CN109800733A - Data processing method and device, electronic equipment - Google Patents
Data processing method and device, electronic equipment Download PDFInfo
- Publication number
- CN109800733A CN109800733A CN201910092534.4A CN201910092534A CN109800733A CN 109800733 A CN109800733 A CN 109800733A CN 201910092534 A CN201910092534 A CN 201910092534A CN 109800733 A CN109800733 A CN 109800733A
- Authority
- CN
- China
- Prior art keywords
- electromyography signal
- test data
- signal test
- presently described
- data
- 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
Abstract
The present invention provides a kind of data processing methods, comprising: each electromyography signal test data is successively input to the object classifiers constructed in advance;When each electromyography signal test data is input to object classifiers, the first test result corresponding with current electromyography signal test data is obtained, and judge whether current electromyography signal test data is first electromyography signal test data;If so, generating ballot queue, and determine corresponding second test result of electromyography signal test data;If it is not, then being matched the corresponding difference value of current electromyography signal test data to obtain the second test result with preset threshold value;Ballot queue is updated according to the first test result and the second test result, and votes current electromyography signal test data to obtain recognition result according to updated ballot queue, improves the accuracy of the recognition result of electromyography signal test data.
Description
Technical field
The present invention relates to electromyography signals to identify field, in particular to a kind of data processing method and device, electronic equipment.
Background technique
In recent years, machine learning also has significant progress, such as based on deep learning with the development of Information technology
Surface electromyogram signal identify field, surface electromyogram signal is the neuromuscular to be got off from muscle surface by electrode leader record
Bioelectrical signals when system activity can react nervimuscular active state to a certain extent, can be known by deep learning
The corresponding muscle activity state of not different surface electromyogram signals produces the fields such as medical science of recovery therapy, sports medical science and bio-mechanical
Raw huge power-assisted.
Through the present invention staff the study found that in the existing method based on deep learning identification surface electromyogram signal, usually
It is building disaggregated model, the electromyography signal collected is inputted into disaggregated model, obtains the recognition result of electromyography signal, however,
Since electromyography signal has unstability and randomness, the recognition result obtained by classifier often will appear larger
Error, therefore, how to solve the problems, such as electromyography signal instability and randomness caused by recognition result inaccuracy become
Those skilled in the art urgently solve the problems, such as.
Summary of the invention
Technical problem to be solved by the invention is to provide a kind of data processing methods, and the electromyography signal obtained in advance is surveyed
It tries data input classifier and obtains the first test result, and determine the corresponding second test knot of the electromyography signal test data
Fruit updates ballot queue according to the first test result and the second test result, and is identified according to updated ballot queue
As a result, improving the accuracy of the recognition result of the electromyography signal test data.
The present invention also provides a kind of data processing equipments, to guarantee the realization and application of the above method in practice.
A kind of data processing method, comprising:
Obtain pre-stored multiple electromyography signal test datas, and drawing according to each electromyography signal test data
Divide sequence, each electromyography signal test data is successively input to the object classifiers constructed in advance;
When each electromyography signal test data is input to the object classifiers, obtains and believe with presently described myoelectricity
Number corresponding first test result of test data, and judge whether presently described electromyography signal test data is first electromyography signal
Test data;The presently described electromyography signal test data is currently to be input to the electromyography signal test of the object classifiers
Data;
If presently described electromyography signal test data is first electromyography signal test data, ballot queue is generated, and really
Determine corresponding second test result of presently described electromyography signal test data, wherein the ballot queue is for storing each institute
State corresponding first test result of electromyography signal test data;
If presently described electromyography signal test data is not first electromyography signal test data, presently described myoelectricity is believed
Number corresponding difference value of test data is matched to obtain corresponding with presently described electromyography signal test data with preset threshold value
The second test result, wherein the difference value characterizes presently described electromyography signal test data and the previous myoelectricity is believed
The difference degree of number test data;
It is updated according to corresponding first test result of presently described electromyography signal test data and second test result
The ballot queue, and vote presently described electromyography signal test data to be known according to updated ballot queue
Other result.
Above-mentioned method, optionally, the storing process of the electromyography signal test data, comprising:
Acquire the electromyography signal that user's execution movement generates;
Using sliding window setting technique, the time sequencing generated according to the electromyography signal divides the electromyography signal,
Multiple electromyography signal test datas are obtained, and each electromyography signal test data obtained to division successively stores.
Above-mentioned method, optionally, before storing the electromyography signal test data, further includes:
Electromyography signal test data to be stored is analyzed, is obtained corresponding with the electromyography signal test data to be stored
Temporal signatures;
According to temporal signatures building two-dimentional myoelectricity characteristic image corresponding with the electromyography signal test data.
Above-mentioned method, optionally, it is described by the corresponding difference value of presently described electromyography signal test data with it is preset
Threshold value is matched to obtain the second test result corresponding with presently described electromyography signal test data, comprising:
Characterization and the current institute according to the corresponding two-dimentional myoelectricity characteristic image of presently described electromyography signal test data
The characterization of the corresponding two-dimentional myoelectricity characteristic image of previous electromyography signal test data of electromyography signal test data is stated, determination is worked as
The corresponding difference value of the preceding electromyography signal test data;
The difference value and preset threshold value are matched to obtain corresponding with presently described electromyography signal test data
Second test result.
Above-mentioned method, optionally, if the presently described electromyography signal test data is first electromyography signal test number
According to then generating ballot queue, determine corresponding second test result of the electromyography signal test data, comprising:
If presently described electromyography signal test data is first electromyography signal test data, ballot queue is generated;
The state for determining presently described ballot queue, the state according to the presently described ballot queue determine presently described
Corresponding second test result of electromyography signal test data is non-action transfer point.
Above-mentioned method, it is optionally, described according to corresponding first test result of presently described electromyography signal test data
And second test result updates the ballot queue, comprising:
When second test result is non-action transfer point, by presently described electromyography signal test data corresponding the
One test result is stored into the ballot queue, to realize the update of the ballot queue.
Above-mentioned method, optionally, it is described by the corresponding difference value of presently described electromyography signal test data with it is preset
Threshold value is matched to obtain second test result, comprising:
If the difference value is less than preset first threshold, it is determined that presently described electromyography signal test data corresponding the
Two test results are movement transfer point;
If the difference value is greater than preset second threshold, it is determined that presently described electromyography signal test data corresponding the
Two test results are non-action transfer point;
If the difference value is more than or equal to preset first threshold and is less than or equal to preset second threshold, according to described in
Ballot queue obtains the voting results of presently described electromyography signal measured data;
If the voting results the first test result corresponding with presently described electromyography signal test data is consistent, it is determined that
Second test result of presently described electromyography signal test data is non-action transfer point;
If the voting results the first test result corresponding with presently described electromyography signal test data is inconsistent, really
Second test result of fixed presently described electromyography signal test data is movement transfer point.
Above-mentioned method, it is optionally, described according to corresponding first test result of presently described electromyography signal test data
And second test result updates the ballot queue, comprising:
When corresponding second test result of presently described electromyography signal test data is non-action transfer point, by current institute
It states corresponding first test result of electromyography signal test data to store into the ballot queue, to realize the ballot queue
It updates;
It, will be described when corresponding second test result of presently described electromyography signal test data is movement transfer point
The quene state of ballot queue is set as empty, and corresponding first test result of presently described electromyography signal test data is stored
Extremely in the ballot queue, to realize the update of the ballot queue.
A kind of data processing equipment, comprising:
Acquiring unit is believed for obtaining pre-stored multiple electromyography signal test datas, and according to each myoelectricity
The stripe sequence of number test data, is successively input to the target classification constructed in advance for each electromyography signal test data
Device;
Judging unit, when for each electromyography signal test data to be input to the object classifiers, obtain with
Corresponding first test result of presently described electromyography signal test data, and whether judge presently described electromyography signal test data
For first electromyography signal test data;The presently described electromyography signal test data is currently to be input to the object classifiers
Electromyography signal test data;
First determination unit, for when presently described electromyography signal test data is first electromyography signal test data,
Ballot queue is generated, and determines corresponding second test result of presently described electromyography signal test data, wherein the ballot team
Column are for storing corresponding first test result of each electromyography signal test data;
Second determination unit, for not being first electromyography signal test data when presently described electromyography signal test data
When, the corresponding difference value of presently described electromyography signal test data is matched to obtain and presently described flesh with preset threshold value
Corresponding second test result of electric signal test data, wherein the difference value characterizes presently described electromyography signal test data
With the difference degree of the previous electromyography signal test data;
Recognition unit, for according to corresponding first test result of presently described electromyography signal test data and described second
Test result updates the ballot queue, and carries out according to updated ballot queue to presently described electromyography signal test data
Ballot is to obtain recognition result.
A kind of electronic equipment, including memory and one perhaps one of them or one of more than one instruction with
Upper instruction is stored in memory, and is configured to execute above-mentioned data processing side by one or more than one processor
Method.
Compared with prior art, the present invention includes the following advantages:
The present invention provides a kind of data processing methods, comprising: pre-stored multiple electromyography signal test datas are obtained,
And according to the stripe sequence of each electromyography signal test data, successively each electromyography signal test data is input to
The object classifiers constructed in advance;When each electromyography signal test data is input to the object classifiers, obtain with
Corresponding first test result of presently described electromyography signal test data, and whether judge presently described electromyography signal test data
For first electromyography signal test data;If presently described electromyography signal test data is first electromyography signal test data, give birth to
At ballot queue, and determine corresponding second test result of presently described electromyography signal test data;If presently described myoelectricity letter
Number test data is not first electromyography signal test data, then by the corresponding difference value of presently described electromyography signal test data with
Preset threshold value is matched to obtain second test result;According to presently described electromyography signal test data corresponding first
Test result and second test result update the ballot queue, and according to updated ballot queue to presently described flesh
Electric signal test data is voted to obtain recognition result, and the standard of the recognition result of the electromyography signal test data is improved
True property.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without any creative labor, it can also be obtained according to these attached drawings
His attached drawing.
Fig. 1 is a kind of method flow diagram of data processing method provided by the invention;
Fig. 2 is an a kind of exemplary diagram of data processing method provided by the invention;
Fig. 3 is a kind of structural schematic diagram of data processing equipment provided by the invention;
Fig. 4 is the structural schematic diagram of a kind of electronic equipment provided by the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The present invention can be used in numerous general or special purpose computing device environment or configurations.Such as: personal computer, service
Device computer, handheld device or portable device, laptop device, multi-processor device including any of the above devices or devices
Distributed computing environment etc..
The embodiment of the invention provides a kind of data processing method, this method can be applied in multiple systems platform, be held
Row main body can be terminal or the processor of various mobile devices, and the method flow diagram of the method is as shown in Figure 1, tool
Body includes:
S101: pre-stored multiple electromyography signal test datas are obtained, and number is tested according to each electromyography signal
According to stripe sequence, each electromyography signal test data is successively input to the object classifiers constructed in advance.
In method provided in an embodiment of the present invention, the object classifiers are by supervised learning, semi-supervised learning or nothing
The classifier of supervised learning.
In method provided in an embodiment of the present invention, the stripe sequence can be time sequencing or put in order.
In method provided in an embodiment of the present invention, the electromyography signal test data can execute a certain movement pair for user
Any one the electromyography signal test data in multiple electromyography signal test datas answered.
S102: when each electromyography signal test data is input to the object classifiers, obtain with it is presently described
Corresponding first test result of electromyography signal test data, and judge whether presently described electromyography signal test data is first flesh
Electric signal test data;The presently described electromyography signal test data is the myoelectricity letter for being currently input to the object classifiers
Number test data.
In method provided in an embodiment of the present invention, by the way that electromyography signal test data is input to the object classifiers,
It can get the classification results of the electromyography signal test data, i.e., tentatively identify that electromyography signal is tested by the object classifiers
The corresponding type of action of data, the classification results are first test result.
S103: if presently described electromyography signal test data is first electromyography signal test data, generating ballot queue,
And determine corresponding second test result of presently described electromyography signal test data, wherein the ballot queue is each for storing
Corresponding first test result of a electromyography signal test data.
In method provided in an embodiment of the present invention, the ballot queue can be used for storing presently described electromyography signal test number
According to the first test result of corresponding first test result or history, the original state of the ballot queue is sky, i.e., does not store and appoint
What data;First test result of history is the of electromyography signal test data before current electromyography signal test data
One test result.
S104:, will be presently described if presently described electromyography signal test data is not first electromyography signal test data
The corresponding difference value of electromyography signal test data, which is matched to obtain with preset threshold value, tests number with presently described electromyography signal
According to corresponding second test result.
In method provided in an embodiment of the present invention, the difference value characterize presently described electromyography signal test data with it is described
The difference degree of the previous electromyography signal test data of presently described electromyography signal test data.
S105: according to corresponding first test result of presently described electromyography signal test data and second test result
The ballot queue is updated, and votes presently described electromyography signal test data to obtain according to updated ballot queue
To recognition result.
In method provided in an embodiment of the present invention, after the ballot queue is updated, according to the ballot queue for storing
Corresponding first test result of one or more electromyography signal test datas vote, known according to voting results
Other result.
In method provided in an embodiment of the present invention, the recognition result is to identify that the electromyography signal test data is corresponding
Movement.
In data processing method provided in an embodiment of the present invention, number is tested by obtaining pre-stored multiple electromyography signals
According to, and according to the stripe sequence of each electromyography signal test data, it is successively that each electromyography signal test data is defeated
Enter to the object classifiers constructed in advance;When each electromyography signal test data is input to the object classifiers, obtain
The first test result corresponding with presently described electromyography signal test data is taken, and judges presently described electromyography signal test data
It whether is first electromyography signal test data;If presently described electromyography signal test data is first electromyography signal test data,
Ballot queue is then generated, and determines corresponding second test result of presently described electromyography signal test data;If presently described flesh
Electric signal test data is not first electromyography signal test data, then by the corresponding difference of presently described electromyography signal test data
Value and preset threshold value are matched to obtain the second test result corresponding with presently described electromyography signal test data;Foundation is worked as
Preceding corresponding first test result of the electromyography signal test data and second test result update the ballot queue, and
It votes presently described electromyography signal test data to obtain recognition result according to updated ballot queue, improves institute
State the accuracy of the recognition result of electromyography signal test data.
In data processing method provided in an embodiment of the present invention, the storing process of the electromyography signal test data can be with
Include:
Acquire the electromyography signal that user's execution movement generates;
Using sliding window setting technique, the time sequencing generated according to the electromyography signal divides the electromyography signal,
Multiple electromyography signal test datas are obtained, and each electromyography signal test data obtained to division successively stores.
In method provided in an embodiment of the present invention, user is acquired by electromyographic signal collection equipment and executes at least one movement
The electromyography signal of generation, user, which can be repeated several times, executes the same movement, and difference can be used when executing identical movement every time
Dynamics, it should be noted that pre-set the row of channels number of the electromyographic signal collection equipment, column port number and adjacent logical
Road spacing.
In method provided in an embodiment of the present invention, the movement can be the limb action of human or animal;Optionally, described
Movement is the gesture motion of human body.
The method that present example body provides, optionally, at least one movement that the movement executes at random for people, one
Act a corresponding electromyography signal.
In method provided in an embodiment of the present invention, executed what is collected individually with the user using sliding window setting technique
From starting to terminating, corresponding continuous electromyography signal is divided deliberate action;Window by presetting sliding window is long and sliding
Dynamic increment divides the continuous electromyography signal by the chronological order that the continuous electromyography signal generates and obtains multiple fleshes
Electric signal test data.
In data processing method provided in an embodiment of the present invention, optionally, before storing the electromyography signal test data,
Further include:
Electromyography signal test data to be stored is analyzed, is obtained corresponding with the electromyography signal test data to be stored
Temporal signatures;
According to temporal signatures building two-dimentional myoelectricity characteristic image corresponding with the electromyography signal test data.
In method provided in an embodiment of the present invention, by the electromyography signal test data by the sequence of division successively store to
In preset test data storage region, surveyed before storing the electromyography signal test data or storing the electromyography signal
After trying data, analyzes the electromyography signal test data and obtain temporal signatures;According to temporal signatures building and the flesh
The corresponding two-dimentional myoelectricity characteristic image of electric signal test data.
In method provided in an embodiment of the present invention, time domain spy is obtained by analyzing the single electromyography signal test data
Sign;The temporal signatures include multiple and electromyography signal test data feature, such as average absolute value MAV, slope sign
Change rate SSC, zero-crossing rate ZC and waveform length WL etc..
In method provided in an embodiment of the present invention, puts in order by acquisition device channels and test the single electromyography signal
The temporal signatures of data are integrated into two-dimentional myoelectricity characteristic image.
It is described by the corresponding difference value of presently described electromyography signal test data and preset threshold value matched to obtain with
Corresponding second test result of presently described electromyography signal test data, comprising:
Characterization and the current institute according to the corresponding two-dimentional myoelectricity characteristic image of presently described electromyography signal test data
The characterization of the corresponding two-dimentional myoelectricity characteristic image of previous electromyography signal test data of electromyography signal test data is stated, determination is worked as
The corresponding difference value of the preceding electromyography signal test data;
The difference value and preset threshold value are matched to obtain corresponding with presently described electromyography signal test data
Second test result.
Data processing method provided in an embodiment of the present invention, if choosing the electromyography signal test data of any action for knowing
Not, then the calculation method of the difference value is specific as follows:
K (t, x)=psnr3(S(t,x-1),S(t,x))
where(1≤t≤M,2≤x≤N(t))
Wherein, K (t, x) indicates the difference value of x-th of electromyography signal test data of t-th of deliberate action, S (t, x) table
Show that the two-dimentional myoelectricity characteristic image of x-th of electromyography signal test data of t-th of deliberate action, psnr are two-dimentional myoelectricity characteristic patterns
Y-PSNR as between, N (t) indicate the number of the electromyography signal test data of t-th of deliberate action, and M is deliberate action
Total number.
In method provided in an embodiment of the present invention, optionally, if choosing the corresponding any number of tests of any movement simultaneously
For identification, then the calculation method of the difference value is specific as follows for data:
K (x)=psnr3(S(x-1),S(x))
where(1≤x≤N)
Wherein, K (x) indicates the difference value of x-th of electromyography signal test data, and S (x) indicates x-th of electromyography signal test
The two-dimentional myoelectricity characteristic image of data, psnr are the Y-PSNRs between two-dimentional myoelectricity characteristic image, and N indicates electromyography signal test
The number of data.
Difference value calculating method provided in an embodiment of the present invention, there is stronger robustness, and calculating process is simple, time delay compared with
It is small, it is suitable for real-time myoelectric control, and equally applicable in the movement consecutive variations of no tranquillization.
In data processing method provided in an embodiment of the present invention, the building process of the object classifiers may include:
It acquires user and executes the electromyography signal that deliberate action generates;It is raw according to the electromyography signal using sliding window setting technique
At time sequencing the electromyography signal is divided, obtain multiple electromyography signal training datas, and obtain to division each
A electromyography signal training data is successively stored;By the electromyography signal test data by the sequential storage divided to pre-
If training data storage region in;It analyzes the electromyography signal training data and obtains training data temporal signatures;According to described in
Training data temporal signatures determine preliminary classification device;It is obtained according to the electromyography signal training data training preliminary classification device
Object classifiers.
The present invention is implemented in the method provided, the average absolute value that the training data temporal signatures of the electromyography signal include
The features such as MAV, slope sign change rate SSC, zero-crossing rate ZC and waveform length WL.
Training data temporal signatures packet in method provided in an embodiment of the present invention, according to the electromyography signal training data
The features such as the average absolute value MAV, slope sign change rate SSC, zero-crossing rate ZC and the waveform length WL that contain determine preliminary classification device
Type;The preliminary classification device can be logistic regression classifier, SVM classifier, K-NN adjacent to classifier and Bayes point
One or more classifiers such as class device.
In method provided in an embodiment of the present invention, the preliminary classification device is trained according to the electromyography signal training data,
The accuracy that the electromyography signal data of input are classified can be reached preset standard, be up to the preliminary classification device of standard
It is determined as object classifiers.
In data processing method provided in an embodiment of the present invention, if the presently described electromyography signal test data is first
Electromyography signal test data then generates ballot queue, and determines corresponding second test result of the electromyography signal test data,
Include:
If presently described electromyography signal test data is first electromyography signal test data, ballot queue is generated;
The state for determining presently described ballot queue, the state according to the presently described ballot queue determine presently described
Corresponding second test result of electromyography signal test data is non-action transfer point.
In method provided in an embodiment of the present invention, if the state of the presently described ballot queue is sky, it is determined that described
Corresponding second test result of electromyography signal test data is non-action transfer point.
In data method provided in an embodiment of the present invention, if presently described electromyography signal test data is first electromyography signal
Test data, then it is described according to corresponding first test result of presently described electromyography signal test data and it is described second test knot
Fruit updates the ballot queue, comprising:
When second test result is non-action transfer point, by presently described electromyography signal test data corresponding the
One test result is stored into the ballot queue, to realize the update of the ballot queue.
In data processing method provided in an embodiment of the present invention, the preset threshold value includes preset first threshold and pre-
If second threshold, it is described match with preset threshold value by the corresponding difference value of presently described electromyography signal test data
To second test result, may include:
If the difference value is less than preset first threshold, it is determined that presently described electromyography signal test data corresponding the
Two test results are movement transfer point;
If the difference value is greater than preset second threshold, it is determined that presently described electromyography signal test data corresponding the
Two test results are non-action transfer point;
If the difference value is more than or equal to preset first threshold and is less than or equal to preset second threshold, according to described in
Ballot queue obtains the voting results of the electromyography signal measured data;
If the voting results the first test result corresponding with presently described electromyography signal test data is consistent, it is determined that
Second test result of presently described electromyography signal test data is non-action transfer point;
If the voting results the first test result corresponding with presently described electromyography signal test data is inconsistent, really
Second test result of fixed presently described electromyography signal test data is movement transfer point.
If the state of the ballot queue is sky, it is determined that the second test result of presently described electromyography signal test data
For non-action transfer point.
It is described to obtain the electromyography signal measured data according to the ballot queue in method provided in an embodiment of the present invention
Voting results may include:
If the first test result corresponding with presently described electromyography signal test data of the ballot queue storage is consistent
The first test result of history quantity, be more than or equal to the first test result corresponding with presently described electromyography signal test data
The quantity of inconsistent the first test result of history, it is determined that the voting results and presently described electromyography signal test data pair
The first test result answered is consistent;
If the first test result corresponding with presently described electromyography signal test data of the ballot queue storage is consistent
The first test result of history data volume, be less than the first test result corresponding with presently described electromyography signal test data not
The quantity of consistent the first test result of history, it is determined that the voting results are corresponding with presently described electromyography signal test data
The first test result it is inconsistent.
It is described corresponding according to presently described electromyography signal test data in data processing method provided in an embodiment of the present invention
The first test result and second test result update the ballot queue, may include:
When corresponding second test result of presently described electromyography signal test data is non-action transfer point, by current institute
It states corresponding first test result of electromyography signal test data to store into the ballot queue, to realize the ballot queue
It updates;
It, will be described when corresponding second test result of presently described electromyography signal test data is movement transfer point
The quene state of ballot queue is set as empty, and corresponding first test result of presently described electromyography signal test data is stored
Extremely in the ballot queue, to realize the update of the ballot queue.
The present invention implements to be applied in the data processing method provided on the surface of the limb action generation of human or animal
The identification of electromyography signal is illustrated by taking the gesture motion for identifying people as an example below, specific as follows, may include:
Various gestures movement is chosen in advance, and the gesture motion includes thumb stretching, extension, and index finger stretches, and middle finger stretching, extension is small
Refer to stretching, extension, thumb and index finger stretch jointly, and index finger and middle finger stretch jointly, middle finger and nameless common stretching, extension, it is nameless and
Little finger of toe stretches jointly, and thumb and little finger of toe stretch jointly, and thumb, index finger and middle finger stretch jointly, middle finger, the third finger and small thumb
Refer to common stretching, extension, thumb, index finger, middle finger and nameless common stretching, extension, index finger, middle finger, the third finger and little finger of toe stretch jointly, greatly
Thumb, index finger, middle finger, the third finger and little finger of toe stretch totally 14 kinds of movements jointly.
The electrod-array of electromyographic signal collection equipment is placed in the position in movement muscle activation region, the electromyography signal
Acquiring equipment and choosing row of channels number is 8, and column port number is 6, and adjacency channel spacing is the flexible high-density electrode array of 14mm.
User executes above-mentioned 14 kinds of deliberate actions one by one, and in executing each action process, acquisition user executes each
The electromyography signal that a movement generates, wherein it chooses two different dynamics ranges and repeats a kind of movement, first
Dynamics is 70%~80% maximal voluntary contractile force, and the second dynamics is 30%~40% maximal voluntary contractile force, uses the respectively
One dynamics and the second dynamics execute each the movement 4 times, keep isometric contraction 6 seconds that dynamics is constant every time, are then kept for 4 seconds
Tranquillization, different action gaps give subject's sufficient time of having a rest.
It is divided using sliding window setting technique and executes the continuous electromyography signal that each deliberate action obtains by acquiring user, every time
It divides user and a kind of primary obtained continuous electromyography signal of deliberate action is executed by a kind of dynamics, wherein setting sliding window
A length of 128 milliseconds of window, sliding increment be set as 64 milliseconds;Continuous electromyography signal flow point is segmented into centainly by sliding window setting technique
The analysis window of quantity obtains user by a kind of dynamics and executes a kind of corresponding multiple sample datas of deliberate action;
Multi-feature extraction is carried out to be configured to two-dimentional flesh the electromyography signal in each channel in each described sample data
Electrical feature image.
The multiple features of extraction include temporal signatures, and the temporal signatures include average absolute value MAV, slope sign change rate
SSC, zero-crossing rate ZC and waveform length WL, the feature of each sample extraction is then put in order be integrated into two-dimentional flesh by channel
Electrical feature image, image a length of 8 is directly proportional to row of channels number, and image width is 6, color channel dimension directly proportional to column port number
It is identical as number of features.
Optionally, sample data is divided into electromyography signal training data and myoelectricity letter according to the first dynamics and the second dynamics
Number test data, if will execute the corresponding sample data of the movement by the first dynamics be determined as electromyography signal training data,
The corresponding sample data of the movement will be then executed by the second dynamics is determined as electromyography signal test data;If second will be passed through
Dynamics executes the corresponding sample data of the movement and is determined as electromyography signal training data, then will be by described in the execution of the first dynamics
It acts corresponding sample data and is determined as electromyography signal test data, by the electromyography signal training data and the electromyography signal
Test data is stored.
Temporal signatures according to sample data determine classifier, train the classifier according to electromyography signal training data,
The classifier for meeting preset requirement is determined as object classifiers, wherein the classifier for meeting preset requirement can be full
The classifier of the default discrimination of foot, is also possible to meet the classifier of default training burden.
According to the electromyography signal training data, first threshold and second threshold are set, wherein first threshold TH1, the
Two threshold values are TH1+TH2;Optionally, the value of TH1 is -1E5, and the value of TH2 is 0.35E5.
Obtain pre-stored multiple electromyography signal test datas, and drawing according to each electromyography signal test data
Divide sequence, each electromyography signal test data is successively input to the object classifiers constructed in advance.The myoelectricity is believed
Number test data is input to the process of the object classifiers, as shown in Fig. 2, specifically including:
S201: when each electromyography signal test data is input to the object classifiers, obtain with it is presently described
Corresponding first test result of electromyography signal test data.
During executing S201, first test result is the classifier according to the electromyography signal test data
Obtained initial recognition result.
S202: judging whether presently described electromyography signal test data is first electromyography signal test data, if so, executing
S203, if it is not, executing S204.
S203: generating ballot queue and determines corresponding second test result of the electromyography signal test data.
S204: it is matched the corresponding difference value of presently described electromyography signal test data to obtain institute with preset threshold value
State the second test result.
S205: judging whether second test result is movement transfer point, if it is not, S206 is then executed, if so, executing
S207。
S206: first test result is stored into the ballot queue, to realize the update of the ballot queue.
S207: it sets empty for the quene state of the ballot queue, and first test result is stored to described
In ballot queue, to realize the update of the ballot queue.
S208: it votes presently described electromyography signal test data to be identified according to updated ballot queue
As a result.
Based on the method that the embodiments of the present invention provide, user will be acquired and execute the continuous myoelectricity letter that index finger stretches
Number, 5 electromyography signal test datas are divided into using sliding window setting technique and are sequentially input to object classifiers, it is assumed that the 1st flesh
The recognition result of electric signal test data and the 2nd electromyography signal test data is all " index finger stretching, extension ", and the first test result
It is all " index finger stretching, extension " that is, there are two characterization " index finger stretching, extension " first test results in ballot queue, if the 3rd myoelectricity is believed
When number test data is input to object classifiers, the first obtained test result is " middle finger stretching, extension ", and the 3rd electromyography signal survey
The difference value for trying data is greater than first threshold, i.e., the second test result of the 3rd electromyography signal test data is non-action conversion
Point, and then the first test result of the 3rd electromyography signal test data is stored and updates ballot team into ballot queue to realize
Column, updated three the first test results of ballot queue for storing, two characterizations " index finger stretching, extension ", a characterization " stretch by middle finger
Exhibition ";Updated ballot queue votes to the 3rd electromyography signal test data, due in updated ballot queue
The the first test result quantity for characterizing " index finger stretching, extension " is greater than the first test result for characterizing " middle finger stretching, extension ", therefore finally obtains the
The recognition result of 3 electromyography signal test datas is " index finger stretching, extension ".
If the first test result of the 3rd electromyography signal test data is " index finger stretching, extension ", difference value is more than or equal to the
One threshold value and be less than or equal to second threshold;There are two characterization " index finger stretching, extension " history first test knots in current ballot queue
Fruit, therefore voting results are consistent with the 3rd the first test result of electromyography signal test data, are " index finger stretching, extension ", therefore sentence
Fixed 3rd electromyography signal test data is non-action transfer point, and the first test result of the 3rd electromyography signal test data is deposited
It puts into ballot queue to realize the update of ballot queue, and the 3rd electromyography signal is tested according to updated ballot queue
Data are voted, and the recognition result for obtaining the 3rd electromyography signal test data is " index finger stretching, extension ".
In method provided in an embodiment of the present invention, during being voted by ballot queue to determine recognition result, if described
The first test result of history of the characterization " index finger stretching, extension " for queue for storing of voting and the history first of characterization " middle finger stretching, extension " are surveyed
Test result is identical, then determines voting results or according to current electromyography signal test data according to the accuracy rate of history recognition result
First test result determines voting results.
In method provided in an embodiment of the present invention, optionally, voted according to ballot queue to determine that electromyography signal is tested
During second test result of data, if the history first of the characterization " index finger stretching, extension " of the ballot queue for storing is tested
As a result identical as the first test result of history of characterization " middle finger stretching, extension ", then first according to current electromyography signal test data is surveyed
Test result determination be for movement transfer point or non-action transfer point, if the first test result of current electromyography signal test data with
First test result of first electromyography signal test data is identical, then the second test result of current electromyography signal test data is
Non-action transfer point, if the first test result of current electromyography signal test data and first electromyography signal test data not phase
Together, then the second test result of current electromyography signal test data is movement transfer point.
Data processing method provided in an embodiment of the present invention, by determining whether the second test result is movement transfer point,
It determines that current electromyography signal test data acts in collection process whether to change, on this basis, judges current myoelectricity
Whether signal testing data correspond to the same movement with previous electromyography signal test data, and the embodiment of the present invention will be two neighboring
All continuous electromyography signal test datas between movement transfer point are regarded as same action classification, one myoelectricity letter of every test
Number test data, all by the recognition result of the electromyography signal test data and a upper movement transfer point to presently described myoelectricity
Corresponding first test result of all electromyography signal test datas between signal testing data carries out most ballots, obtains
The final recognition result of current electromyography signal test data, has strong robustness.
The derivatization process of above-mentioned each concrete implementation mode and each implementation, all falls in the scope of protection of the present invention.
Corresponding with method described in Fig. 1, the embodiment of the invention also provides a kind of data processing equipments, for Fig. 1
The specific implementation of middle method, data processing equipment provided in an embodiment of the present invention can be set with application computer terminal or various movements
In standby, structural schematic diagram is as shown in figure 3, specifically include:
Acquiring unit 301, for obtaining pre-stored multiple electromyography signal test datas, and according to each myoelectricity
Each electromyography signal test data is successively input to the target classification constructed in advance by the stripe sequence of signal testing data
Device;
Judging unit 302 when for each electromyography signal test data to be input to the object classifiers, obtains
The first test result corresponding with presently described electromyography signal test data, and judge that presently described electromyography signal test data is
No is first electromyography signal test data;The presently described electromyography signal test data is currently to be input to the target classification
The electromyography signal test data of device;
First determination unit 303, for being first electromyography signal test data when presently described electromyography signal test data
When, ballot queue is generated, and determine corresponding second test result of presently described electromyography signal test data, wherein the throwing
Ticket queue is for storing corresponding first test result of each electromyography signal test data;
Second determination unit 304, for not being first electromyography signal test number when presently described electromyography signal test data
According to when, the corresponding difference value of presently described electromyography signal test data is matched with preset threshold value obtain with it is presently described
Corresponding second test result of electromyography signal test data, wherein the difference value characterizes presently described electromyography signal and tests number
According to the difference degree with the previous electromyography signal test data;
Recognition unit 305, for according to corresponding first test result of presently described electromyography signal test data and described
Second test result updates the ballot queue, and according to updated ballot queue to presently described electromyography signal test data
It votes to obtain recognition result.
In data processing equipment provided in an embodiment of the present invention, processor is by obtaining pre-stored multiple electromyography signals
Test data, and according to the stripe sequence of each electromyography signal test data, successively each electromyography signal is tested
Data are input to the object classifiers constructed in advance;Each electromyography signal test data is input to the object classifiers
When, the first test result corresponding with presently described electromyography signal test data is obtained, and judge that presently described electromyography signal is surveyed
Try whether data are first electromyography signal test data;If presently described electromyography signal test data is first electromyography signal test
Data then generate ballot queue, and determine corresponding second test result of the electromyography signal test data;If presently described flesh
Electric signal test data is not first electromyography signal test data, then by the corresponding difference of presently described electromyography signal test data
Value and preset threshold value are matched to obtain the second test result corresponding with presently described electromyography signal test data;Foundation is worked as
Preceding corresponding first test result of the electromyography signal test data and second test result update the ballot queue, and
It votes presently described electromyography signal test data to obtain recognition result according to updated ballot queue, improves institute
State the accuracy of the recognition result of electromyography signal test data.
It is described before the storage of the electromyography signal test data in data processing equipment provided in an embodiment of the present invention
Acquiring unit may include:
First acquisition subelement, the electromyography signal generated for acquiring user's execution movement;
First storing sub-units, using sliding window setting technique, the time sequencing according to electromyography signal generation is to the flesh
Electric signal is divided, and obtains multiple electromyography signal test datas, and test number to each electromyography signal that division obtains
According to successively being stored.
It is described to obtain before storing the electromyography signal test data in data processing equipment provided in an embodiment of the present invention
Unit is taken, can also include:
First analysis subelement obtains and the myoelectricity to be stored for analyzing electromyography signal test data to be stored
The corresponding temporal signatures of signal testing data;
Subelement is constructed, according to temporal signatures building two-dimentional myoelectricity feature corresponding with the electromyography signal test data
Image.
It is described that presently described electromyography signal test data is corresponding in data processing equipment provided in an embodiment of the present invention
Difference value and preset threshold value are matched to obtain the second test result mistake corresponding with presently described electromyography signal test data
Cheng Zhong, second determination unit may include:
First determines subelement, for according to the corresponding two-dimentional myoelectricity characteristic image of presently described electromyography signal test data
Characterization and the presently described electromyography signal test data the corresponding two-dimentional myoelectricity of previous electromyography signal test data it is special
The characterization for levying image, determines the corresponding difference value of presently described electromyography signal test data;
First coupling subelement, for being matched to obtain and presently described myoelectricity the difference value with preset threshold value
Corresponding second test result of signal testing data.
In data processing equipment provided in an embodiment of the present invention, if the presently described electromyography signal test data is first
Electromyography signal test data then generates ballot queue, determines corresponding second test result of the electromyography signal test data
First determination unit, comprising:
Subelement is generated, for giving birth to when presently described electromyography signal test data is first electromyography signal test data
At ballot queue;
Third determines subelement, for determining the state of presently described ballot queue, according to the presently described ballot team
The state of column determines that corresponding second test result of presently described electromyography signal test data is non-action transfer point.
It is described corresponding according to presently described electromyography signal test data in data processing equipment provided in an embodiment of the present invention
The first test result and second test result update it is described ballot queue recognition unit, comprising:
First updates subelement, is used for when second test result is non-action transfer point, by presently described myoelectricity
Corresponding first test result of signal testing data is stored into the ballot queue, to realize the update of the ballot queue.
It is described that presently described electromyography signal test data is corresponding in data processing equipment provided in an embodiment of the present invention
Difference value is matched to obtain the second determination unit of second test result with preset threshold value, comprising:
4th determines subelement: for determining presently described myoelectricity when the difference value is less than preset first threshold
Corresponding second test result of signal testing data is movement transfer point;
5th determines subelement, for determining presently described myoelectricity when the difference value is greater than preset second threshold
Corresponding second test result of signal testing data is non-action transfer point;
6th determines subelement, for when the difference value is more than or equal to preset first threshold and is less than or equal to preset
When second threshold, the voting results of presently described electromyography signal measured data are obtained according to the ballot queue;
7th determines subelement, for when the voting results and presently described electromyography signal test data corresponding first
When test result is consistent, determine that the second test result of presently described electromyography signal test data is non-action transfer point;
8th determines subelement, for when the voting results and presently described electromyography signal test data corresponding first
Test result is inconsistent, it is determined that the second test result of presently described electromyography signal test data is movement transfer point.
It is described corresponding according to presently described electromyography signal test data in data processing equipment provided in an embodiment of the present invention
The first test result and second test result update it is described ballot queue recognition unit, comprising:
Second updates subelement, for being non-dynamic when corresponding second test result of presently described electromyography signal test data
When making transfer point, corresponding first test result of presently described electromyography signal test data is stored into the ballot queue,
To realize the update of the ballot queue;
Third updates subelement, for being when corresponding second test result of presently described electromyography signal test data
When acting transfer point, set empty for the quene state of the ballot queue, and by presently described electromyography signal test data pair
The first test result answered is stored into the ballot queue, to realize the update of the ballot queue.
The embodiment of the invention also provides a kind of storage medium, the storage medium includes the instruction of storage, wherein in institute
It states the equipment where controlling the storage medium when instruction operation and executes above-mentioned data processing method.
The embodiment of the invention also provides a kind of electronic equipment, structural schematic diagram is as shown in figure 4, specifically include memory
401 and one perhaps more than one 402 one of them or more than one instruction of instruction 402 be stored in memory 401
In, and be configured to by one or more than one processor 403 execute the one or more instruction 402 comprising use
In the instruction performed the following operation:
Obtain pre-stored multiple electromyography signal test datas, and drawing according to each electromyography signal test data
Divide sequence, each electromyography signal test data is successively input to the object classifiers constructed in advance;
When each electromyography signal test data is input to the object classifiers, obtains and believe with presently described myoelectricity
Number corresponding first test result of test data, and judge whether presently described electromyography signal test data is first electromyography signal
Test data;The presently described electromyography signal test data is currently to be input to the electromyography signal test of the object classifiers
Data;
If presently described electromyography signal test data is first electromyography signal test data, ballot queue is generated, and really
Determine corresponding second test result of presently described electromyography signal test data, wherein the ballot queue is for storing each institute
State corresponding first test result of electromyography signal test data;
If presently described electromyography signal test data is not first electromyography signal test data, presently described myoelectricity is believed
Number corresponding difference value of test data is matched to obtain corresponding with presently described electromyography signal test data with preset threshold value
The second test result, wherein the difference value characterizes presently described electromyography signal test data and the previous myoelectricity is believed
The difference degree of number test data;
It is updated according to corresponding first test result of presently described electromyography signal test data and second test result
The ballot queue, and vote presently described electromyography signal test data to be known according to updated ballot queue
Other result.
It should be noted that all the embodiments in this specification are described in a progressive manner, each embodiment weight
Point explanation is the difference from other embodiments, and the same or similar parts between the embodiments can be referred to each other.
For device class embodiment, since it is basically similar to the method embodiment, so being described relatively simple, related place ginseng
See the part explanation of embodiment of the method.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
For convenience of description, it is divided into various units when description apparatus above with function to describe respectively.Certainly, implementing this
The function of each unit can be realized in the same or multiple software and or hardware when invention.
As seen through the above description of the embodiments, those skilled in the art can be understood that the present invention can
It realizes by means of software and necessary general hardware platform.Based on this understanding, technical solution of the present invention essence
On in other words the part that contributes to existing technology can be embodied in the form of software products, the computer software product
It can store in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that a computer equipment
(can be personal computer, server or the network equipment etc.) executes the certain of each embodiment or embodiment of the invention
Method described in part.
A kind of data processing method provided by the present invention and device are described in detail above, it is used herein
A specific example illustrates the principle and implementation of the invention, and the above embodiments are only used to help understand originally
The method and its core concept of invention;At the same time, for those skilled in the art, according to the thought of the present invention, specific
There will be changes in embodiment and application range, in conclusion the content of the present specification should not be construed as to of the invention
Limitation.
Claims (10)
1. a kind of data processing method characterized by comprising
Pre-stored multiple electromyography signal test datas are obtained, and suitable according to the division of each electromyography signal test data
Each electromyography signal test data is successively input to the object classifiers constructed in advance by sequence;
When each electromyography signal test data is input to the object classifiers, obtains and surveyed with presently described electromyography signal
Corresponding first test result of data is tried, and judges whether presently described electromyography signal test data is first electromyography signal test
Data;The presently described electromyography signal test data is the electromyography signal test number for being currently input to the object classifiers
According to;
If presently described electromyography signal test data is first electromyography signal test data, ballot queue is generated, and determination is worked as
Corresponding second test result of the preceding electromyography signal test data, wherein the ballot queue is for storing each flesh
Corresponding first test result of electric signal test data;
If presently described electromyography signal test data is not first electromyography signal test data, presently described electromyography signal is surveyed
The examination corresponding difference value of data and preset threshold value are matched to obtain corresponding with presently described electromyography signal test data the
Two test results, wherein the difference value characterizes presently described electromyography signal test data and the previous electromyography signal is surveyed
Try the difference degree of data;
According to described in corresponding first test result of presently described electromyography signal test data and second test result update
Ballot queue, and vote presently described electromyography signal test data according to updated ballot queue to obtain identification knot
Fruit.
2. the method according to claim 1, wherein the storing process of the electromyography signal test data, comprising:
Acquire the electromyography signal that user's execution movement generates;
Using sliding window setting technique, the time sequencing generated according to the electromyography signal divides the electromyography signal, obtains
Multiple electromyography signal test datas, and each electromyography signal test data obtained to division successively stores.
3. according to the method described in claim 2, it is characterized in that, before storing the electromyography signal test data, further includes:
Electromyography signal test data to be stored is analyzed, time domain corresponding with the electromyography signal test data to be stored is obtained
Feature;
According to temporal signatures building two-dimentional myoelectricity characteristic image corresponding with the electromyography signal test data to be stored.
4. according to the method described in claim 3, it is characterized in that, described that presently described electromyography signal test data is corresponding
Difference value and preset threshold value are matched to obtain the second test result corresponding with presently described electromyography signal test data, packet
It includes:
Characterization and the presently described flesh according to the corresponding two-dimentional myoelectricity characteristic image of presently described electromyography signal test data
The characterization of the corresponding two-dimentional myoelectricity characteristic image of the previous electromyography signal test data of electric signal test data, determines current institute
State the corresponding difference value of electromyography signal test data;
The difference value and preset threshold value are matched to obtain and presently described electromyography signal test data corresponding second
Test result.
5. the method according to claim 1, wherein if the presently described electromyography signal test data is first
Electromyography signal test data then generates ballot queue, determines corresponding second test result of the electromyography signal test data, packet
It includes:
If presently described electromyography signal test data is first electromyography signal test data, ballot queue is generated;
The state for determining presently described ballot queue, the state according to the presently described ballot queue determine presently described myoelectricity
Corresponding second test result of signal testing data is non-action transfer point.
6. according to the method described in claim 5, it is characterized in that, described corresponding according to presently described electromyography signal test data
The first test result and second test result update the ballot queue, comprising:
When second test result is non-action transfer point, presently described electromyography signal test data corresponding first is surveyed
Test result is stored into the ballot queue, to realize the update of the ballot queue.
7. the method according to claim 1, wherein described that presently described electromyography signal test data is corresponding
Difference value is matched to obtain second test result with preset threshold value, comprising:
If the difference value is less than preset first threshold, it is determined that presently described electromyography signal test data corresponding second is surveyed
Test result is movement transfer point;
If the difference value is greater than preset second threshold, it is determined that presently described electromyography signal test data corresponding second is surveyed
Test result is non-action transfer point;
If the difference value is more than or equal to preset first threshold and is less than or equal to preset second threshold, according to the ballot
Queue obtains the voting results of presently described electromyography signal measured data;
If the voting results the first test result corresponding with presently described electromyography signal test data is consistent, it is determined that current
Second test result of the electromyography signal test data is non-action transfer point;
If the voting results the first test result corresponding with presently described electromyography signal test data is inconsistent, it is determined that when
Second test result of the preceding electromyography signal test data is movement transfer point.
8. the method according to the description of claim 7 is characterized in that described corresponding according to presently described electromyography signal test data
The first test result and second test result update the ballot queue, comprising:
When corresponding second test result of presently described electromyography signal test data is non-action transfer point, by presently described flesh
Corresponding first test result of electric signal test data is stored into the ballot queue, to realize the ballot queue more
Newly;
When corresponding second test result of presently described electromyography signal test data is movement transfer point, by the ballot
The quene state of queue is set as empty, and corresponding first test result of presently described electromyography signal test data is stored to institute
It states in ballot queue, to realize the update of the ballot queue.
9. a kind of data processing equipment characterized by comprising
Acquiring unit is surveyed for obtaining pre-stored multiple electromyography signal test datas, and according to each electromyography signal
The stripe sequence for trying data, is successively input to the object classifiers constructed in advance for each electromyography signal test data;
Judging unit when for each electromyography signal test data to be input to the object classifiers, obtains and current
Corresponding first test result of the electromyography signal test data, and judge presently described electromyography signal test data whether headed by
A electromyography signal test data;The presently described electromyography signal test data is the flesh for being currently input to the object classifiers
Electric signal test data;
First determination unit, for generating when presently described electromyography signal test data is first electromyography signal test data
Ballot queue, and determine corresponding second test result of presently described electromyography signal test data, wherein the ballot queue is used
In storing corresponding first test result of each electromyography signal test data;
Second determination unit, for inciting somebody to action when presently described electromyography signal test data is not first electromyography signal test data
The corresponding difference value of presently described electromyography signal test data is matched to obtain with preset threshold value and presently described myoelectricity is believed
Number corresponding second test result of test data, wherein the difference value characterizes presently described electromyography signal test data with before
The difference degree of one electromyography signal test data;
Recognition unit, for according to corresponding first test result of presently described electromyography signal test data and second test
As a result the ballot queue is updated, and is voted according to updated ballot queue presently described electromyography signal test data
To obtain recognition result.
10. a kind of electronic equipment, including memory and one perhaps one of them or one of more than one instruction with
Upper instruction is stored in memory, and is configured to be executed by one or more than one processor as claim 1~8 is any
Data processing method described in one.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910092534.4A CN109800733B (en) | 2019-01-30 | 2019-01-30 | Data processing method and device and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910092534.4A CN109800733B (en) | 2019-01-30 | 2019-01-30 | Data processing method and device and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109800733A true CN109800733A (en) | 2019-05-24 |
CN109800733B CN109800733B (en) | 2021-03-09 |
Family
ID=66560630
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910092534.4A Active CN109800733B (en) | 2019-01-30 | 2019-01-30 | Data processing method and device and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109800733B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110403609A (en) * | 2019-09-03 | 2019-11-05 | 北京海益同展信息科技有限公司 | Movement velocity analysis method, device and wearable device |
CN111401166A (en) * | 2020-03-06 | 2020-07-10 | 中国科学技术大学 | Robust gesture recognition method based on electromyographic information decoding |
CN113688802A (en) * | 2021-10-22 | 2021-11-23 | 季华实验室 | Gesture recognition method, device and equipment based on electromyographic signals and storage medium |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8747336B2 (en) * | 2005-10-16 | 2014-06-10 | Bao Tran | Personal emergency response (PER) system |
CN105022471A (en) * | 2014-04-23 | 2015-11-04 | 王建勤 | Device and method for carrying out gesture recognition based on pressure sensor array |
US20150366504A1 (en) * | 2014-06-20 | 2015-12-24 | Medibotics Llc | Electromyographic Clothing |
CN105654037A (en) * | 2015-12-21 | 2016-06-08 | 浙江大学 | Myoelectric signal gesture recognition method based on depth learning and feature images |
CN107273798A (en) * | 2017-05-11 | 2017-10-20 | 华南理工大学 | A kind of gesture identification method based on surface electromyogram signal |
CN107590432A (en) * | 2017-07-27 | 2018-01-16 | 北京联合大学 | A kind of gesture identification method based on circulating three-dimensional convolutional neural networks |
CN107679516A (en) * | 2017-10-25 | 2018-02-09 | 中国科学院合肥物质科学研究院 | Lower extremity movement recognition methods based on multiple dimensioned Gauss Markov random field model |
CN108268844A (en) * | 2018-01-17 | 2018-07-10 | 上海术理智能科技有限公司 | Movement recognition method and device based on surface electromyogram signal |
CN108491077A (en) * | 2018-03-19 | 2018-09-04 | 浙江大学 | A kind of surface electromyogram signal gesture identification method for convolutional neural networks of being divided and ruled based on multithread |
US20180307314A1 (en) * | 2012-06-14 | 2018-10-25 | Medibotics Llc | Circumferential Array of Electromyographic (EMG) Sensors |
CN108703824A (en) * | 2018-03-15 | 2018-10-26 | 哈工大机器人(合肥)国际创新研究院 | A kind of bionic hand control system and control method based on myoelectricity bracelet |
CN108983973A (en) * | 2018-07-03 | 2018-12-11 | 东南大学 | A kind of humanoid dexterous myoelectric prosthetic hand control method based on gesture identification |
CN109213305A (en) * | 2017-06-29 | 2019-01-15 | 沈阳新松机器人自动化股份有限公司 | A kind of gesture identification method based on surface electromyogram signal |
-
2019
- 2019-01-30 CN CN201910092534.4A patent/CN109800733B/en active Active
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8747336B2 (en) * | 2005-10-16 | 2014-06-10 | Bao Tran | Personal emergency response (PER) system |
US20180307314A1 (en) * | 2012-06-14 | 2018-10-25 | Medibotics Llc | Circumferential Array of Electromyographic (EMG) Sensors |
CN105022471A (en) * | 2014-04-23 | 2015-11-04 | 王建勤 | Device and method for carrying out gesture recognition based on pressure sensor array |
US20150366504A1 (en) * | 2014-06-20 | 2015-12-24 | Medibotics Llc | Electromyographic Clothing |
CN105654037A (en) * | 2015-12-21 | 2016-06-08 | 浙江大学 | Myoelectric signal gesture recognition method based on depth learning and feature images |
CN107273798A (en) * | 2017-05-11 | 2017-10-20 | 华南理工大学 | A kind of gesture identification method based on surface electromyogram signal |
CN109213305A (en) * | 2017-06-29 | 2019-01-15 | 沈阳新松机器人自动化股份有限公司 | A kind of gesture identification method based on surface electromyogram signal |
CN107590432A (en) * | 2017-07-27 | 2018-01-16 | 北京联合大学 | A kind of gesture identification method based on circulating three-dimensional convolutional neural networks |
CN107679516A (en) * | 2017-10-25 | 2018-02-09 | 中国科学院合肥物质科学研究院 | Lower extremity movement recognition methods based on multiple dimensioned Gauss Markov random field model |
CN108268844A (en) * | 2018-01-17 | 2018-07-10 | 上海术理智能科技有限公司 | Movement recognition method and device based on surface electromyogram signal |
CN108703824A (en) * | 2018-03-15 | 2018-10-26 | 哈工大机器人(合肥)国际创新研究院 | A kind of bionic hand control system and control method based on myoelectricity bracelet |
CN108491077A (en) * | 2018-03-19 | 2018-09-04 | 浙江大学 | A kind of surface electromyogram signal gesture identification method for convolutional neural networks of being divided and ruled based on multithread |
CN108983973A (en) * | 2018-07-03 | 2018-12-11 | 东南大学 | A kind of humanoid dexterous myoelectric prosthetic hand control method based on gesture identification |
Non-Patent Citations (2)
Title |
---|
YU DU: ""Semi-supervised learning for surface EMG-based gesture recognition"", 《INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE》 * |
陈香 等: ""一种基于融合多传感器信息的手语手势识别方法"", 《航天医学与医学工程》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110403609A (en) * | 2019-09-03 | 2019-11-05 | 北京海益同展信息科技有限公司 | Movement velocity analysis method, device and wearable device |
CN111401166A (en) * | 2020-03-06 | 2020-07-10 | 中国科学技术大学 | Robust gesture recognition method based on electromyographic information decoding |
CN113688802A (en) * | 2021-10-22 | 2021-11-23 | 季华实验室 | Gesture recognition method, device and equipment based on electromyographic signals and storage medium |
CN113688802B (en) * | 2021-10-22 | 2022-04-01 | 季华实验室 | Gesture recognition method, device and equipment based on electromyographic signals and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109800733B (en) | 2021-03-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wood et al. | On the variability of manual spike sorting | |
Verikas et al. | Electromyographic patterns during golf swing: Activation sequence profiling and prediction of shot effectiveness | |
Amma et al. | Advancing muscle-computer interfaces with high-density electromyography | |
Brodu et al. | Exploring two novel features for EEG-based brain–computer interfaces: Multifractal cumulants and predictive complexity | |
US20170143226A1 (en) | Action Recognition Method and Device Based on Surface Electromyography Signal | |
CN109800733A (en) | Data processing method and device, electronic equipment | |
EP3836836B1 (en) | Real-time spike detection and identification | |
Tot et al. | Software platform for learning about brain wave acquisition and analysis | |
CN108681685A (en) | A kind of body work intension recognizing method based on human body surface myoelectric signal | |
CN111048209A (en) | Health assessment method and device based on living body face recognition and storage medium thereof | |
Donovan et al. | Simple space-domain features for low-resolution sEMG pattern recognition | |
CN104679967B (en) | A kind of method for judging psychological test reliability | |
Xu et al. | Real-time finger force prediction via parallel convolutional neural networks: a preliminary study | |
Moore et al. | Depth sensor-based in-home daily activity recognition and assessment system for stroke rehabilitation | |
Ang et al. | Facial expression recognition through pattern analysis of facial muscle movements utilizing electromyogram sensors | |
Zhang et al. | Activity recognition of the torso based on surface electromyography for exoskeleton control | |
Castiglioni et al. | Assessing sample entropy of physiological signals by the norm component matrix algorithm: Application on muscular signals during isometric contraction | |
Liu et al. | Using deep learning for fusion of eye and mouse movement based user authentication | |
Carrino et al. | Gesture segmentation and recognition with an EMG-based intimate approach-an accuracy and usability study | |
Jaber et al. | Elicitation hybrid spatial features from HD-sEMG signals for robust classification of gestures in real-time | |
CN112957018A (en) | Heart state detection method and device based on artificial intelligence | |
CN114548165A (en) | Electromyographic mode classification method capable of crossing users | |
Iapăscurtă | Combining Algorithmic Information Dynamics Concepts and Machine Learning for Electroencephalography Analysis: What Can We Get? | |
Džaja et al. | Accelerometer-based algorithm for the segmentation and classification of repetitive human movements during workouts | |
Di Girolamo et al. | A comparison of sEMG temporal and spatial information in the analysis of continuous movements |
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 |