CN109276245B - Surface electromyogram signal feature processing and joint angle prediction method and system - Google Patents

Surface electromyogram signal feature processing and joint angle prediction method and system Download PDF

Info

Publication number
CN109276245B
CN109276245B CN201811292141.XA CN201811292141A CN109276245B CN 109276245 B CN109276245 B CN 109276245B CN 201811292141 A CN201811292141 A CN 201811292141A CN 109276245 B CN109276245 B CN 109276245B
Authority
CN
China
Prior art keywords
joint angle
characteristic vector
electromyographic
signal
prediction
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201811292141.XA
Other languages
Chinese (zh)
Other versions
CN109276245A (en
Inventor
周曦
罗洋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing Zhongke Yuncong Technology Co Ltd
Original Assignee
Chongqing Zhongke Yuncong Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing Zhongke Yuncong Technology Co Ltd filed Critical Chongqing Zhongke Yuncong Technology Co Ltd
Priority to CN201811292141.XA priority Critical patent/CN109276245B/en
Publication of CN109276245A publication Critical patent/CN109276245A/en
Application granted granted Critical
Publication of CN109276245B publication Critical patent/CN109276245B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/389—Electromyography [EMG]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
    • A61B5/1071—Measuring physical dimensions, e.g. size of the entire body or parts thereof measuring angles, e.g. using goniometers
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/45—For evaluating or diagnosing the musculoskeletal system or teeth
    • A61B5/4528—Joints
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Pathology (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Artificial Intelligence (AREA)
  • Signal Processing (AREA)
  • Physiology (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Dentistry (AREA)
  • Psychiatry (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Evolutionary Computation (AREA)
  • Orthopedic Medicine & Surgery (AREA)
  • Rheumatology (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The invention provides a surface electromyogram signal feature processing and joint angle prediction method and a system, wherein the method comprises the following steps: collecting electromyographic signals and joint angle signals; acquiring a first characteristic vector in a time domain according to the acquired electromyographic signals; resampling the characteristic vector and the collected joint angle signal to obtain a second characteristic vector; performing time difference compensation processing on the second characteristic vector to obtain processed electromyographic signal characteristics; training the processed electromyographic signal characteristics and the corresponding joint angles to obtain a prediction model, and predicting the joint angles in real time; the invention can enable the characteristic vector to fully reflect the motion information of the human body at the same time, improve the prediction precision, fully acquire the information contained in the electromyographic signal, reduce the data volume participating in training and quickly improve the prediction speed of the model.

Description

Surface electromyogram signal feature processing and joint angle prediction method and system
Technical Field
The invention relates to the field of electronics, in particular to a surface electromyogram signal feature processing and joint angle prediction method and system.
Background
The cooperative action between limbs is mainly performed by the central motor system sending commands to skeletal muscles, wherein α neurons in spinal cord can transmit motion signals to branch muscle fibers and contract due to stimulation, each α neuron and the fiber tissue to which the neuron belongs are called a motion unit, electromyographic signals are closely connected with α neurons, and the electromyographic signals are a series of time-series electric potentials generated on the outer layer of skin after passing through muscle fibers, skin and fat equivalent conductors.
However, the surface electromyographic signal is a weak and unstable sine-like signal, wherein the amplitude of the sine-like signal is mainly 0.01-100 mv, the frequency of the sine-like signal is distributed in 13-500Hz, and the electromyographic signal needs to be converted into the motion information of the human body by adopting a proper method. At present, joint angles are predicted based on electromyographic signals and mainly divided into two stages: a training phase and a testing phase of the model. In both stages, first, characteristic values are extracted in the time domain from a section of the raw electromyographic signal. And (3) making a feature value extracted in the model training stage and a joint angle in the most front moment in a feature window into a data set, and then training a machine learning model such as a support vector machine, a neural network, a linear regressor and the like to obtain a prediction model of the joint angle. And in the testing stage, the joint angle is directly predicted in real time through the extracted myoelectric characteristics. However, the following disadvantages exist in the feature processing procedure:
(1) the eigenvalues and joint angles are considered to be simultaneous. The joint angle and the electromyographic signal characteristic are not synchronous in real time due to factors such as electromyographic characteristic lag electromyographic signal, electromyographic delay and filtering, and the electromyographic signal characteristic value cannot reflect the movement information at a certain moment. Therefore, the time-asynchronous electromyographic signal characteristic value and joint angle training model is adopted, so that the prediction performance of the model is poor.
(2) And a large amount of data is adopted to train the least square support vector machine, so that the model is large and the prediction real-time performance is poor. In order to ensure that the model can sufficiently reflect the mapping relation between the electromyographic signals and the joint angles, the existing method usually adopts a large amount of data to train the model. However, for the least square support vector machine, the more training data is involved, the larger the model is, and the prediction speed of the model is affected.
Disclosure of Invention
In view of the above-mentioned shortcomings of the prior art, the present invention provides a method and system for surface electromyography characteristic processing and joint angle prediction to solve the above-mentioned problems.
The invention provides a surface electromyogram signal feature processing and joint angle prediction method, which comprises the following steps:
collecting electromyographic signals and joint angle signals;
acquiring a first characteristic vector in a time domain according to the acquired electromyographic signals;
resampling the characteristic vector and the collected joint angle signal to obtain a second characteristic vector;
performing time difference compensation processing on the second characteristic vector to obtain processed electromyographic signal characteristics;
training the processed electromyographic signal characteristics and the corresponding joint angles to obtain a prediction model,
and predicting the joint angle in real time according to the prediction model.
Further, the time difference compensation processing includes translating the second feature vector along the forward time direction on a time axis, where the translation amount is the time difference between the second feature vector and the joint angle.
Further, the translation amount is obtained by the following formula:
Tlag＝PΔt＝N/2Δt+Tfil-TEMD
wherein, TlagFor translation, Δ T is the sampling period, P is the number of sampling periods, N is the sampling window size, TfilIs electromyographic signal channelLag time of filtering, TEMDThe time when the electromyographic signal leads the joint angle is shown.
Further, the characteristic vector at least comprises one or a combination of absolute integral value, zero crossing point, ramp change, absolute standard deviation value and signal wavelength.
Further, normalization processing is carried out on the processed electromyographic signal characteristics, and a prediction model of the processed electromyographic signal characteristics and the corresponding joint angles is established, wherein the prediction model is a least square support vector machine model.
Further, the optimized objective function expression of the prediction model is as follows:
Figure BDA0001850192290000021
wherein U is weight of the support vector machine, phi is kernel function in the support vector machine, ekC is a penalty term coefficient for the allowable error between the fitted value and the actual value.
The invention also provides a surface electromyogram signal feature processing and joint angle prediction system, which comprises:
the collecting unit is used for collecting electromyographic signals and joint angle signals;
the processing unit is used for processing the acquired data; the processing unit acquires a first characteristic vector in a time domain according to the collected electromyographic signals; resampling the characteristic vector and the collected joint angle signal to obtain a second characteristic vector; performing time difference compensation processing on the second characteristic vector to obtain processed electromyographic signal characteristics;
and the prediction model is used for training the processed electromyographic signal characteristics and the corresponding joint angles and predicting the joint angles in real time.
Further, the collection unit comprises an electromyographic signal collection module and a joint angle collection module, the electromyographic signal collection module is arranged on the ulnar wrist flexor of the testee, and the joint angle collection module is arranged at the wrist joint of the testee.
The invention also provides a computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the method of any one of the above.
The present invention also provides an electronic terminal, comprising: a processor and a memory;
the memory is adapted to store a computer program and the processor is adapted to execute the computer program stored by the memory to cause the terminal to perform the method as defined in any one of the above.
The invention has the beneficial effects that: according to the method and the system for surface electromyogram signal feature processing and joint angle prediction, the joint angle and the feature vector are synchronized on a time axis by introducing the time difference compensation factor into the feature vector, so that the feature vector can fully reflect the motion information of a human body at the same time; in addition, the characteristic vector and the joint angle are resampled, so that the information contained in the electromyographic signal can be fully acquired, the data volume participating in training is reduced, the prediction speed of the model can be rapidly increased, the method is simple to debug, the model is easy to transplant to embedded equipment, and the applicability is strong.
Drawings
FIG. 1 is a schematic flow chart of a surface electromyogram signal feature processing and joint angle prediction method in an embodiment of the invention.
FIG. 2 is a schematic structural diagram of a surface electromyogram signal feature processing and joint angle prediction system in an embodiment of the present invention.
FIG. 3 is a schematic diagram of resampling in the surface electromyography signal feature processing and joint angle prediction method in the embodiment of the invention.
FIG. 4 is a schematic diagram of a process flow of feature processing in a training stage in the method for surface electromyography feature processing and joint angle prediction according to the embodiment of the present invention.
FIG. 5 is a schematic diagram of an algorithm flow of a surface electromyography signal feature processing and joint angle prediction method in an embodiment of the invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
In the following description, numerous details are set forth to provide a more thorough explanation of embodiments of the present invention, however, it will be apparent to one skilled in the art that embodiments of the present invention may be practiced in real time without these specific details, and in other embodiments, structures and devices are shown in block diagram form rather than in detail in order to avoid obscuring the embodiments of the present invention.
As shown in fig. 1, the method for surface electromyogram signal feature processing and joint angle prediction in this embodiment includes:
collecting electromyographic signals and joint angle signals;
acquiring a first characteristic vector in a time domain according to the acquired electromyographic signals;
resampling the characteristic vector and the collected joint angle signal to obtain a second characteristic vector;
performing time difference compensation processing on the second characteristic vector to obtain processed electromyographic signal characteristics;
training the processed electromyographic signal characteristics and the corresponding joint angles to obtain a prediction model,
and predicting the joint angle in real time according to the prediction model.
In this embodiment, as shown in fig. 2, first, the raw electromyographic signal x of the flexor carpi ulnaris at time k Δ t is collectedkAnd collecting the joint angle ymea(k) The sampling frequency is 1KHz, then a feature vector F (k) is extracted from the electromyographic signals in a time domain, and then an absolute integral value IEMG (k), a zero crossing point ZC (k), a slope change SSC (k), an absolute standard deviation DASDV (k) and a signal wavelength WL (k) are calculated in the window size N;
the size of the window N in this embodiment is about (300- & 600).
Each time domain feature in f (k) is represented by the following formulas (1) to (5):
the absolute value integral IEMG expression is:
Figure BDA0001850192290000041
wherein xkRaw electromyographic signal expressed as k Δ t
The zero crossing ZC expression is:
Figure BDA0001850192290000042
the ramping SSC expression is:
Figure BDA0001850192290000043
the absolute standard deviation value DASDV expression is:
Figure BDA0001850192290000051
the signal wavelength WL expression is:
Figure BDA0001850192290000052
in the embodiment, the characteristics F (k) of the electromyographic signal segment which is long enough for a period of time and the measured joint angle ymea (k) are resampled, as shown in fig. 3, the frequency band distribution of the electromyographic signal in the embodiment is 13-500Hz, the sampling frequency is 1000Hz, and the characteristic vector resampling period T is (5-10) sampling periods Deltat; in the feature extraction in the above steps, the electromyographic feature calculation is calculated based on the sampling frequency of 1000Hz, sufficient motion information is retained in the feature vector, and the above processing procedure is shown in fig. 4.
In the present embodiment, time difference compensation is introduced in the resampled features f (k), i.e. the f (k) vector is right-shifted by T on the time axislagObtaining an F (k-P) electromyographic signal feature vector in P delta T units, wherein a delay factor TlagThe size expression is shown as formula (6):
Tlag＝PΔt＝N/2Δt+Tfil-TEMDformula (6)
In this embodiment, after the processed electromyographic signal characteristics are obtained, a line 0-1 normalization method is adopted to map the characteristic values into the distribution of [0-1 ]. The formula for 0-1 normalization is shown in equation (7):
Figure BDA0001850192290000053
wherein, FiIs a time sequence of the ith characteristic of the electromyographic signal characteristic vector F; fi(k) Is represented by FiThe value at time k, where k is 1 … … N.
In the embodiment, the processed electromyographic signal feature vector F (k-P) corresponds to the joint angle ymea(k) Training a least square support vector machine model; the least squares support vector machine parameter derivation is shown in figure 5.
The present embodiment adopts the structure risk minimization principle according to the LSSVM, and the optimization objective function is expressed as equation (8), and the equation is constrained as equation (9).
Figure BDA0001850192290000054
y(k)＝UTΦ(F(k))+b+ekFormula (9)
Wherein U is the weight of the support vector machine; phi is a kernel function in the support vector machine; e.g. of the typekIs the allowable error between the fitted value and the actual value; and C is a penalty term coefficient.
The Lagrange multiplier method is adopted to process the formula as follows:
Figure BDA0001850192290000061
wherein, αiLagrange multiplier coefficients.
The bias derivation of U, phi, ek and b is respectively calculated under the KKT condition as shown in the formula:
Figure BDA0001850192290000063
Figure BDA0001850192290000064
Figure BDA0001850192290000065
and finally, the derivation is generalized to a linear equation shown in the formula (15), and the parameters of the LSSVM model can be obtained by solving the formula (15).
Wherein, 1v ═ 1,1, …,1] TM × 1, α ═ α 1, α 2, …, α M ] T, Ω ij ═ Φ (f (i)), Φ (f (i)) ═ K [ (f (i)), f (j)) ], Y ═ Y (1), …, Y (K)), …, Y (M)) ] T, where K is expressed as gaussian kernel function, and its expression is shown as follows:
Figure BDA0001850192290000067
can be obtained by solving linear equationsTo parameter b, αiThen the final equation for predicting the joint angle from the electromyographic signal characteristics is as follows:
Figure BDA0001850192290000068
in this embodiment, the bandwidth δ of the gaussian kernel function and the penalty term coefficient C are hyper-parameters of two handedness equations, and cannot be obtained by solving through a support vector machine. The parameters of each measured object are determined respectively, and are summarized as shown in table 1:
Figure BDA0001850192290000069
Figure BDA0001850192290000071
TABLE 1
In the testing stage, the characteristics F (k) are extracted by collecting electromyographic signals, the characteristics F (k) are directly input into a trained least square method support vector machine model as shown in a formula (16), and the joint angle is calculated in real time.
In another embodiment, the difference is that an RBF neural network is used as a prediction model, and the Root Mean Square Error (RMSE) and the Correlation Coefficient (CC) are used as evaluation indexes, and the expressions are respectively as follows:
Figure BDA0001850192290000072
wherein the content of the first and second substances,
Figure BDA0001850192290000073
expressed as the actual predicted angle; y is expressed as a predicted joint angle; m represents the number of calculation points.
Figure BDA0001850192290000074
Wherein Cov represents y and
Figure BDA0001850192290000075
calculating covariance; d represents y and
Figure BDA0001850192290000076
the respective variance.
The overall prediction accuracy and speed result of the algorithm in the embodiment are shown in table 2, the model prediction RMSE subjected to electromyographic signal characteristic processing is greatly reduced, and the CC value is improved, which indicates that the prediction accuracy is improved. In the aspect of training speed, the speed of the LSSVM after feature processing is increased by 380 times, and the speed of the RBF neural network is increased by 5000 times; in the aspect of predicting the speed, the speed of the LSSVM after the characteristic processing is improved by 32 times, and the speed of the RBF neural network is improved by 25 times; the prediction speed of the LSSVM for feature processing is fastest.
Figure BDA0001850192290000077
TABLE 2
Correspondingly, a surface electromyogram signal characteristic processing and joint angle prediction system is also provided, which comprises:
the collecting unit is used for collecting electromyographic signals and joint angle signals;
the processing unit is used for processing the acquired data; the processing unit acquires a first characteristic vector in a time domain according to the collected electromyographic signals; resampling the characteristic vector and the collected joint angle signal to obtain a second characteristic vector; performing time difference compensation processing on the second characteristic vector to obtain processed electromyographic signal characteristics;
and the prediction model is used for training the processed electromyographic signal characteristics and the corresponding joint angles and predicting the joint angles in real time.
The collecting unit in this embodiment includes an electromyographic signal collecting module 2 and a joint angle collecting module 1, the electromyographic signal collecting module 2 is arranged in the ulnar wrist flexor of the testee, the joint angle collecting module 1 is arranged in the wrist joint of the testee, the electromyographic signal collecting module 2 in this embodiment is a collecting module based on a dry electrode, the ulnar wrist flexor is arranged in the dry electrode, a maxon encoder is adopted for the joint angle, the joint angle is installed at the joint of two connecting rods, the two connecting rods are tightly tied up with the hand and the forearm respectively, as shown in fig. 2, the collecting time in this embodiment is about 10s and about 10000 points, and finally the single chip microcomputer 3 sends data to the computer 4 through a USB.
In the embodiment, the direction of the finger in the hand of the object to be tested and the forearm should keep a straight line, and then the wrist moves in flexion and extension at the speed of 0.35rad/s to 4rad/s, in order to ensure the quality of the electromyographic signal, the embodiment adopts the dry electrode, and the dry electrode is more convenient to use, does not need to replace the electrode slice for many times, and is more convenient to use; compared with a disposable gel wet electrode, the cost is lower, and the signal-to-noise ratio of the electromyographic signals detected by the electrode is high through experiments; the tested object should keep a relaxed state and keep a correct sitting posture; the experimental subject needs to rest for about 2 minutes in every two experimental intervals to relieve muscle fatigue and tension, and preferably, a plurality of adults without any history of neuromuscular diseases are selected as the experimental subject, and each subject carries out five experiments.
The present embodiment also provides a computer-readable storage medium on which a computer program is stored, which when executed by a processor implements any of the methods in the present embodiments.
The present embodiment further provides an electronic terminal, including: a processor and a memory;
the memory is used for storing computer programs, and the processor is used for executing the computer programs stored by the memory so as to enable the terminal to execute the method in the embodiment.
The computer-readable storage medium in the present embodiment can be understood by those skilled in the art as follows: all or part of the steps for implementing the above method embodiments may be performed by hardware associated with a computer program. The aforementioned computer program may be stored in a computer readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
The electronic terminal provided by the embodiment comprises a processor, a memory, a transceiver and a communication interface, wherein the memory and the communication interface are connected with the processor and the transceiver and are used for completing mutual communication, the memory is used for storing a computer program, the communication interface is used for carrying out communication, and the processor and the transceiver are used for operating the computer program so that the electronic terminal can execute the steps of the method.
In this embodiment, the Memory may include a Random Access Memory (RAM), and may also include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory.
The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
Note that in the corresponding figures of embodiments, where signals are represented by lines, some lines are thicker, to indicate more constituent signal paths (constituent signal paths) and/or one or more ends of some lines have arrows, to indicate primary information flow direction, these designations are not intended to be limiting, and indeed, the use of such lines in connection with one or more example embodiments facilitates easier circuit or logic unit routing, and any represented signal (as determined by design requirements or preferences) may actually comprise one or more signals that may be conveyed in either direction and may be implemented in any suitable type of signal scheme.
Unless otherwise specified the use of the ordinal adjectives "first", "second", etc., to describe a common object, merely indicate that different instances of like objects are being referred to, and are not intended to imply that the objects so described must be in a given sequence, either temporally, spatially, in ranking, or in any other manner.
Reference in the specification to "an embodiment," "one embodiment," "some embodiments," or "other embodiments" means that a particular feature, structure, or characteristic described in connection with the embodiments is included in at least some embodiments, but not necessarily all embodiments. The various appearances of "an embodiment," "one embodiment," or "some embodiments" are not necessarily all referring to the same embodiments. If the specification states a component, feature, structure, or characteristic "may", "might", or "could" be included, that particular component, feature, structure, or characteristic is not necessarily included. If the specification or claim refers to "a" or "an" element, that does not mean there is only one of the element. If the specification or claim refers to "a further" element, that does not preclude there being more than one of the further element.
While the present invention has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art in light of the foregoing description. For example, other memory structures (e.g., dynamic ram (dram)) may use the discussed embodiments. The embodiments of the invention are intended to embrace all such alternatives, modifications and variances that fall within the broad scope of the appended claims.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.

Claims (10)

1. A surface electromyogram signal feature processing and joint angle prediction method is characterized by comprising the following steps:
collecting electromyographic signals and joint angle signals;
acquiring a first characteristic vector in a time domain according to the acquired electromyographic signals;
resampling the first characteristic vector and the collected joint angle signal to obtain a second characteristic vector;
performing time difference compensation processing on the second characteristic vector to obtain processed electromyographic signal characteristics;
training the processed electromyographic signal characteristics and the corresponding joint angles to obtain a prediction model,
and predicting the joint angle in real time according to the prediction model.
2. The method for surface electromyogram signal feature processing and joint angle prediction according to claim 1, wherein: and the time difference compensation processing comprises the step of translating the second characteristic vector along the positive time direction on a time axis, wherein the translation amount is the time difference between the second characteristic vector and the joint angle.
3. The method for surface electromyogram signal feature processing and joint angle prediction according to claim 2, wherein the translation amount is obtained by the following formula:
Tlag＝PΔt＝N/2Δt+Tfil-TEMD
wherein, TlagFor translation, Δ T is the sampling period, P is the number of sampling periods, N is the sampling window size, TfilLag time, T, filtered for electromyographic signalsEMDThe time when the electromyographic signal leads the joint angle is shown.
4. The method for surface electromyogram signal feature processing and joint angle prediction according to claim 1, wherein: the first characteristic vector at least comprises one or a combination of several of absolute integral value, zero crossing point, ramp change, absolute standard deviation value and signal wavelength.
5. The method for surface electromyographic signal feature processing and joint angle prediction according to claim 1, wherein the processed electromyographic signal features are normalized to establish a prediction model of the processed electromyographic signal features and corresponding joint angles, and the prediction model is a least squares support vector machine model.
6. The method for surface electromyogram signal feature processing and joint angle prediction according to claim 5, wherein: the optimized objective function expression of the prediction model is as follows:
Figure FDA0002261850930000011
wherein U is the weight of the support vector machine, ekAs the allowable error between the fitted value and the actual value, C is the coefficient of the penalty term, UTIs the transpose of U.
7. A surface electromyogram signal feature processing and joint angle prediction system is characterized by comprising:
the collecting unit is used for collecting electromyographic signals and joint angle signals;
the processing unit is used for processing the acquired data; the processing unit acquires a first characteristic vector in a time domain according to the collected electromyographic signals; resampling the first characteristic vector and the collected joint angle signal to obtain a second characteristic vector; performing time difference compensation processing on the second characteristic vector to obtain processed electromyographic signal characteristics;
and the prediction model is used for training the processed electromyographic signal characteristics and the corresponding joint angles and predicting the joint angles in real time.
8. The system for surface electromyogram signal feature processing and joint angle prediction according to claim 7, wherein: the collecting unit comprises an electromyographic signal collecting module and a joint angle collecting module, the electromyographic signal collecting module is arranged on the ulnar wrist flexor of the testee, and the joint angle collecting module is arranged at the wrist joint of the testee.
9. A computer-readable storage medium having stored thereon a computer program, characterized in that: the program when executed by a processor implements the method of any one of claims 1 to 6.
10. An electronic terminal, comprising: a processor and a memory;
the memory is for storing a computer program and the processor is for executing the computer program stored by the memory to cause the terminal to perform the method of any of claims 1 to 6.
CN201811292141.XA 2018-11-01 2018-11-01 Surface electromyogram signal feature processing and joint angle prediction method and system Active CN109276245B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811292141.XA CN109276245B (en) 2018-11-01 2018-11-01 Surface electromyogram signal feature processing and joint angle prediction method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811292141.XA CN109276245B (en) 2018-11-01 2018-11-01 Surface electromyogram signal feature processing and joint angle prediction method and system

Publications (2)

Publication Number Publication Date
CN109276245A CN109276245A (en) 2019-01-29
CN109276245B true CN109276245B (en) 2020-02-07

Family

ID=65174277

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811292141.XA Active CN109276245B (en) 2018-11-01 2018-11-01 Surface electromyogram signal feature processing and joint angle prediction method and system

Country Status (1)

Country Link
CN (1) CN109276245B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109965847B (en) * 2019-04-08 2023-11-07 清华大学 Server and signal analysis system
CN109984763A (en) * 2019-04-10 2019-07-09 福州大学 Method based on artificial neural network intelligent predicting human synovial torque
CN111079927B (en) * 2019-12-12 2022-07-08 福州大学 Patella pain detection system based on extreme learning machine
CN116619422A (en) * 2023-06-21 2023-08-22 哈尔滨理工大学 Motion capture and wrist joint angle estimation mechanical arm mapping method based on human body characteristic signals

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102799937A (en) * 2012-06-26 2012-11-28 天津大学 Lower limb movement track predication method under fusion of information of myoelectricity signal and joint angle
WO2018022602A1 (en) * 2016-07-25 2018-02-01 Ctrl-Labs Corporation Methods and apparatus for predicting musculo-skeletal position information using wearable autonomous sensors
CN107856014A (en) * 2017-11-08 2018-03-30 浙江工业大学 Mechanical arm posture control method based on gesture identification
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

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3003226A2 (en) * 2013-06-03 2016-04-13 The Regents of The University of Colorado, A Body Corporate Postural control of a multi-function prosthesis
US11000211B2 (en) * 2016-07-25 2021-05-11 Facebook Technologies, Llc Adaptive system for deriving control signals from measurements of neuromuscular activity

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102799937A (en) * 2012-06-26 2012-11-28 天津大学 Lower limb movement track predication method under fusion of information of myoelectricity signal and joint angle
WO2018022602A1 (en) * 2016-07-25 2018-02-01 Ctrl-Labs Corporation Methods and apparatus for predicting musculo-skeletal position information using wearable autonomous sensors
CN107856014A (en) * 2017-11-08 2018-03-30 浙江工业大学 Mechanical arm posture control method based on gesture identification
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

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
A Novel Method for Elbow Joint Continuous Prediction using EMG and Musculoskeletal Model;Muye Pang et al;《International Conference on Robotics and Biomimetics(ROBIO)》;20131130;第1240-1245页 *
Continuous estimation of finger joint angles under different static wrist motions from surface EMG signals;Lizhi Pan et al;《Biomedical Signal Processing and Control》;20140907;第265-271页 *
Continuous estimation of joint angle from electromyography using multiple time-delayed features and random forests;Feiyun Xiao et al;《Biomedical Signal Processing and Control》;20170830;第303-311页 *
Estimation of Lower Limb Periodic Motions from sEMG Using Least Squares Support Vector Regression;Q.L.Li et al;《Neural Process Lett》;20141026;第371-388页 *
Surface EMG based continuous estimation of human lower limb joint angles by using deep belief networks;Jiangcheng Chen et al;《Biomedical Signal Processing and Control》;20171016;第335-342页 *
原空间最小二乘支持向量机;梁锦锦 等;《计算机工程与设计》;20140731;第35卷(第7期);第2541-2544页、第2611页 *
基于鲁棒最小二乘支持向量机的齿轮磨损预测;衷路生 等;《北京工业大学学报》;20140731;第40卷(第7期);第1028-1034页、第1047页 *

Also Published As

Publication number Publication date
CN109276245A (en) 2019-01-29

Similar Documents

Publication Publication Date Title
CN109276245B (en) Surface electromyogram signal feature processing and joint angle prediction method and system
CN107378944B (en) Multidimensional surface electromyographic signal artificial hand control method based on principal component analysis method
CN109262618B (en) Muscle cooperation-based upper limb multi-joint synchronous proportional myoelectric control method and system
Narayan SEMG signal classification using KNN classifier with FD and TFD features
Asghar et al. Review on electromyography based intention for upper limb control using pattern recognition for human-machine interaction
Kumar et al. Towards identification of finger flexions using single channel surface electromyography–able bodied and amputee subjects
JP5252432B2 (en) Finger joint angle estimation device
Tang et al. Continuous estimation of human upper limb joint angles by using PSO-LSTM model
CN114533086A (en) Motor imagery electroencephalogram decoding method based on spatial domain characteristic time-frequency transformation
CN113990441A (en) Lower limb knee joint active muscle myoelectric fitting method based on biodynamics
CN110353704A (en) Mood assessments method and apparatus based on wearable ECG monitoring
Wang et al. Prediction of sEMG-based tremor joint angle using the RBF neural network
Zhang et al. Real‐time and user‐independent feature classification of forearm using EMG signals
CN105796091B (en) A kind of intelligent terminal for removing electrocardiosignal vehicle movement noise
Anantamek et al. Recognition of yoga poses using EMG signals from lower limb muscles
CN111110268B (en) Human body muscle sound signal prediction method based on random vector function connection network technology
Silva-Acosta et al. Automatic estimation of continuous elbow flexion–extension movement based on electromyographic and electroencephalographic signals
CN114159080B (en) Training and recognition method and device for upper limb rehabilitation robot movement intention recognition model
CN106983511B (en) Method and device for identifying muscle strength and muscle tension state mutation points
Zhang et al. Human motor function estimation based on EMG signal fractal dimension standard deviation
CN110633644A (en) Human body joint angle prediction method based on electromyographic wavelet packet decomposition and GABP
Bai et al. Intelligent prosthetic arm force control based on sEMG analysis and BPNN classifier
Gao et al. sEMG-angle estimation using feature engineering techniques for least square support vector machine
CN113520413B (en) Lower limb multi-joint angle estimation method based on surface electromyogram signals
Lu et al. A comparison of contributions of individual muscle and combination muscles to interaction force prediction using KPCA-DRSN model

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
CB02 Change of applicant information
CB02 Change of applicant information

Address after: 401122 5 stories, Block 106, West Jinkai Avenue, Yubei District, Chongqing

Applicant after: Chongqing Zhongke Yuncong Technology Co., Ltd.

Address before: 401122 5 stories, Block 106, West Jinkai Avenue, Yubei District, Chongqing

Applicant before: CHONGQING ZHONGKE YUNCONG TECHNOLOGY CO., LTD.

GR01 Patent grant
GR01 Patent grant