CN115590535B - Time window adjusting method, device and equipment for electroencephalogram signal identification and storage medium - Google Patents

Time window adjusting method, device and equipment for electroencephalogram signal identification and storage medium Download PDF

Info

Publication number
CN115590535B
CN115590535B CN202211462180.6A CN202211462180A CN115590535B CN 115590535 B CN115590535 B CN 115590535B CN 202211462180 A CN202211462180 A CN 202211462180A CN 115590535 B CN115590535 B CN 115590535B
Authority
CN
China
Prior art keywords
time window
identification
time
adjusting
ssvep
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202211462180.6A
Other languages
Chinese (zh)
Other versions
CN115590535A (en
Inventor
牛兰
宾剑雄
康晓洋
张立华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CN202211462180.6A priority Critical patent/CN115590535B/en
Publication of CN115590535A publication Critical patent/CN115590535A/en
Application granted granted Critical
Publication of CN115590535B publication Critical patent/CN115590535B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/369—Electroencephalography [EEG]
    • A61B5/377—Electroencephalography [EEG] using evoked responses
    • A61B5/378—Visual stimuli
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/369—Electroencephalography [EEG]
    • A61B5/372—Analysis of electroencephalograms

Abstract

The invention relates to the technical field of electroencephalogram signal identification, and particularly discloses a time window adjusting method, a time window adjusting device, equipment and a storage medium for electroencephalogram signal identification, wherein the method comprises the following steps: s1, obtaining a first identification result; s2, when the first identification result is correct, reducing the time window according to the dynamic step length, and returning to the step S1 until the first identification result is wrong, wherein the dynamic step length is a time window adjusting step length which is negatively related to the using times of the dynamic step length; s3, increasing a time window according to the dynamic step length; s4, acquiring a second identification result; s5, when the second recognition result is correct, returning to the step S4 until the second recognition result reaches a preset threshold value continuously and correctly, and fixing a time window; when the second recognition result is wrong, returning to the step S3; the method realizes the coarse adjustment and fine adjustment processes of the time window based on the dynamic step length, so that the time window can be adaptively optimized towards the direction with high identification accuracy and high information transmission efficiency.

Description

Time window adjusting method, device and equipment for electroencephalogram signal identification and storage medium
Technical Field
The application relates to the technical field of electroencephalogram signal identification, in particular to a time window adjusting method, device, equipment and storage medium for electroencephalogram signal identification.
Background
With the research and development of brain science, brain-computer interface engineering has gained wide attention in the fields of clinical medical treatment, entertainment and leisure, smart home and the like.
The brain-computer interface based on the SSVEP is favored by its advantages of non-invasive, easy training, relatively low cost, etc. The brain-computer interface based on the SSVEP induces the brain-computer response by adopting periodic visual stimulation with specific frequency and carries out feature extraction and target identification.
Target recognition accuracy and information transfer rate are two important indexes for measuring the brain-computer interface system based on SSVEP. In the application process of the traditional SSVEP, a fixed time window algorithm is usually adopted for signal processing, namely, visual stimulation with fixed time duration is adopted for different targets, and signal processing and characteristic identification are carried out on data with the fixed window length; however, in practical applications, the fixed time window is likely to cause problems such as low recognition accuracy and low information transmission efficiency due to scene changes and user changes, and cannot be adjusted according to actual use conditions to achieve both recognition accuracy and information transmission efficiency.
In view of the above problems, no effective technical solution exists at present.
Disclosure of Invention
The application aims to provide a time window adjusting method, a device, equipment and a storage medium for electroencephalogram signal identification, so that the time window can be adaptively optimized and adjusted, and the subsequent electroencephalogram signal identification processing can take the identification accuracy and the information transmission efficiency into consideration.
In a first aspect, the present application provides a time window adjusting method for electroencephalogram signal identification, which is applied to SSVEP electroencephalogram signal identification processing, and the method includes the following steps:
s1, acquiring and identifying electroencephalogram data acquired based on SSVEP induction according to a time window to acquire a first identification result;
s2, when the first identification result is correct, reducing the time window according to a dynamic step length, and returning to the step S1 until the first identification result is wrong, wherein the dynamic step length is a time window adjusting step length with the time length being in negative correlation with the use times of the dynamic step length;
s3, increasing the time window according to the dynamic step length;
s4, acquiring and identifying electroencephalogram data acquired based on SSVEP induction according to the time window to acquire a second identification result;
s5, when the second recognition result is correct, returning to the step S4, and when the continuous correct times of the second recognition result reach a preset threshold value, fixing the time window; and returning to the step S3 when the second recognition result is wrong.
According to the time window adjusting method for electroencephalogram signal identification, electroencephalogram data acquired based on SSVEP induction are acquired and identified based on time windows with different time lengths, the adjusting direction of the time windows is guided according to the identification result, and the coarse adjustment of the steps S1-S2 and the fine adjustment of the steps S3-S5 are achieved based on the dynamic step length, so that the time windows can be adaptively optimized towards the direction with high identification accuracy and high information transmission efficiency.
In step S1 and step S4, the step of acquiring and identifying the electroencephalogram data acquired based on SSVEP induction according to the time window includes:
acquiring electroencephalogram data according to the time window stimulation induction based on SSVEP;
filtering the electroencephalogram data;
and performing feature extraction and target identification on the electroencephalogram data after filtering processing.
The time window adjusting method for electroencephalogram signal identification is characterized in that the initial value of the time window is the median value between the upper limit and the lower limit of the time window.
In this example, the upper limit of the time window is set according to the worst information transmission efficiency, and the lower limit of the time window is set according to the worst identification accuracy, so that the method of the present application sets the initial value of the time window to the median between the upper limit of the time window and the lower limit of the time window, so that the time window can be optimally adjusted under the condition that the values of both the identification accuracy and the information transmission efficiency are initially satisfied, and the adjustment efficiency of the method provided by the present application is effectively improved.
The time window adjusting method for electroencephalogram signal identification is characterized in that the initial value of the dynamic step is 1/20-1/5 of the difference between the upper limit and the lower limit of the time window.
The time window adjusting method for electroencephalogram signal identification is characterized in that the initial value of the dynamic step length is set according to the sampling frequency of the electroencephalogram signal.
The time window adjusting method for electroencephalogram signal identification is characterized in that the dynamic step length is the product of a preset fixed time step length and a decreasing function related to the using times.
The time window adjusting method for electroencephalogram signal identification is characterized in that the subtraction function is an inverse proportion function or a power function which has a negative value and changes with respect to the number of times of use.
In this example, the inverse proportion function and the power function with the power being a negative value are both decreasing functions with a large early-stage change amplitude and a small late-stage change amplitude, so that steps S1-S2 in the method of the present application can achieve coarse adjustment with a larger amplitude and steps S3-S5 can achieve optimization with a higher precision, and the method of the present application can obtain a time window that takes recognition accuracy and information transmission efficiency into account.
In a second aspect, the present application further provides a time window adjusting device for electroencephalogram signal identification, which is applied to the SSVEP electroencephalogram signal identification process, and the device includes:
the first extraction and identification module is used for acquiring and identifying electroencephalogram data acquired based on SSVEP induction according to a time window so as to acquire a first identification result;
the first adjusting module is used for reducing the time window according to the dynamic step length when the first identification result is correct, and re-triggering the first extraction identification module to operate until the first identification result is wrong, wherein the dynamic step length is the time window adjusting step length with the time length being negatively correlated with the use times of the dynamic step length;
the second adjusting module is used for increasing the time window according to the dynamic step length;
the second extraction and identification module is used for acquiring and identifying the electroencephalogram data acquired based on SSVEP induction according to the time window so as to acquire a second identification result;
and the output module is used for re-triggering the second extraction identification module to operate when the second identification result is correct, fixing the time window until the continuous correct times of the second identification result reach a preset threshold value, and is also used for re-triggering the second regulation module to operate when the second identification result is wrong.
The time window adjusting device for electroencephalogram signal identification acquires and identifies electroencephalogram data acquired based on SSVEP induction based on time windows of different time lengths, guides the adjusting direction of the time windows according to an identification result, and realizes the coarse adjustment of the steps S1-S2 and the fine adjustment of the steps S3-S5 based on dynamic step length, so that the time windows can be self-adaptively optimized towards the direction with high identification accuracy and high information transmission efficiency.
In a third aspect, the present application further provides an electronic device, comprising a processor and a memory, where the memory stores computer readable instructions, and the computer readable instructions, when executed by the processor, perform the steps of the method as provided in the first aspect.
In a fourth aspect, the present application also provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, performs the steps of the method as provided in the first aspect.
From the above, the time window adjusting method for electroencephalogram signal identification, the time window adjusting device, the equipment and the storage medium are provided, wherein the time window adjusting method for electroencephalogram signal identification acquires and identifies electroencephalogram data acquired based on SSVEP induction based on time windows with different time lengths, guides the adjusting direction of the time window according to the identification result, and realizes the coarse adjustment of the steps S1-S2 and the fine adjustment of the steps S3-S5 based on the dynamic step length, so that the time window can be adaptively optimized towards the direction with high identification accuracy and high information transmission efficiency, and the subsequent electroencephalogram signal identification processing can take both the identification accuracy and the information transmission efficiency into consideration.
Drawings
Fig. 1 is a flowchart of a time window adjusting method for electroencephalogram signal identification according to an embodiment of the present application.
Fig. 2 is an operation logic diagram of a time window adjusting method for electroencephalogram signal identification according to an embodiment of the present application.
Fig. 3 is a schematic structural diagram of a time window adjusting device for electroencephalogram signal identification provided in the embodiment of the present application.
Fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Reference numerals: 201. a first extraction and identification module; 202. a first adjustment module; 203. a second conditioning module; 204. a second extraction and identification module; 205. an output module; 301. a processor; 302. a memory; 303. a communication bus.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined or explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
The target identification accuracy and the information transmission rate are two important indexes for measuring the brain-computer interface system based on the SSVEP. In practical application, the more effective response signal data acquired by target identification each time is more beneficial to improving the identification accuracy, but the longer the target induction time is, the lower the information transmission rate is.
The traditional SSVEP estimates the optimal time window parameter according to a large amount of off-line training data results and is applied to an on-line experiment; however, when the off-line training device is applied to an actual scene, the fixed time window parameters of the off-line training are not satisfactory in the actual application due to the fact that different application scenes have different background noises and different attention degrees to be tested and other factors.
In a first aspect, referring to fig. 1-2, some embodiments of the present application provide a time window adjusting method for electroencephalogram signal identification, which is applied in an SSVEP electroencephalogram signal identification process, and the method includes the following steps:
s1, acquiring and identifying electroencephalogram data acquired based on SSVEP induction according to a time window to acquire a first identification result;
s2, when the first identification result is correct, reducing the time window according to the dynamic step length, and returning to the step S1 until the first identification result is wrong, wherein the dynamic step length is a time window adjusting step length with the time length being in negative correlation with the use times of the dynamic step length;
s3, increasing a time window according to the dynamic step length;
s4, acquiring and identifying the brain electrical data acquired based on SSVEP induction according to the time window to acquire a second identification result;
s5, when the second recognition result is correct, returning to the step S4 until the continuous correct times of the second recognition result reach a preset threshold value, and fixing a time window; when the second recognition result is erroneous, the process returns to step S3.
Specifically, SSVEP (Steady-State Visual occupied Potentials) refers to a continuous response related to a stimulation frequency (at a fundamental frequency or a multiple frequency of the stimulation frequency) generated by a Visual cortex of a human brain when a Visual stimulation with a fixed frequency is applied to generate a corresponding electroencephalogram signal; the existing brain-computer interface based on SSVEP generally utilizes a time window with fixed length to induce and generate an electroencephalogram signal, the identification accuracy is affected due to insufficient data volume of the electroencephalogram signal caused by too small setting of the time window, and the identification efficiency is affected due to too much data volume needing to be transmitted caused by too large setting of the time window; the brain-computer interface operation based on the SSVEP is generally provided with stimulation identification time of a single command, the time is defined as T, the stimulation identification time of the single command comprises stimulation time (Ta) of a target and rest time (Tb) between two times of target stimulation, namely T = Ta + Tb, the time window determined by the electroencephalogram signal identification time window adjusting method in the embodiment of the application is the stimulation time in the stimulation identification time, and the process of obtaining the optimal time window is the stimulation identification time required by optimally identifying the single command by reducing the stimulation time Ta as far as possible while ensuring that the brain-computer interface based on the SSVEP has high identification accuracy.
More specifically, the brain-computer interface based on the SSVEP has a preset time window, but in the actual use process, the change of the scene and the user may cause the preset time window to no longer have the advantages of high identification accuracy and low data transmission amount (i.e., high information transmission efficiency), thereby affecting the identification process of the whole brain-computer signal; therefore, the time window adjusting method for electroencephalogram signal identification in the embodiment of the present application is particularly suitable for the case where the SSVEP-based brain-computer interface is used for the first time, the user changes, the application scene changes, and the like, that is, the time window adjusting method for electroencephalogram signal identification in the embodiment of the present application is particularly suitable for the case where the SSVEP-based brain-computer interface is used for the first time, the case where the user changes, or the case where the application scene changes, and the like, so as to obtain the optimal time window suitable for the current user and the application scene, and the time window enables the current user to perform electroencephalogram signal identification processing considering both the identification accuracy and the information transmission efficiency in the application scene.
More specifically, the time window used for the first time in step S1 may be a time window with a preset length, or a time window used before a brain-computer interface based on SSVEP, and the time window adjusting method for electroencephalogram signal identification according to the embodiment of the present application aims to obtain electroencephalogram data through repeated stimulation and induction, and implement adaptive adjustment of the time window according to the accuracy of the identification result of the electroencephalogram data, so as to obtain the time window taking recognition accuracy and information transmission efficiency into consideration as the time window for subsequent SSVEP electroencephalogram signal identification processing.
More specifically, it should be understood that the first recognition result and the second recognition result are both recognition results of the electroencephalogram data intercepted with respect to the corresponding time window, and the recognition results may be correct or incorrect; step S2 and step S5 are processes of adjusting the time window based on the first recognition result and based on the second recognition result, respectively, and therefore, it should be understood that step S2 and step S5 respectively include a judgment process regarding the correctness of the first recognition result and a judgment process regarding the correctness of the second recognition result; it should also be understood that, in order to ensure the accuracy of the determination process, the time window adjustment method for electroencephalogram signal identification in the embodiment of the present application performs induction based on a known command to acquire electroencephalogram data, and preferably performs induction based on the same known single command, so as to ensure that the determination on the correctness of the first identification result and the second identification result is accurate, and therefore, in the embodiment of the present application, the process of determining the accuracy of the first identification result and the second identification result is equivalent to a process of comparing the first identification result and the second identification result with a previously calibrated identification result.
More specifically, the dynamic step is a time length value for adjusting the length of each time window, the time length of the dynamic step is set to a change value associated with the number of times of using the dynamic step itself, so that the time window is adjusted based on different time length values each time, wherein the time window adjusting process includes a decreasing process in step S2 and an increasing process in step S3, and the dynamic step is a negative correlation setting based on the number of times of using, so that the variation of each adjustment of the time window is smaller and smaller, and the time window can gradually narrow the adjusting range to tend to change the time window which takes into account the recognition accuracy and the information transmission efficiency.
More specifically, since the dynamic step size becomes smaller with the number of times of use, in the embodiment of the present application, step S1 to step S2 are equivalent to a coarse adjustment process of a time window, and step S3 to step S5 are equivalent to a fine adjustment process of the time window; based on the foregoing, too much electroencephalogram data may be generated if the time window is too long, and the influence of the redundant electroencephalogram data on the accuracy of the recognition result is not great, so that the time window has redundant useless time length, therefore, steps S1 to S2 in the embodiment of the present application can greatly and efficiently remove the useless time length in the time window, and adjust the time window to a state in which the accuracy is slightly lost, and then steps S3 to S5 are used to finely adjust the time length value of the increased time window, so as to gradually obtain the time window in which the recognition accuracy meets the use requirement, thereby implementing adaptive adjustment of the time window.
More specifically, in the step S2, the error of the first recognition result is taken as an end point of the coarse tuning process, which indicates that all useless time lengths are removed from the current time window, but the data volume is not enough to ensure accurate identification of the electroencephalogram data; when the correct second identification result is generated for the first time in the step S5, the data size of the electroencephalogram data acquired based on the current time window can preliminarily guarantee the accuracy of electroencephalogram data identification, but if the identification accuracy needs to be guaranteed to be effective for a long time, repeated verification of electroencephalogram data identification needs to be carried out, so that the step S5 needs to return to the step S4 to repeatedly acquire the second identification result for accuracy verification, and the requirement of high identification accuracy can be met only when the continuous correct times of the second identification result reach the preset threshold value and the current time window can be considered to have high information transmission efficiency.
More specifically, when the second recognition result in step S5 is incorrect, it indicates that the current time window does not meet the requirement of recognition accuracy, so that it needs to be adjusted appropriately according to step S3.
More specifically, the preset threshold is a value of the number of times of correct continuous identification set according to the use requirement, and may be set according to the actual identification accuracy or the historical identification accuracy or the use instruction of the brain-computer interface based on the SSVEP.
More specifically, the process of fixing the time window in step S5 is a process of setting the current time window as a time window for the subsequent electroencephalogram signal identification processing, that is, fixing the output of the time window as a time window for the electroencephalogram signal identification processing.
According to the time window adjusting method for electroencephalogram signal identification, electroencephalogram data obtained based on SSVEP induction are obtained and identified based on time windows of different time lengths, the adjusting direction of the time windows is guided according to the identification result, the coarse adjustment of the steps S1-S2 and the fine adjustment of the steps S3-S5 are achieved based on the dynamic step length, the time windows can be self-adaptively optimized towards the direction with high identification accuracy and high information transmission efficiency, and the subsequent electroencephalogram signal identification processing is achieved while the identification accuracy and the information transmission efficiency are taken into consideration.
In some preferred embodiments, the method of the embodiment of the present application is provided with an upper time window limit and a lower time window limit, which are respectively used for defining the maximum value and the minimum value after the time window is adjusted, and when the time window exceeds the upper time window limit and the lower time window limit, it indicates that there is a defect or problem in the current adjustment process, and it is necessary to terminate the operation of the whole method, and perform maintenance on the brain-computer interface based on SSVEP.
Specifically, the upper limit and the lower limit of the time window are respectively defined as Tmax and Tmin, which are both set based on the use requirement of the system (the brain-computer interface of the SSVEP) and the device operation parameters, wherein the Tmax is set in consideration of the information transmission rate requirement of the system; when the Tmax is set to be too large, the time window has a larger upper limit, which easily causes that the adjusted time window is too long to cause that the information transmission rate of the system is obviously reduced; if the Tmin is set to be too small, the time window has a smaller lower limit, and the adjusted time window is too short, so that the identification accuracy of the system is obviously reduced.
In some preferred embodiments, in step S1 and step S4, the step of acquiring and identifying the acquired electroencephalogram data based on SSVEP induction according to the time window includes:
acquiring electroencephalogram data according to time window stimulation induction based on SSVEP;
filtering the electroencephalogram data;
and performing feature extraction and target identification on the electroencephalogram data after filtering processing.
Specifically, the filtering process is a preprocessing process of electroencephalogram data, and needs to be selected according to the type of the electroencephalogram data, if four-target SSVEP data (stimulation frequency 6Hz-12 Hz) with the sampling frequency of 305Hz are selected, a second-order Butterworth filter is selected to filter the electroencephalogram data, and the lowest cut-off frequency and the highest cut-off frequency of the filter used in the filtering process are respectively set to be 4Hz and 30Hz; the method of the embodiment of the application can be used for filtering the electroencephalogram data, so that the phenomenon that noise exists in the electroencephalogram data to influence the identification accuracy can be avoided.
More specifically, in the embodiment of the present application, the feature extraction and target identification processing is performed by using a typical correlation analysis (CCA), and the processing mainly includes: respectively extracting two representative comprehensive variables (which are linear combinations of the variables in the two variable groups) from the two groups of variables, and reflecting the overall correlation between the two groups of indexes by using the correlation between the two comprehensive variables so as to judge whether the contents of the two groups of data representations are consistent; the method of the embodiment of the application can analyze the overall correlation of the acquired electroencephalogram data and the pre-calibrated reference data based on the algorithm so as to acquire the identification result; in some other embodiments, the process of obtaining the first recognition result and/or the second recognition result by the method according to the embodiments of the present application may be further simplified to analyze the overall correlation between the electroencephalogram data and the reference data corresponding to the previously calibrated recognition result based on the above algorithm to determine whether the contents represented by the electroencephalogram data and the reference data are consistent to generate the first recognition result and/or the second recognition result containing the error information or the correct information.
More specifically, the process of applying the above typical correlation analysis algorithm to the method of the embodiment of the present application is as follows:
the acquired electroencephalogram data is assumed to be multidimensional data
Figure DEST_PATH_IMAGE001
The reference data is data->
Figure DEST_PATH_IMAGE002
Means for combining data>
Figure 130749DEST_PATH_IMAGE001
And data->
Figure 375785DEST_PATH_IMAGE002
Linearly transformed into one-dimensional>
Figure DEST_PATH_IMAGE003
And &>
Figure DEST_PATH_IMAGE004
And satisfies the following conditions:
Figure DEST_PATH_IMAGE005
（1）
wherein:
Figure DEST_PATH_IMAGE006
（2）
Figure DEST_PATH_IMAGE007
is SCorresponding target frequency, <' > or >>
Figure DEST_PATH_IMAGE008
In order to be able to sample the frequency,Th is the number of cycles of 2 pi for the number of data points collected corresponding to the time window, i.e., H =1,2,3 … … N;
the optimization goal to be able to write a CCA is to maximize ρ: (b:)
Figure 456655DEST_PATH_IMAGE003
,/>
Figure 383023DEST_PATH_IMAGE004
)，ρ(/>
Figure 728553DEST_PATH_IMAGE003
,/>
Figure 832907DEST_PATH_IMAGE004
) Is->
Figure 896678DEST_PATH_IMAGE003
And &>
Figure 626736DEST_PATH_IMAGE004
Can represent the correlation of the two to obtain the corresponding projection vector->
Figure DEST_PATH_IMAGE009
,/>
Figure DEST_PATH_IMAGE010
I.e. by
Figure DEST_PATH_IMAGE011
（3）
Before projection, the raw data is generally normalized and the data is normalized
Figure 168051DEST_PATH_IMAGE001
And data->
Figure 692574DEST_PATH_IMAGE002
Normalization is performed to convert to data with a mean of 0 and a variance of 1, and is performed as follows:
Figure DEST_PATH_IMAGE012
（4）
wherein the content of the first and second substances,covin the form of a function of the covariance,Din the form of a function of the variance,Eis a mean function;
by
Figure 259952DEST_PATH_IMAGE001
And &>
Figure 528123DEST_PATH_IMAGE002
The condition that the mean value of the converted data is 0 can be obtained:
Figure DEST_PATH_IMAGE013
（5）
therefore, formula (3) can be rewritten as follows:
Figure DEST_PATH_IMAGE014
（6）
therefore, the specific operation flow of the typical correlation analysis algorithm is as follows:
inputting: are each data
Figure 395716DEST_PATH_IMAGE001
And data->
Figure 356718DEST_PATH_IMAGE002
Wherein->
Figure 129502DEST_PATH_IMAGE001
And &>
Figure 686517DEST_PATH_IMAGE002
Are all greater than 1;
and (3) outputting: ρ, data
Figure 861146DEST_PATH_IMAGE001
And data->
Figure 727471DEST_PATH_IMAGE002
Projection vector(s)>
Figure 987551DEST_PATH_IMAGE009
And &>
Figure 339467DEST_PATH_IMAGE010
Wherein ρ can be taken directly as recognition result or the decision data ≥ based on ρ>
Figure 368603DEST_PATH_IMAGE001
And data->
Figure 671409DEST_PATH_IMAGE002
Whether the characterized contents are consistent or not to generate a corresponding recognition result;
the specific calculation process is as follows: calculating data
Figure 153205DEST_PATH_IMAGE001
Variance of (2)S XX Data->
Figure 317602DEST_PATH_IMAGE002
Variance of (2)S YY ，/>
Figure 201244DEST_PATH_IMAGE002
And &>
Figure 674951DEST_PATH_IMAGE001
Covariance ofS XY =S YY T；
Then calculate the matrix
Figure DEST_PATH_IMAGE015
；
For matrixMSingular value decomposition is carried out to obtain a maximum singular value rho and a singular vector corresponding to the maximum singular value;
the reference signal frequency corresponding to the maximum singular value ρ output in each test is the target frequency of system identification (the electroencephalogram signal identification processing process is a process of identifying the target frequency), i.e. the identification result.
In some preferred embodiments, the initial value of the time window is the median value between the upper time window limit and the lower time window limit.
Specifically, in the application embodiment, since it is not specifically clear that the difference strength caused by the application of the scene change and the user change is applied, the initial value of the time window should be subjected to the initialization processing, so that the optimal adjustment of the time window is performed based on the same reference each time the method of the application embodiment is used.
More specifically, as can be seen from the foregoing, the upper limit of the time window is set according to the worst information transmission efficiency, and the lower limit of the time window is set according to the worst identification accuracy, so that the method according to the embodiment of the present application sets the initial value of the time window to the median between the upper limit of the time window and the lower limit of the time window, so that the time window can be optimally adjusted under the condition that both the identification accuracy and the information transmission efficiency are initially satisfied, thereby effectively improving the adjustment efficiency of the method according to the embodiment of the present application.
In some preferred embodiments, the upper limit of the time window is 5S, and the lower limit of the time window is 1S, so that the initial value of the time window is preferably 3S, and the above setting of the time window can meet the use of a general brain-computer interface based on SSVEP.
In some preferred embodiments, the initial value of the dynamic step size is 1/20 to 1/5 of the difference between the upper and lower time window limits.
Specifically, the dynamic step should be set such that the time window does not exceed the upper limit of the time window and the lower limit of the time window in the course of initial coarse adjustment for several times, and therefore, the method of the embodiment of the present application sets the initial value of the dynamic step to 1/20-1/5 of the difference between the upper limit of the time window and the lower limit of the time window, so that the length of the dynamic step of each adjustment of the time window is moderate, and the time window has a sufficient number of adjustment spaces.
In some preferred embodiments, the initial value of the dynamic step is set according to the sampling frequency of the brain electrical signal.
Specifically, in the process of electroencephalogram signal identification processing, the number of data points contained in electroencephalogram data acquired each time is determined by the sampling frequency and the length of a time window; therefore, in the method of the embodiment of the present application, the length of the dynamic step determines the number of changes of data points in the electroencephalogram data acquired corresponding to the adjusted time window, and the number of increases or decreases of the data points corresponding to the time window can be directly analyzed by combining the length of the dynamic step with the sampling frequency, where if the sampling frequency is 305Hz and the dynamic step increased by the time window is 0.4S, 305 × 0.4=122 data points are added to the electroencephalogram data acquired corresponding to the adjusted time window; therefore, the initial value of the dynamic step is set according to the sampling frequency of the electroencephalogram signal, and the variation of data points in the electroencephalogram data which are correspondingly collected in the initial stage of the time window can be accurately regulated according to actual use requirements.
In some preferred embodiments, the dynamic step size is the product of a predetermined fixed time step and a decreasing function with respect to the number of uses.
Specifically, in the embodiment of the present application, the dynamic step is divided into two parts, so that a user can conveniently set a suitable dynamic step according to a use requirement, so that the time window adjusting method for electroencephalogram signal identification in the embodiment of the present application has different advantages, for example, designing a reduction function with a larger variation amplitude can improve the efficiency of adjusting the time window by the method in the embodiment of the present application, and for example, designing a reduction function with a smaller variation amplitude can improve the compromise effect of the time window obtained by the method in the embodiment of the present application.
In some preferred embodiments, the decreasing function is an inverse proportional function or a power function that is negative with respect to the change in the number of uses and that varies with respect to the number of uses.
Specifically, the inverse proportion function and the power function with the power being a negative value are both decreasing functions with a large early-stage variation amplitude and a small late-stage variation amplitude, so that steps S1 to S2 in the method of the embodiment of the present application can achieve coarse adjustment with a larger amplitude, and steps S3 to S5 can achieve optimization with higher precision, so that the method of the embodiment of the present application can obtain a time window that takes into account both the identification accuracy and the information transmission efficiency.
More specifically, the subtraction function is a subtraction function within the interval (0, + ∞), and when the subtraction function appears as an inverse proportional function with respect to the change in the number of uses, satisfies:
y=k/x （7）
wherein x is the number of times of use, k is a constant greater than 0, and y is the output value of a subtraction function;
when the subtraction function appears as a power function in which the power is negative and varies with respect to the number of uses, it satisfies:
y=kx a （8）
wherein x is the number of times of use, k is a constant greater than 0, a is a constant less than 0, and y is the output value of the subtraction function.
In a second aspect, please refer to fig. 3, some embodiments of the present application further provide a time window adjusting device for electroencephalogram signal identification, which is applied in the SSVEP electroencephalogram signal identification process, and the device includes:
the first extraction and identification module 201 is configured to acquire and identify electroencephalogram data acquired based on SSVEP induction according to a time window to acquire a first identification result;
the first adjusting module 202 is configured to reduce the time window according to the dynamic step length when the first identification result is correct, and re-trigger the first extraction identification module to operate until the first identification result is incorrect, where the dynamic step length is a time window adjusting step length in which the time length is negatively correlated with the number of times of use of the dynamic step length;
a second adjusting module 203, configured to increase the time window according to the dynamic step;
the second extraction and identification module 204 is configured to acquire and identify electroencephalogram data acquired based on SSVEP induction according to a time window to acquire a second identification result;
the output module 205 is configured to, when the second recognition result is correct, re-trigger the second extraction recognition module to operate until the consecutive correct times of the second recognition result reach a preset threshold, and fix the time window; and the second regulating module is also used for re-triggering the second regulating module to operate when the second identification result is wrong.
The time window adjusting device for electroencephalogram signal identification obtains and identifies electroencephalogram data obtained based on SSVEP induction based on time windows with different time lengths, guides the adjustment direction of the time windows according to identification results, and realizes the coarse adjustment of the steps S1-S2 and the fine adjustment of the steps S3-S5 based on dynamic step length, so that the time windows can be adaptively optimized towards the direction with high identification accuracy and high information transmission efficiency, and the subsequent electroencephalogram signal identification processing can take the identification accuracy and the information transmission efficiency into consideration.
In some preferred embodiments, the time window adjusting device for brain electrical signal identification in the embodiments of the present application is used to execute the time window adjusting method for brain electrical signal identification provided in the first aspect described above.
In a third aspect, referring to fig. 4, some embodiments of the present application further provide a schematic structural diagram of an electronic device, where the present application provides an electronic device, including: the processor 301 and the memory 302, the processor 301 and the memory 302 being interconnected and communicating with each other via a communication bus 303 and/or other form of connection mechanism (not shown), the memory 302 storing computer readable instructions executable by the processor 301, the processor 301 executing the computer readable instructions when the electronic device is operated to perform the method of any of the alternative implementations of the above-described embodiments.
In a fourth aspect, the present application provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the computer program performs the method in any optional implementation manner of the foregoing embodiments. The storage medium may be implemented by any type of volatile or nonvolatile storage device or combination thereof, such as a Static Random Access Memory (SRAM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), an Erasable Programmable Read-Only Memory (EPROM), a Programmable Read-Only Memory (PROM), a Read-Only Memory (ROM), a magnetic Memory, a flash Memory, a magnetic disk, or an optical disk.
In summary, the present application provides a time window adjusting method, an apparatus, a device, and a storage medium for electroencephalogram signal identification, wherein the time window adjusting method for electroencephalogram signal identification in the present application embodiment acquires and identifies electroencephalogram data acquired based on SSVEP induction based on time windows of different time lengths, guides an adjustment direction of the time window according to an identification result, and implements the coarse adjustment of step S1 to step S2 and the fine adjustment of step S3 to step S5 based on a dynamic step size, so that the time window can be adaptively optimized toward a direction with high identification accuracy and high information transmission efficiency, so that the subsequent electroencephalogram signal identification processing can take into account the identification accuracy and the information transmission efficiency.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
In addition, units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.
In this document, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made to the present application by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims (10)

1. A time window adjusting method for brain electrical signal identification is applied to SSVEP brain electrical signal identification processing, and is characterized in that the time window is stimulation time in stimulation identification time, and the method comprises the following steps:
s1, acquiring and identifying electroencephalogram data acquired based on SSVEP induction according to a time window to acquire a first identification result;
s2, when the first identification result is correct, reducing the time window according to a dynamic step length, and returning to the step S1 until the first identification result is wrong, wherein the dynamic step length is a time window adjusting step length with the time length being in negative correlation with the use times of the dynamic step length;
s3, increasing the time window according to the dynamic step length;
s4, acquiring and identifying the brain electrical data acquired based on SSVEP induction according to the time window to acquire a second identification result;
s5, when the second recognition result is correct, returning to the step S4, and fixing the time window until the continuous correct times of the second recognition result reach a preset threshold value; and returning to the step S3 when the second recognition result is wrong.
2. The method for adjusting time window for electroencephalogram signal identification according to claim 1, wherein in step S1 and step S4, the step of acquiring and identifying electroencephalogram data acquired based on SSVEP induction from the time window comprises:
acquiring electroencephalogram data according to the time window stimulation induction based on SSVEP;
filtering the electroencephalogram data;
and performing feature extraction and target identification on the electroencephalogram data after filtering processing.
3. The method of adjusting a time window for brain electrical signal identification according to claim 1, wherein the initial value of the time window is the median between the upper time window limit and the lower time window limit.
4. The method for adjusting time window of electroencephalogram signal identification according to claim 1, wherein the initial value of the dynamic step is 1/20-1/5 of the difference between the upper limit of the time window and the lower limit of the time window.
5. The method for adjusting time window of electroencephalogram signal identification according to claim 4, wherein the initial value of the dynamic step is set according to the sampling frequency of the electroencephalogram signal.
6. The method for adjusting a time window for brain electrical signal identification according to any one of claims 1-5, wherein said dynamic step size is a product of a preset fixed time step and a decreasing function with respect to said number of uses.
7. The method of adjusting a time window for brain electrical signal identification according to claim 6, wherein said decreasing function is an inverse proportional function or a power function with a negative value with respect to the change of the number of times of use and a power function with respect to the change of the number of times of use.
8. A time window adjusting device for brain electrical signal identification is applied to SSVEP brain electrical signal identification processing, and is characterized in that the time window is a stimulation time in stimulation identification time, and the device comprises:
the first extraction and identification module is used for acquiring and identifying electroencephalogram data acquired based on SSVEP induction according to a time window so as to acquire a first identification result;
the first adjusting module is used for reducing the time window according to the dynamic step length when the first identification result is correct, and re-triggering the first extraction identification module to operate until the first identification result is wrong, wherein the dynamic step length is the time window adjusting step length with the time length negatively correlated with the use times of the dynamic step length;
the second adjusting module is used for increasing the time window according to the dynamic step length;
the second extraction and identification module is used for acquiring and identifying the electroencephalogram data acquired based on SSVEP induction according to the time window so as to acquire a second identification result;
and the output module is used for re-triggering the second extraction identification module to operate when the second identification result is correct, fixing the time window until the continuous correct times of the second identification result reach a preset threshold value, and is also used for re-triggering the second regulation module to operate when the second identification result is wrong.
9. An electronic device comprising a processor and a memory, said memory storing computer readable instructions which, when executed by said processor, perform the steps of the method according to any one of claims 1 to 7.
10. A storage medium on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1-7.
CN202211462180.6A 2022-11-17 2022-11-17 Time window adjusting method, device and equipment for electroencephalogram signal identification and storage medium Active CN115590535B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211462180.6A CN115590535B (en) 2022-11-17 2022-11-17 Time window adjusting method, device and equipment for electroencephalogram signal identification and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211462180.6A CN115590535B (en) 2022-11-17 2022-11-17 Time window adjusting method, device and equipment for electroencephalogram signal identification and storage medium

Publications (2)

Publication Number Publication Date
CN115590535A CN115590535A (en) 2023-01-13
CN115590535B true CN115590535B (en) 2023-03-31

Family

ID=84852971

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211462180.6A Active CN115590535B (en) 2022-11-17 2022-11-17 Time window adjusting method, device and equipment for electroencephalogram signal identification and storage medium

Country Status (1)

Country Link
CN (1) CN115590535B (en)

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
AU2008212131A1 (en) * 2007-02-09 2008-08-14 Agency For Science, Technology And Research A system and method for processing brain signals in a BCI system
JP5711056B2 (en) * 2011-06-21 2015-04-30 アイシン精機株式会社 EEG interface system
KR20210053631A (en) * 2019-11-04 2021-05-12 한국과학기술원 Method of entraining gamma brain wave using visible light and apparatus for providing visible stimulation
CN112099619B (en) * 2020-08-11 2024-02-06 东南大学 Time window length self-adaptive selection method for mixed vision brain-computer interface
CN115153592A (en) * 2022-07-14 2022-10-11 北京理工大学 Brain-eye synchronous mechanical arm control system with non-fixed stimulation duration

Also Published As

Publication number Publication date
CN115590535A (en) 2023-01-13

Similar Documents

Publication Publication Date Title
Kantz et al. Dimension estimates and physiological data
Au Uncertainty law in ambient modal identification—Part I: Theory
Mandic et al. On the characterization of the deterministic/stochastic and linear/nonlinear nature of time series
Khare et al. Classification of emotions from EEG signals using time‐order representation based on the S‐transform and convolutional neural network
CN110047519B (en) Voice endpoint detection method, device and equipment
CN109087670A (en) Mood analysis method, system, server and storage medium
CN111700718B (en) Method and device for recognizing holding gesture, artificial limb and readable storage medium
CN109635864B (en) Fault-tolerant control method and device based on data
Rice et al. Tests for conditional heteroscedasticity of functional data
CN111061877A (en) Text theme extraction method and device
CN115590535B (en) Time window adjusting method, device and equipment for electroencephalogram signal identification and storage medium
Vanella Stochastic forecasting of demographic components based on principal component analyses
CN109388781B (en) Method and device for processing measurement data
Gui et al. Exploring missing data prediction in medical monitoring: A performance analysis approach
Subbotin A multi-horizon scale for volatility
CN102081124B (en) System and method for identifying high-speed peripheral equipment interconnected signal
CN111091099A (en) Scene recognition model construction method, scene recognition method and device
CN109308660B (en) Credit assessment scoring model evaluation method, apparatus, device and storage medium
CN112612844A (en) Data processing method, device, equipment and storage medium
CN109783745B (en) Method, device and computer equipment for personalized typesetting of pages
CN116028556B (en) Intelligent processing method for transmission data of AR terminal system
CN116403080B (en) Face clustering evaluation method, system, computer and readable storage medium
JP7202260B2 (en) HYPER-PARAMETER MANAGEMENT DEVICE, HYPER-PARAMETER MANAGEMENT SYSTEM AND HYPER-PARAMETER MANAGEMENT METHOD
CN110827811A (en) Voice control method and device for household electrical appliance
CN113791391B (en) Interference characteristic parameter identification method, device, storage and 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