sampled sequence of each subsignal section；
The electroencephalogramsignal signal acquisition module successively merges the sampled sequence according to the segment number of the corresponding subsignal section,
Generate the EEG signals.
5. according to the method described in claim 4, it is characterized in that, parse the EEG signals in the data processing module,
Before determining target instruction target word selected by the user, further includes:
The data processing module generates multiple training instructions, and controls the stimulus to the sense organ module and generate and refer to comprising the training
The training selection interface of order；
The electroencephalogramsignal signal acquisition module obtains training signal of the user based on the trained selection interface feedback；
The data processing module is based on parsing by training signal described in preset gradually linear discriminant analysis arithmetic analysis
As a result the two-value linear classification model about the user is established；The two-value linear classification model is for parsing the user's
The EEG signals.
6. according to the method described in claim 5, it is characterized in that, the data processing module parses the EEG signals, really
Target instruction target word selected by the fixed user, comprising:
The data processing module is calculated separately each described in the EEG signals by the two-value linear classification model
The classification value of sampled sequence；
The data processing module determines the target instruction target word based on all classification values.
7. method according to claim 1-6, which is characterized in that obtain and use in the body-sensing signal acquisition module
The body-sensing data at family, and the input block of the body-sensing data correlation is identified as before target block, further includes:
The body-sensing acquisition position that the body-sensing signal acquisition module obtains the user is in the body-sensing a reference value of non-athletic state；
The body-sensing signal acquisition module is constructed based on the body-sensing a reference value and triggers decision function about the body-sensing of the user；
The body-sensing decision function is for generating the body-sensing data；The body-sensing decision function specifically:
Wherein, Action is the body-sensing decision function；For the body-sensing a reference value；Th0For preset activation threshold value；
AcczFor the body-sensing data to be obtained；Sign is the body-sensing transfer function of the body-sensing signal acquisition module.
8. a kind of system of data input characterized by comprising electroencephalogramsignal signal acquisition module, body-sensing signal acquisition module, sense
Official's stimulating module and data processing module；
The stimulus to the sense organ module, for according to preset block partitioning algorithm, generating and exporting comprising multiple input blocks
Numeric keypad；Each input block includes at least one input instruction；
The body-sensing signal acquisition module, for obtaining the body-sensing data of user, and by the input area of the body-sensing data correlation
Block is identified as target block；
The stimulus to the sense organ module, the input instruction for including based on the target block, generates instruction selection interface；
The electroencephalogramsignal signal acquisition module, the EEG signals fed back for obtaining the user based on described instruction selection interface；
The data processing module determines target instruction target word selected by the user for parsing the EEG signals.
9. system according to claim 8, which is characterized in that the body-sensing signal acquisition module is also used to:
The body-sensing signal acquisition module, for calculating between the body-sensing data and the associated normal data of each input block
Matching degree；
The body-sensing signal acquisition module, for by the corresponding input block identification of the highest normal data of the matching degree
For the target block.
10. system according to claim 8, which is characterized in that the electroencephalogramsignal signal acquisition module is also used to:
The electroencephalogramsignal signal acquisition module is obtained for being filtered by preset bandpass filter to original signal
Filtering signal；
The electroencephalogramsignal signal acquisition module, for instructing corresponding excitation week according to input each in described instruction selection interface
The filtering signal is divided into multiple subsignal sections by the phase；
The electroencephalogramsignal signal acquisition module determines sampling frequency for the frequency acquisition according to the original signal, and is based on institute
It is down-sampled to each subsignal section progress respectively to state sampling frequency, generates the corresponding sampling sequence of each subsignal section
Column；
The electroencephalogramsignal signal acquisition module, for by the sampled sequence according to the segment number of the corresponding subsignal section successively
Merge, generates the EEG signals.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910304787.3A CN110174938B (en) | 2019-04-16 | 2019-04-16 | Data input method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910304787.3A CN110174938B (en) | 2019-04-16 | 2019-04-16 | Data input method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110174938A true CN110174938A (en) | 2019-08-27 |
| CN110174938B CN110174938B (en) | 2022-03-18 |
Family
ID=67690038
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910304787.3A Active CN110174938B (en) | 2019-04-16 | 2019-04-16 | Data input method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110174938B (en) |
Citations (7)
| 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 |
| BR102012027982A2 (en) * | 2012-10-31 | 2014-10-21 | Vieira De Oliveira Junior Marcio | SYSTEM APPLIED ON BRAIN SIGNAL COMMAND DEVICE FOR ACTION FUNCTION OF ADAPTED MOUSE FOR PEOPLE WITH PHYSICAL DISABILITY |
| US20170185150A1 (en) * | 2015-07-20 | 2017-06-29 | Beijing University Of Technology | Wireless input system based on steady-state visual-evoked potentials |
| CN107239137A (en) * | 2017-04-25 | 2017-10-10 | 中国农业大学 | A kind of character input method and device based on dummy keyboard |
| CN108181995A (en) * | 2018-01-31 | 2018-06-19 | 京东方科技集团股份有限公司 | interactive system, method and device |
| CN109471530A (en) * | 2018-10-22 | 2019-03-15 | 吉林大学 | Brain control input method based on Steady State Visual Evoked Potential and Mental imagery |
| CN109582131A (en) * | 2018-10-29 | 2019-04-05 | 中国航天员科研训练中心 | The asynchronous mixing brain-machine interface method of one kind and system |
-
2019
- 2019-04-16 CN CN201910304787.3A patent/CN110174938B/en active Active
Patent Citations (7)
| 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 |
| BR102012027982A2 (en) * | 2012-10-31 | 2014-10-21 | Vieira De Oliveira Junior Marcio | SYSTEM APPLIED ON BRAIN SIGNAL COMMAND DEVICE FOR ACTION FUNCTION OF ADAPTED MOUSE FOR PEOPLE WITH PHYSICAL DISABILITY |
| US20170185150A1 (en) * | 2015-07-20 | 2017-06-29 | Beijing University Of Technology | Wireless input system based on steady-state visual-evoked potentials |
| CN107239137A (en) * | 2017-04-25 | 2017-10-10 | 中国农业大学 | A kind of character input method and device based on dummy keyboard |
| CN108181995A (en) * | 2018-01-31 | 2018-06-19 | 京东方科技集团股份有限公司 | interactive system, method and device |
| CN109471530A (en) * | 2018-10-22 | 2019-03-15 | 吉林大学 | Brain control input method based on Steady State Visual Evoked Potential and Mental imagery |
| CN109582131A (en) * | 2018-10-29 | 2019-04-05 | 中国航天员科研训练中心 | The asynchronous mixing brain-machine interface method of one kind and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110174938B (en) | 2022-03-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108829245B (en) | A kind of virtual sand table intersection control routine based on multi-modal brain-machine interaction technology | |
| Ebrahimi et al. | Brain-computer interface in multimedia communication | |
| Barbosa et al. | Activation of a mobile robot through a brain computer interface | |
| CN101201696B (en) | Chinese input BCI system based on P300 brain electric potential | |
| CN110123313B (en) | Self-training brain-computer interface system and related training method | |
| CN101159086B (en) | Calling device based on brain electric information detection | |
| CN109471530A (en) | Brain control input method based on Steady State Visual Evoked Potential and Mental imagery | |
| CN108958486A (en) | Mood brain machine interface system based on recognition of face brain electricity N170 ingredient | |
| CN108334766B (en) | Electronic device, unlocking method and Related product | |
| CN109326341A (en) | A kind of rehabilitation motion guiding method and apparatus | |
| CN108415560A (en) | Electronic device, method of controlling operation thereof and Related product | |
| CN104267807A (en) | Hand action mechanomyography based man-machine interaction method and interaction system | |
| CN110141258A (en) | A kind of emotional state detection method, equipment and terminal | |
| CN106774851A (en) | Sense of touch finger motion rehabilitation system and method based on brain-computer interface | |
| CN113208593A (en) | Multi-modal physiological signal emotion classification method based on correlation dynamic fusion | |
| CN110472595B (en) | Electroencephalogram recognition model construction method and device and recognition method and device | |
| Zhang et al. | Spatial-temporal neural network for P300 detection | |
| CN110688013A (en) | English keyboard spelling system and method based on SSVEP | |
| CN108491792A (en) | Office scene human-computer interaction Activity recognition method based on electro-ocular signal | |
| CN110174938A (en) | A kind of method and system of data input | |
| CN108491074A (en) | Electronic device, exercising support method and Related product | |
| Bi et al. | A SSVEP brain-computer interface with the hybrid stimuli of SSVEP and P300 | |
| CN112783314B (en) | Brain-computer interface stimulation paradigm generating and detecting method, system, medium and terminal based on SSVEP | |
| Bandara et al. | Differentiation of signals generated by eye blinks and mouth clenching in a portable brain computer interface system | |
| CN113552941A (en) | Multi-sensory-mode BCI-VR (binary-coded decimal-alternating Current-virtual Voltage regulator) control method and system and VR equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |