CN109213305A - A kind of gesture identification method based on surface electromyogram signal - Google Patents
A kind of gesture identification method based on surface electromyogram signal Download PDFInfo
- Publication number
- CN109213305A CN109213305A CN201710518604.9A CN201710518604A CN109213305A CN 109213305 A CN109213305 A CN 109213305A CN 201710518604 A CN201710518604 A CN 201710518604A CN 109213305 A CN109213305 A CN 109213305A
- Authority
- CN
- China
- Prior art keywords
- electromyogram signal
- surface electromyogram
- feature extraction
- kinds
- gesture identification
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J19/00—Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/017—Gesture based interaction, e.g. based on a set of recognized hand gestures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention discloses a kind of gesture identification methods based on surface electromyogram signal, utilize sensor collection surface electromyography signal data;Feature extraction is carried out using seven kinds of feature extraction combinational algorithms to surface electromyogram signal data;To data after carrying out feature extraction using three kinds of different classifiers progress comparison of classification;Calculate the bat of the classification results of every group of surface electromyogram signal data, using the combination of feature extraction value-based algorithm and pattern recognition classifier device corresponding to maximum value in the bat of classification results as optimum combination, determine which kind of gesture the surface electromyogram signal belongs to.The embodiment of the present invention is classified using seven kinds of feature extraction combinational algorithms and three kinds of classifiers, to effectively improve the swiftness and stability of gesture identification, applied to can bring better user experience in Dextrous Hand control.
Description
Technical field
The present invention relates to a kind of gesture identification method for providing and optimizing based on surface electromyogram signal is related to, can be applied to take
The Dextrous Hand control field of business robot, the motion intention of the sensor real-time judge user by being mounted on forearm, control spirit
Dab hand is moved according to the intention of user.
Background technique
Surface electromyogram signal (Surface Electromyography, abbreviation SEMG) is one kind and neuron-muscular activity phase
The bioelectrical signals of pass.When movement instruction is conducted via central nervous system to related muscle fibre, muscle fibre can be caused to power on
Position changes the contraction of concurrent myogenic fiber, superposition of the potential change at the skin surface on time of origin and space and form table
Facial muscle electric signal can be acquired by surface myoelectric electrode.Surface electromyogram signal contains the mode of contraction of muscle and shrinks strong
Information, the different limb actions such as degree correspond to different electromyography signals, can determine this by analyzing surface electromyogram signal
Specific action mode corresponding to signal.
Surface electromyogram signal is a kind of non-linear bio signal, and with weak output signal, structure is complicated, non-stationary, robust
Property difference feature.Since it has non-implantable and easy availability, surface electromyogram signal can be used as input signal used in a variety of people
In machine interaction (human-computer interation, HCI) equipment, such as pass through the control of gesture identification progress Dextrous Hand.
The surface electromyogram signal controller of standard is typically based on threshold value and the amplitude of signal to realize identification (such as palm to simple action
Opening and close).It realizes and multiple surface electrodes is needed to accurately identifying for different gestures, increase signal dimension and complexity
Degree, and influence the processing of surface electromyogram signal.
Currently, the research of gesture identification is concentrated mainly on three Signal Pretreatment, feature extraction and pattern-recognition fields.Letter
Number pretreatment strategy is usually filtered to the sample of acquisition and dimensionality reduction, such as principal component analysis and improves signal-to-noise ratio.Feature extraction
Strategy is to be calculated the signal data of extraction to extract feature vector from the time domain of signal or frequency domain, such as being averaged for signal
Amplitude (MAV) and slope sign variation number (SSC) etc., and the classification to feature vector is realized with the methods of linear discriminant device.It is right
Method for distinguishing, including K closest (KNN), artificial neural network (ANN), support vector machines (SVM) are known in muscle electrical signal pattern
Deng.But it is directed to three above field, how to select optimal feature extracting method is still huge problem.
Therefore it needs to provide a kind of gesture identification method of optimization based on surface electromyogram signal, to improve the fast of identification
Speed and stability judge that user movement is intended in real time, improve user experience.
Summary of the invention
The problems of gesture identification is carried out for based on surface electromyogram signal, the present invention proposes a kind of using seven kinds of spies
Sign extracts combinational algorithm and three kinds of classifiers carry out the matched method based on surface electromyogram signal identification gesture.The side of this method
Case is as follows:
A kind of gesture identification method based on surface electromyogram signal, including step 1: sensor collection surface myoelectricity is utilized
Signal data;Step 2: feature extraction is carried out using seven kinds of feature extraction combinational algorithms to the surface electromyogram signal data;Step
Rapid three: to data after carrying out feature extraction using three kinds of different classifiers progress comparison of classification;Step 4: every group of surface is calculated
The bat of the classification results of electromyography signal data, by feature corresponding to maximum value in the bat of classification results
The combination of extraction algorithm and pattern recognition classifier device determines which kind of gesture the surface electromyogram signal belongs to as optimum combination.
Preferably, seven kinds of feature extraction combinational algorithms are based on three kinds of Willison amplitude, root mean square and variance spies
Extraction algorithm is levied as combination foundation.
Preferably, seven kinds of feature extraction combinational algorithms are respectively algorithm root mean square-Willison amplitude, root mean square-
Willison amplitude-variance, root mean square, root mean square-variance, Willison amplitude, variance-Willison amplitude and variance.
Preferably, the calculation formula of the Willison amplitude arithmetic is as follows, wherein W be sampling window width,
xiFor in i-th of sample of sampling window:
Preferably, the threshold value in Willison amplitude arithmetic formula is 75 μ V.
Preferably, the calculation formula of the root mean square algorithm is as follows, and wherein W is the surface electromyogram signal sampling of acquisition
Points, XiFor the range value of i-th of surface electromyogram signal sampled point:
Preferably, the calculation formula of the variance algorithm is as follows, and wherein W is the surface electromyogram signal sampled point of acquisition
Number, xiFor the range value of i-th of surface electromyogram signal sampled point, M is mean value:
Preferably, three kinds of different classifiers include the closest value of K, artificial neural network and support vector machines.
Preferably, in step 1 surface myoelectric data sample collected include the acquisition of four sensors five kinds of gestures,
And every kind of gesture repeats ten times.
Preferably, five kinds of gestures include clench fist, index finger stretches out that other fingers close, index finger and middle finger stretch out other hands
Finger closes, index finger and thumb close the stretching of other fingers, the five fingers opening.
As can be seen from the above technical solutions, the embodiment of the present invention has the advantage that
Gesture identification method of the embodiment of the present invention based on surface electromyogram signal, it is worthwhile by using seven kinds of feature extraction groups
Method and three kinds of classifiers are classified, to effectively improve the swiftness and stability of gesture identification, are applied to dexterous manual
System, and then accurately judge the intention of user, effectively improve the experience of user.Seven kinds of features in the embodiment of the present invention mention
Taking combinational algorithm is temporal signatures extraction algorithm, can make to calculate simplicity, effectively realize real-time control.The embodiment of the present invention
Optimum combination is chosen by the way of permutation and combination as control algolithm, and there is pure science, improve the accuracy of identification.
Detailed description of the invention
Fig. 1 is that a kind of schematic diagram of electromyography signal line system model is provided in the embodiment of the present invention;
Fig. 2 is that a kind of flow diagram based on surface electromyogram signal identification gesture method is provided in the embodiment of the present invention;
Fig. 3 is five kinds of gesture motion schematic diagrames of offer in the embodiment of the present invention;
Fig. 4 is the raw-data map that the embodiment of the present invention acquires five kinds of gestures of Fig. 3 using four surface myoelectric sensors.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people
The model that the present invention protects all should belong in member's every other embodiment obtained without making creative work
It encloses.
Description and claims of this specification and term " first ", " second ", " third " " in above-mentioned attached drawing
The (if present)s such as four " are to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should manage
The data that solution uses in this way are interchangeable under appropriate circumstances, so that the embodiments described herein can be in addition to illustrating herein
Or the sequence other than the content of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that
Cover it is non-exclusive include, for example, containing the process, method, system, product or equipment of a series of steps or units need not limit
In step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, produce
The other step or units of product or equipment inherently.
Surface electromyogram signal is a kind of non-linear bio signal, and with weak output signal, structure is complicated, non-stationary, robust
Property difference feature.Utilize sensor collection surface electromyography signal, it usually needs mathematical modeling is carried out to surface electromyogram signal.Fig. 1
It is that a kind of schematic diagram of electromyography signal line system model is provided in the embodiment of the present invention.Since collected surface electromyogram signal is
A kind of random signal has very strong uncertainty, and simple mathematical function relationship can not be to its accurate description.In the embodiment
The middle mathematical feature that surface electromyogram signal is estimated according to statistical method.The mathematical model of surface electromyogram signal and its processing method are tight
Close association.The present embodiment establishes the mathematics of surface electromyogram signal from macroscopic perspective according to the Biological Principles of surface electromyogram signal
Model;The electric pulse that nervous centralis is issued is as the input signal of system, and each pulse is mutually indepedent, and probability having the same
Distribution;Skeleton muscle group is integrally equivalent to transmission function G, establishes the mathematical model of surface electromyogram signal.
Fig. 2 is a kind of flow diagram that gesture method is identified based on surface electromyogram signal that one embodiment of the invention provides.
In this embodiment, gesture identification method includes 4 steps in total.
Step S1: sensor collection surface electromyography signal data are utilized.
In a preferred embodiment, surface myoelectric data collected include that 4 sensors acquire five kinds of gestures, and every kind
Gesture repeats ten times.Specific five kinds of gestures as shown in figure 3, include CLOSE: clench fist, ONE: index finger stretch out other fingers close,
TWO: index finger and middle finger stretch out other fingers close, THREE: index finger and thumb close other fingers stretch out, OPEN: the five fingers
It opens.Wherein 4 sensors correspond to five kinds of gestures collected surface electromyogram signal initial data it is as shown in Figure 4.Fig. 4 by
20 collecting sample compositions show respectively 5 kinds of differences and act corresponding surface electromyogram signal;Wherein, the horizontal axis table of every part of figure
Show the time, unit is millisecond (ms);The longitudinal axis indicates voltage magnitude, and unit is millivolt (mV).
5 kinds of gestures shown in Fig. 3 are denoted as one group, every group of each gesture motion is repeated 10 times each duration at random
2s forms 10 groups of data.In experiment, 5 parts are divided by 10 groups, wherein 4 parts are used to train, portion is used to test.
The various movements of hand are each skeletal muscle synergistic effects as a result, skeletal muscle required for different gesture motion
Type and shrinkage degree difference.In gesture change procedure, the shrinkage degree of forearm skeletal muscle is the function of actuation time, but one
For a little skeletal muscle in addition also with wrist towards related, mathematical feature is complex, therefore is extracting surface electromyogram signal feature
When need to gesture motion carry out it is following two specification: (1) skeletal muscle is most short to the process of predetermined action state by relaxation state
Not less than 1.5s;(2) each gesture motion returns relaxation state after keeping 2s.
Above-mentioned two regulation purposes are preferably to extract surface electromyogram signal characteristic value, and also to avoid different
The influence of body otherness.Bradypragia or the too fast temporal signatures that can all influence surface electromyogram signal.Meeting above-mentioned item regulation
Afterwards, extracted feature vector is of less demanding to calculation amount, may be implemented to extract validity feature value with the less time.
Step S2: feature extraction is carried out using seven kinds of feature extraction combinational algorithms to surface electromyogram signal data.Wherein, seven
Kind feature extraction combinational algorithm is calculated based on three kinds of Willison amplitude (WA), root mean square (RMS) and variance (VR) feature extractions
Method is as combination foundation.It is respectively root mean square-by 7 kinds of algorithms that three of the above feature extraction algorithm permutation and combination goes out
Willison amplitude (RMS-WA), root mean square-Willison amplitude-variance (RMS-WA-VAR), root mean square (RMS), root mean square-
Variance (RMS-VAR), Willison amplitude (WA), variance-Willison amplitude (VAR-WA) and variance (VAR).
Willison amplitude (WA) can calculate the change frequency of surface electromyogram signal amplitude, when amplitude variation is more than pre-
Determine effective when threshold value.Willison amplitude (WA) feature extraction being mixed into avoid outside noise, the contraction of it and skeletal muscle
Degree is related.The expression of Willison amplitude (WA) is as shown in formula 1:
In the formula, W is width, the x of sampling windowiFor in i-th of sample of sampling window.Philipson et al.
Research think threshold value to be chosen for 50-100 μ V the most suitable.The threshold value that the present invention chooses is 75 μ V.
Root mean square (RMS) is a kind of analysis method for being widely used in surface electromyogram signal feature extraction.When gesture motion
Between do not exceed 2s generally, therefore it is fine every the second moment stationarity of middle electromyography signal in the short time.Surface electromyogram signal is with flat
Steady signal is handled, and the root-mean-square value of signal can reflect the amplitude Characteristics of random signal very well, root mean square (RMS) is specifically counted
Operator expression formula is as shown in formula 2:
Wherein, W is the surface electromyogram signal sampling number of acquisition, XiFor the amplitude of i-th of surface electromyogram signal sampled point
Value.Root mean square reflects the virtual value of surface electromyogram signal to a certain extent, mainly shows each muscle group in limb action mistake
The size of dynamics is contributed in journey.
Variance (VAR) indicates the degree that surface electromyogram signal and average value deviate, and has reacted skeletal action process surface
The variation range and severe degree of electromyography signal amplitude.The specific calculation expression of variance (VAR) is as shown in formula 3:
Wherein W is the surface electromyogram signal sampling number of acquisition, xiFor the amplitude of i-th of surface electromyogram signal sampled point
Value, M is mean value.
Step S3: to data after carrying out feature extraction using three kinds of different classifiers progress comparison of classification, three kinds of differences
Classifier include the closest value of K (KNN), artificial neural network (ANN) and support vector machines (SVM).
K nearest neighbor algorithm (KNN) is a kind of sorting algorithm of Case-based Reasoning, and basic thought is: using a kind of metric calculation
The distance between sample to be sorted and all training samples find the K neighbour nearest apart from sample to be sorted, then according to this
Classification belonging to K neighbour carries out the classification that most ballots determine sample to be sorted.
Artificial neural network (ANN) includes a variety of neural network algorithms, uses BP neural network algorithm in the present embodiment.
BP neural network algorithm is learnt using error backpropagation algorithm, is successfully used for function approximation, pattern-recognition and data
The fields such as excavation.Part is improved based on the supervised learning that gradient declines using BP neural network algorithm in pattern-recognition to search
Without hesitation can, have good learning ability.
Support vector machines (SVM) is being developed by the optimal classification surface of linear classification problem.Support vector machines (SVM)
Basic principle is that sample space is mapped in a high-dimensional feature space by a Nonlinear Mapping p, so that in original sample
The problem of the problem of Nonlinear separability, is converted into the linear separability in feature space in this space.Supported in the embodiment of the present invention to
The kernel function that amount machine (SVM) is selected is gaussian radial basis function.
Step S4: calculating the bat of the classification results of every group of surface electromyogram signal data, by the flat of classification results
As optimum combination, determining should for the combination of feature extraction algorithm and pattern recognition classifier device corresponding to maximum value in equal accuracy
Which kind of gesture surface electromyogram signal belongs to.
It will be carried out using the collected surface electromyogram signal data of sensor by seven kinds of different feature extracting methods special
Sign is extracted and carries out comparison of classification by three kinds of different classifications devices.In conjunction with above-mentioned preferred embodiment mode, each group of emulation
As a result it is all stored in one 7 × 3 matrix.Wherein, the numerical value of the i-th row jth column indicates to select ith feature vector sum the
The bat of classification results after j classifier.In order to be assessed classification results, the embodiment of the present invention defines one
Assess parameter FS.Due to being formula (4) there are the expression of 10 groups of experiments and 3 different classifications devices, appraisal procedure
It is shown, take FSiMaximum value be FMAX, shown in the expression of FMAX such as formula (5).
FMAX=maxiFSi(formula 5)
The combination of feature extraction algorithm and pattern recognition classifier device corresponding to parameter FS maximum value will be assessed as optimal
Combination, so that it is determined which kind of gesture the surface electromyogram signal belongs to.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
It can be combined in any suitable manner in one or more embodiment or examples.In addition, without conflicting with each other, the skill of this field
Art personnel can tie the feature of different embodiments or examples described in this specification and different embodiments or examples
It closes and combines.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (10)
1. a kind of gesture identification method based on surface electromyogram signal, which is characterized in that comprising steps of
Step 1: sensor collection surface electromyography signal data are utilized;
Step 2: feature extraction is carried out using seven kinds of feature extraction combinational algorithms to the surface electromyogram signal data;
Step 3: to data after carrying out feature extraction using three kinds of different classifiers progress comparison of classification;
Step 4: calculating the bat of the classification results of every group of surface electromyogram signal data, by the average standard of classification results
The combination of feature extraction algorithm corresponding to maximum value and pattern recognition classifier device determines surface flesh as optimum combination in exactness
Which kind of gesture electric signal belongs to.
2. a kind of gesture identification method based on surface electromyogram signal according to claim 1, which is characterized in that described seven
Kind feature extraction combinational algorithm is based on three kinds of Willison amplitude, root mean square and variance feature extraction algorithms as combination base
Plinth.
3. a kind of gesture identification method based on surface electromyogram signal according to claim 2, which is characterized in that described seven
Kind of feature extraction combinational algorithm is respectively algorithm root mean square-Willison amplitude, root mean square-Willison amplitude-variance, square
Root, root mean square-variance, Willison amplitude, variance-Willison amplitude and variance.
4. a kind of gesture identification method based on surface electromyogram signal according to claim 2, which is characterized in that described
The calculation formula of Willison amplitude arithmetic is as follows, and wherein W is width, the x of sampling windowiFor the i-th of sampling window
A sample:
5. a kind of gesture identification method based on surface electromyogram signal according to claim 4, which is characterized in that
The threshold value in Willison amplitude arithmetic formula is 75 μ V.
6. a kind of gesture identification method based on surface electromyogram signal according to claim 2, which is characterized in that described equal
The calculation formula of root algorithm is as follows, and wherein W is the surface electromyogram signal sampling number of acquisition, XiFor i-th of surface flesh
The range value of electric signal sampled point:
7. a kind of gesture identification method based on surface electromyogram signal according to claim 2, which is characterized in that the side
The calculation formula of difference algorithm is as follows, and wherein W is the surface electromyogram signal sampling number of acquisition, xiFor i-th of surface myoelectric
The range value of signal sampling point, M is mean value:
8. a kind of gesture identification method based on surface electromyogram signal according to claim 1, which is characterized in that described three
The different classifier of kind includes the closest value of K, artificial neural network and support vector machines.
9. a kind of gesture identification method based on surface electromyogram signal according to claim 1, which is characterized in that step 1
In surface myoelectric data collected include that four sensors acquire five kinds of gestures, and every kind of gesture repeats ten times.
10. a kind of gesture identification method based on surface electromyogram signal according to claim 9, which is characterized in that described
Five kinds of gestures include clench fist, index finger stretch out other fingers close, index finger and middle finger stretch out other fingers close, index finger and thumb it is tight
Close the stretching of other fingers, the five fingers open.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710518604.9A CN109213305A (en) | 2017-06-29 | 2017-06-29 | A kind of gesture identification method based on surface electromyogram signal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710518604.9A CN109213305A (en) | 2017-06-29 | 2017-06-29 | A kind of gesture identification method based on surface electromyogram signal |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109213305A true CN109213305A (en) | 2019-01-15 |
Family
ID=64960802
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710518604.9A Withdrawn CN109213305A (en) | 2017-06-29 | 2017-06-29 | A kind of gesture identification method based on surface electromyogram signal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109213305A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109800733A (en) * | 2019-01-30 | 2019-05-24 | 中国科学技术大学 | Data processing method and device, electronic equipment |
CN110443309A (en) * | 2019-08-07 | 2019-11-12 | 浙江大学 | A kind of electromyography signal gesture identification method of combination cross-module state association relation model |
CN110473603A (en) * | 2019-07-09 | 2019-11-19 | 大连理工大学 | A kind of body-building householder method based on electromyography signal |
CN110826625A (en) * | 2019-11-06 | 2020-02-21 | 南昌大学 | Finger gesture classification method based on surface electromyographic signals |
CN111046731A (en) * | 2019-11-11 | 2020-04-21 | 中国科学院计算技术研究所 | Transfer learning method and recognition method for gesture recognition based on surface electromyogram signals |
CN111209885A (en) * | 2020-01-13 | 2020-05-29 | 腾讯科技(深圳)有限公司 | Gesture information processing method and device, electronic equipment and storage medium |
CN112894882A (en) * | 2020-12-30 | 2021-06-04 | 哈尔滨工业大学芜湖机器人产业技术研究院 | Robot fault detection system based on industrial internet |
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 (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102622605A (en) * | 2012-02-17 | 2012-08-01 | 国电科学技术研究院 | Surface electromyogram signal feature extraction and action pattern recognition method |
CN103440498A (en) * | 2013-08-20 | 2013-12-11 | 华南理工大学 | Surface electromyogram signal identification method based on LDA algorithm |
US20140198944A1 (en) * | 2013-01-14 | 2014-07-17 | Qualcomm Incorporated | Use of emg for subtle gesture recognition on surfaces |
-
2017
- 2017-06-29 CN CN201710518604.9A patent/CN109213305A/en not_active Withdrawn
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102622605A (en) * | 2012-02-17 | 2012-08-01 | 国电科学技术研究院 | Surface electromyogram signal feature extraction and action pattern recognition method |
US20140198944A1 (en) * | 2013-01-14 | 2014-07-17 | Qualcomm Incorporated | Use of emg for subtle gesture recognition on surfaces |
CN103440498A (en) * | 2013-08-20 | 2013-12-11 | 华南理工大学 | Surface electromyogram signal identification method based on LDA algorithm |
Non-Patent Citations (1)
Title |
---|
胡俊: "基于表面肌电信号的最优化手势特征识别分类方法", 《第十二届沈阳科学学术年会论文集(理工农医)》 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109800733B (en) * | 2019-01-30 | 2021-03-09 | 中国科学技术大学 | Data processing method and device and electronic equipment |
CN109800733A (en) * | 2019-01-30 | 2019-05-24 | 中国科学技术大学 | Data processing method and device, electronic equipment |
CN110473603B (en) * | 2019-07-09 | 2022-04-12 | 大连理工大学 | Body-building assisting method based on electromyographic signals |
CN110473603A (en) * | 2019-07-09 | 2019-11-19 | 大连理工大学 | A kind of body-building householder method based on electromyography signal |
CN110443309A (en) * | 2019-08-07 | 2019-11-12 | 浙江大学 | A kind of electromyography signal gesture identification method of combination cross-module state association relation model |
CN110826625A (en) * | 2019-11-06 | 2020-02-21 | 南昌大学 | Finger gesture classification method based on surface electromyographic signals |
CN110826625B (en) * | 2019-11-06 | 2022-04-12 | 南昌大学 | Finger gesture classification method based on surface electromyographic signals |
CN111046731A (en) * | 2019-11-11 | 2020-04-21 | 中国科学院计算技术研究所 | Transfer learning method and recognition method for gesture recognition based on surface electromyogram signals |
CN111046731B (en) * | 2019-11-11 | 2023-07-25 | 中国科学院计算技术研究所 | Transfer learning method and recognition method for gesture recognition based on surface electromyographic signals |
CN111209885A (en) * | 2020-01-13 | 2020-05-29 | 腾讯科技(深圳)有限公司 | Gesture information processing method and device, electronic equipment and storage medium |
CN111209885B (en) * | 2020-01-13 | 2023-05-30 | 腾讯科技(深圳)有限公司 | Gesture information processing method and device, electronic equipment and storage medium |
US11755121B2 (en) | 2020-01-13 | 2023-09-12 | Tencent Technology (Shenzhen) Company Limited | Gesture information processing method and apparatus, electronic device, and storage medium |
CN112894882A (en) * | 2020-12-30 | 2021-06-04 | 哈尔滨工业大学芜湖机器人产业技术研究院 | Robot fault detection system based on industrial internet |
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 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109213305A (en) | A kind of gesture identification method based on surface electromyogram signal | |
Sapsanis et al. | EMG based classification of basic hand movements based on time-frequency features | |
Karnam et al. | EMGHandNet: A hybrid CNN and Bi-LSTM architecture for hand activity classification using surface EMG signals | |
Shen et al. | Movements classification of multi-channel sEMG based on CNN and stacking ensemble learning | |
Shenoy et al. | Online electromyographic control of a robotic prosthesis | |
CN106293057A (en) | Gesture identification method based on BP neutral net | |
Baldacchino et al. | Simultaneous force regression and movement classification of fingers via surface EMG within a unified Bayesian framework | |
CN106383579A (en) | EMG and FSR-based refined gesture recognition system and method | |
Jaramillo et al. | Real-time hand gesture recognition with EMG using machine learning | |
CN110710970B (en) | Method and device for recognizing limb actions, computer equipment and storage medium | |
Neacsu et al. | Automatic EMG-based hand gesture recognition system using time-domain descriptors and fully-connected neural networks | |
CN102961203A (en) | Method for identifying surface electromyography (sEMG) on basis of empirical mode decomposition (EMD) sample entropy | |
CN103440498A (en) | Surface electromyogram signal identification method based on LDA algorithm | |
CN110413107A (en) | Bionic mechanical hand interaction control method based on electromyography signal pattern-recognition and particle group optimizing | |
CN111844032B (en) | Electromyographic signal processing and exoskeleton robot control method and device | |
Yang et al. | sEMG-based continuous hand gesture recognition using GMM-HMM and threshold model | |
Li et al. | Transfer learning-based muscle activity decoding scheme by low-frequency sEMG for wearable low-cost application | |
BAKIRCIOĞLU et al. | Classification of EMG signals using convolution neural network | |
Yu et al. | Attenuating the impact of limb position on surface EMG pattern recognition using a mixed-LDA classifier | |
Zhou et al. | sEMG-based hand motion recognition by means of multi-class adaboost algorithm | |
Lin et al. | A normalisation approach improves the performance of inter-subject sEMG-based hand gesture recognition with a ConvNet | |
CN111923048A (en) | Electromyographic signal classification and exoskeleton robot control method and device | |
Kurzynski et al. | Multiple classifier system applied to the control of bioprosthetic hand based on recognition of multimodal biosignals | |
Wang et al. | Multi-finger myoelectric signals for controlling a virtual robotic prosthetic hand | |
Teng et al. | Evaluation of synergy-based hand gesture recognition method against force variation for robust myoelectric control |
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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190115 |
|
WW01 | Invention patent application withdrawn after publication |