CN108903948A - A kind of human body muscle signals analysis system - Google Patents
A kind of human body muscle signals analysis system Download PDFInfo
- Publication number
- CN108903948A CN108903948A CN201810518942.7A CN201810518942A CN108903948A CN 108903948 A CN108903948 A CN 108903948A CN 201810518942 A CN201810518942 A CN 201810518942A CN 108903948 A CN108903948 A CN 108903948A
- Authority
- CN
- China
- Prior art keywords
- muscle
- muscle signals
- analysis
- signals
- mmg
- 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- Withdrawn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
-
- 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
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1123—Discriminating type of movement, e.g. walking or running
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/22—Ergometry; Measuring muscular strength or the force of a muscular blow
- A61B5/224—Measuring muscular strength
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physiology (AREA)
- Dentistry (AREA)
- Physical Education & Sports Medicine (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The present invention devises a set of muscle signals analysis system, it is related to human action mode identification technology.The device is mainly made of muscle signals acquisition sensor, muscle signals collection plate, muscle signals filter circuit, muscle signals processing module and muscle signals analysis module.The system can extract fast and reliablely human body muscle signals, in conjunction with technologies such as filtering, signal segmentation and normalization, time frequency analysis, machine learning, complete MMG signal processing and feature extraction, realize human action pattern-recognition, muscle strength and muscular fatigue evaluation and test, it is simple with structure, feature easy to use, low in cost.
Description
Technical field
The present invention relates to muscle signals analysis fields, and in particular to a kind of human action mode based on muscle signals feature
Identification, muscle strength analysis and muscular fatigue analysis and research method and system design.
Background technique
During manufacturing artificial limb and bionic application for disabled person, people are used for constantly looking for optimal signal source
Correlative study work, such as:Electromyography signal (Electromyography, EMG), EEG signals
(Electroencephalography, MMG), sound signal source etc. are a kind of new during nearest signal source is found
Signal source appears in researcher's eye:Muscle signals (Mechanomyography, MMG), muscle signals are that human body is doing difference
When movement, parts of body muscle vibrates " sound " signal that a kind of human ear issued is not heard.
For muscle analysis, at present both at home and abroad using most still EMG signals, as a kind of relatively early research muscular movement
The signal source of situation, EMG signal can show that the dielectric displacement in human muscle's motion process changes well, however, EMG signal
There is also obvious disadvantages while being widely used property:Weak output signal, general EMG acquisition big vulnerable to interference, acquisition difficulty
Costly and cumbersome acquisition equipment is needed, collecting flowchart is not also smooth.
The short slab of EMG signal studies MMG signal to be accelerated as EMG signal, MMG
Signal is also a kind of signal source for human muscle's characteristic, it is worth mentioning at this point that, the acquisition equipment of MMG signal is light, acquisition
Process is succinct, at low cost, the extensive advantage of application prospect, and the interference that MMG signal is subject to is smaller, meanwhile, at home and abroad
In MMG research, there is the achievement of some substances.
A possibility that these prospects are brought, be based on MMG signal carry out human action pattern-recognition, muscle strength analysis and
Where the power of muscular fatigue analysis.
Summary of the invention
The purpose of the present invention is design it is a set of based on muscle signals carry out human action pattern-recognition, muscle strength analysis and
The multifunctional unit system of muscular fatigue analysis, its utilitarian design is convenient cheap, by extracting the feature of muscle signals, data
Comprehensively and accuracy is high.
To solve the problems of background technique, the present invention uses following technical scheme:The system mainly includes
Muscle signals acquire sensor 1, control box 2 and host computer procedure three parts, wherein control box 2 by muscle signals collection plate 3,
Muscle signals filter circuit 4 and power supply 7 form, and host computer procedure is divided into muscle signals processing module 5 and muscle signals analysis mould
Block 6 is installed on computer or tablet computer end;Muscle signals acquire one end of sensor 1 and one end of muscle signals collection plate 3
Connection, and the other end of muscle signals collection plate 3 is connected to the input terminal of 4 module of muscle signals filter circuit, muscle signals filter
The output end of wave circuit 4 by USB patchcord or bluetooth connection to computer or tablet computer end, by be installed on computer or
The muscle signals processing module 5 and muscle signals analysis module 6 at tablet computer end are completed to MMG signal processing, human action mould
Formula identification, muscle strength analysis and muscular fatigue analysis.
The operation principle of the present invention is that the white patch for acquiring sensor 1 is against in tested after the completion of system assembles
At the tested muscle of person;When subject makes different movements, the collected muscle signals of sensor are acquired by muscle signals
Plate 3 is converted into digital signal, and since there are certain interference signals for original MMG signal, therefore, it is necessary to be filtered by muscle signals
Circuit 4 is filtered MMG signal, and then by USB patchcord or Bluetooth communication, collected MMG signal is transmitted
To computer or tablet computer end, it is based on muscle signals processing module 5 and muscle signals analysis module 6, carries out human action mould
Formula identification, muscle strength analysis and muscular fatigue analysis;MMG signal processing module 5 uses signal adaptive Length discrepancy segmentation side
Method accurately analyzes each movement MMG signal, obtains frequency spectrum and frequecy characteristic;MMG signal analysis module 6 uses support vector machines
(SVM), caffee learning method carry out human action pattern-recognition;It is in positive based on contraction of muscle strength and muscle signals amplitude
Principle is closed, by carrying out time frequency analysis and feature extraction to MMG signal, the peak torque of muscle, mean power is analyzed and bends and stretches
The muscle strengths index such as flesh ratio;Drop to initial value 70% with Maximum isometric voluntary contraction power MVC as muscular fatigue judgment basis, is based on
The mapping relations of MMG signal averaging frequency and amplitude and muscle Maximum isometric voluntary contraction power carry out muscle in conjunction with Time-Frequency Analysis Method
Fatigue determines and process analysis procedure analysis.
The muscle signals acquisition sensor 1 and muscle signals collection plate 3 need power supply power supply, and this patent uses 12V
Power supply, capacity 2800mAh.
The muscle signals filter circuit 4 is used to filter the noise signal of interference band:Since muscle signals mainly collect
In between 2-50Hz, it is exactly the noise in order to filter 50Hz or more that low-pass filter circuit shown in Fig. 2, which is arranged, obtain 0-50Hz
Effective MMG signal of range.
The muscle signals processing module 5 uses adaptive Length discrepancy dividing method, to continuously MMG signal into
Row dividing processing extracts the individual part feature of muscle;It is then based on Fast Fourier Transform (FFT) method, draws the frequency of MMG signal
Spectrogram.
The muscle signals analysis module 6 uses support vector machines(SVM), caffee learning method combine, carry out
Human action pattern-recognition;It is positively correlated principle based on contraction of muscle strength and muscle signals amplitude, analyzes the peak force of muscle
Square, mean power and the muscle strengths index such as bend and stretch flesh ratio;Drop to initial value 70% with Maximum isometric voluntary contraction power MVC as muscle
Tired judgment basis, the mapping relations based on MMG signal averaging frequency and amplitude and muscle Maximum isometric voluntary contraction power, in conjunction with time-frequency
Analysis method carries out muscular fatigue judgement and process analysis procedure analysis.
The collected MMG signal of control box 2 is transferred to computer by USB serial communication or Bluetooth communication or puts down
On plate computer, it is then based on muscle signals processing module 5 and muscle signals analysis module 6 carries out MMG signal processing, human action
Identification, muscle strength analysis and muscular fatigue analysis.
It is had the advantages that compared with existing well-known technique using technical solution provided by the invention:
(1)A kind of human body muscle signals analysis system of the invention, structure is simple, easy to use, by human action pattern-recognition,
Muscle strength analysis analyzes three kinds of functions with muscular fatigue and is integrated in a system, has centainly initiative.Price it is human-oriented and
Easy assembly and disassembly are easy to part replacement and maintenance.
(2)A kind of human body muscle signals analysis system of the invention compares traditional mode recognition methods, using svm classifier
Device and caffe algorithm carry out movement recognition to muscle signals, more novel.
(3)A kind of human body muscle signals analysis system of the invention is in based on contraction of muscle strength and muscle signals amplitude
It is positively correlated principle, by carrying out time frequency analysis and feature extraction to MMG signal, the peak torque of muscle can be directly calculated, be averaged
Power and bend and stretch the muscle strengths evaluation index such as flesh ratio, what mean power referred to muscle in the unit time does work value, reflects flesh
The efficiency and muscular strength size of meat acting, bend and stretch flesh ratio(F/E)The ratio of two groups of muscle group peak force squares is reflected, reflects joint motion
In balance of muscle force situation between two groups of antagonistic muscle groups.
(4)A kind of human body muscle signals analysis system of the invention, based on MMG signal averaging frequency and amplitude and human body flesh
The mapping relations of meat Maximum isometric voluntary contraction power drop to initial value 70% with Maximum isometric voluntary contraction power MVC to determine that muscular fatigue is opened
Begin, in conjunction with Time-Frequency Analysis Method, quantitative analysis muscular fatigue process.
Fig. 1 is the structural schematic diagram of the invention system.
Fig. 2 was the low-pass filter circuit schematic diagram of noise filtering.
The invention will be further described with the following Examples.
Referring to Fig.1, the system structure is mainly by muscle signals acquisition sensor 1, control box 2 and host computer procedure group
At wherein control box 2 is made of muscle signals collection plate 3, muscle signals filter circuit 4 and power supply 7;Host computer procedure can divide
For muscle signals processing module 5 and muscle signals analysis module 6, it is installed on computer and plate computer end.
Referring to Fig.1, muscle signals acquisition sensor 1 uses piezoelectric acceleration transducer, it, which has, acquires stable, product
The high-quality, advantages such as frequency response is wide, the light small and exquisite, high reliablity of sensor, high sensitivity, economic cost are low, fully meet
The requirement of accurate acquiring muscle signals, in acquisition, subject againsts sensor at respective muscle, and with common wrist guard or
Person's adhesive tape fixes sensor.
Referring to Fig.1, the main function of the muscle signals filter circuit 4 is the interference waveform for filtering 50Hz or more, is obtained
Effective MMG signal is taken, this patent uses second order voltage controlled voltage source low-pass filter circuit device.Particular circuit configurations are as shown in Figure 2.
Referring to Fig.1, the muscle signals collection plate 3 can realize that muscle signals A/D is converted, and this patent uses 16 points
The high-precision adc of resolution.
Referring to Fig.1, muscle signals processing module 5 mainly completes its function on computers, passes through Fast Fourier Transform (FFT)
MMG signal spectrum figure is drawn, the frequency information of MMG signal is obtained from spectrogram;The movement of different people different parts is collected
MMG signal amplitude have differences, handle MMG signal for convenience, it is more accurate to make to extract MMG signal characteristic, needs to adopting
The MMG signal collected is normalized, since collected MMG signal is continuously, it is difficult to analyze the list of muscle
Therefore a motion characteristic is split MMG signal using adaptive Length discrepancy dividing method, novel adaptive Length discrepancy
Partitioning algorithm seeks adaptive envelope to MMG signal, starting and the final position of movement are judged by envelope, is dividing
Its secondary envelope is sought on the basis of analysis MMG signal first enveloped line, and is asked by the diff function to secondary coenvelope line
The method of maximum and minimum, the minimum of searching movement starting and final value is originated as action action frame on secondary coenvelope line
And end position;Adaptivenon-uniform sampling is carried out to MMG signal by two positions, obtains preferable MMG signalizing activity frame.
Referring to Fig.1, muscle signals analysis module 6 mainly completes its function on computers, extracts MMG signal characteristic, into
Pedestrian's body movement recognition, muscular fatigue analysis and muscle strength analysis:The wavelet analysis method pair combined using time-frequency domain
Movement frame signal is handled, and by distortionless wavelet transform, finds out spy of the 6 layers of wavelet coefficient as movement frame signal
Value indicative is combined using SVM classifier and caffe learning algorithm, carries out human action pattern-recognition, and recognition accuracy is high;It is logical
It crosses and time frequency analysis and feature extraction is carried out to MMG signal, in conjunction with muscle strength in equidistant muscle contraction and muscle signals width
Value is positively correlated principle, the muscle strengths evaluation index such as extrapolates the peak torque of muscle, mean power and bend and stretch flesh ratio;Base
In MMG signal averaging frequency and the mapping relations of amplitude and human muscle's Maximum isometric voluntary contraction power, as Maximum isometric voluntary contraction power MVC
Drop to initial value 70%, muscle signals average frequency starts to occur being remarkably decreased to determine that muscular fatigue starts, in conjunction with time-frequency domain point
Analysis method obtains MMG signal time-domain diagram and its frequency-power spectrum figure, quantitative analysis muscular fatigue feature.
Referring to Fig.1, power supply 7 is used to drive sensor and collection plate.
Referring to Fig.1, after use, power supply is disconnected, sensor is removed from acquisition muscle, turn-off data connection
Line completes each section disassembly.
Claims (5)
1. the present invention is a kind of human body muscle signals analysis system, characteristic is its function integration, has and is based on flesh message
Number(MMG)Human action pattern-recognition, muscle strength analysis and muscular fatigue analyze three kinds of functions;The system mainly includes
Muscle signals acquire sensor 1, control box 2 and host computer procedure three parts, wherein control box 2 by muscle signals collection plate 3,
Muscle signals filter circuit 4 and power supply 7 form, and host computer procedure is divided into muscle signals processing module 5 and muscle signals analysis mould
Block 6 is installed on computer or tablet computer end;It is adopted with muscle signals in control box 2 one end of muscle signals acquisition sensor 1
The one end for collecting plate 3 is connected, and the other end of muscle signals collection plate 3 is connected to the input terminal of 4 module of muscle signals filter circuit,
The output end of muscle signals filter circuit 4 is connected to computer or tablet computer end by USB patchcord or bluetooth serial ports, passes through
Muscle signals processing module 5 and muscle signals analysis module 6 on computer or tablet computer end complete the place to MMG signal
Reason, movement recognition, muscle strength analysis and muscular fatigue analysis.
2. MMG signal processing module 5 according to claim 1 is using adaptive Length discrepancy dividing method to continuous action
MMG signal is split, and draws the MMG signal spectrum figure of different movements.
3. the wavelet analysis method for requiring the muscle signals analysis module 6 to combine using time-frequency domain according to right 1 is to dynamic
It is handled as frame signal, by distortionless wavelet transform, calculated 6 layers of wavelet coefficient is special as movement frame signal
Value indicative, then dimension-reduction treatment is carried out to these coefficients composition eigenmatrix, logarithm is combined using SVM classifier and caffe algorithm
According to classifying, the pattern-recognition of human action is completed.
4. requiring the muscle signals analysis module 6 by mentioning to MMG signal progress time frequency analysis and feature according to right 1
It takes, is positively correlated principle based on contraction of muscle strength and muscle signals amplitude, peak torque, the mean power of quantitative analysis muscle
With bend and stretch the muscle strengths index such as flesh ratio.
5. requiring the muscle signals analysis module 6 to drop to initial value 70% with Maximum isometric voluntary contraction power MVC according to right 1 to be
Muscular fatigue judgment basis, the mapping relations based on MMG signal averaging frequency and amplitude and muscle Maximum isometric voluntary contraction power, in conjunction with
Time-Frequency Analysis Method carries out muscular fatigue judgement and process analysis procedure analysis.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810518942.7A CN108903948A (en) | 2018-05-28 | 2018-05-28 | A kind of human body muscle signals analysis system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810518942.7A CN108903948A (en) | 2018-05-28 | 2018-05-28 | A kind of human body muscle signals analysis system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108903948A true CN108903948A (en) | 2018-11-30 |
Family
ID=64409654
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810518942.7A Withdrawn CN108903948A (en) | 2018-05-28 | 2018-05-28 | A kind of human body muscle signals analysis system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108903948A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109646903A (en) * | 2018-12-14 | 2019-04-19 | 中国计量大学 | Training effect evaluation method, apparatus and system |
CN111449641A (en) * | 2020-04-20 | 2020-07-28 | 浙江大学 | Evaluation device and evaluation method for muscle function state based on photoelectric signal detection |
CN111466945A (en) * | 2020-04-26 | 2020-07-31 | 重庆大学 | Muscle sound signal detection sensing device capable of automatically adjusting contact pressure |
CN112263254A (en) * | 2020-06-11 | 2021-01-26 | 复旦大学附属华山医院 | Human body energy consumption prediction system based on surface electromyogram signal sensor and prediction method thereof |
-
2018
- 2018-05-28 CN CN201810518942.7A patent/CN108903948A/en not_active Withdrawn
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109646903A (en) * | 2018-12-14 | 2019-04-19 | 中国计量大学 | Training effect evaluation method, apparatus and system |
CN111449641A (en) * | 2020-04-20 | 2020-07-28 | 浙江大学 | Evaluation device and evaluation method for muscle function state based on photoelectric signal detection |
CN111449641B (en) * | 2020-04-20 | 2021-07-20 | 浙江大学 | Evaluation device and evaluation method for muscle function state based on photoelectric signal detection |
CN111466945A (en) * | 2020-04-26 | 2020-07-31 | 重庆大学 | Muscle sound signal detection sensing device capable of automatically adjusting contact pressure |
CN112263254A (en) * | 2020-06-11 | 2021-01-26 | 复旦大学附属华山医院 | Human body energy consumption prediction system based on surface electromyogram signal sensor and prediction method thereof |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108903948A (en) | A kind of human body muscle signals analysis system | |
CN107361766B (en) | Emotion electroencephalogram signal identification method based on EMD domain multi-dimensional information | |
Istenič et al. | Multiscale entropy-based approach to automated surface EMG classification of neuromuscular disorders | |
CN102426651B (en) | Human body forearm surface electromyogram signal acquisition and pattern recognition system | |
Aziz et al. | Electromyography (EMG) data-driven load classification using empirical mode decomposition and feature analysis | |
WO2018214529A1 (en) | Electromyographic signal acquisition method and apparatus | |
CN103892829B (en) | Eye movement signal identification system based on common spatial mode and identification method thereof | |
CN103092971B (en) | A kind of sorting technique for brain-computer interface | |
CN102930284A (en) | Surface electromyogram signal pattern recognition method based on empirical mode decomposition and fractal | |
CN103793058A (en) | Method and device for classifying active brain-computer interaction system motor imagery tasks | |
CN102715903A (en) | Method for extracting electroencephalogram characteristic based on quantitative electroencephalogram | |
CN112137616B (en) | Consciousness detection device for multi-sense brain-body combined stimulation | |
CN105956624A (en) | Motor imagery electroencephalogram classification method based on space-time-frequency optimization feature sparse representation | |
CN107212883B (en) | A kind of mechanical arm writing device and control method based on brain electric control | |
CN109497997A (en) | Based on majority according to the seizure detection and early warning system of acquisition | |
CN114532993B (en) | Automatic detection method for electroencephalogram high-frequency oscillation signals of epileptic patients | |
CN110353704A (en) | Mood assessments method and apparatus based on wearable ECG monitoring | |
CN107480716A (en) | Method and system for identifying saccade signal by combining EOG and video | |
Samuel et al. | Decoding movement intent patterns based on spatiotemporal and adaptive filtering method towards active motor training in stroke rehabilitation systems | |
Nasehi et al. | A novel effective feature selection algorithm based on S-PCA and wavelet transform features in EEG signal classification | |
Too et al. | Exploring the relation between EMG pattern recognition and sampling rate using spectrogram | |
CN105686827B (en) | A kind of electromyography signal processing and feature extracting method based on microcontroller | |
CN111616680A (en) | Automatic mental load identification method and system | |
Wang et al. | Classification of surface electromyographic signals by means of multifractal singularity spectrum | |
CN114098768B (en) | Cross-individual surface electromyographic signal gesture recognition method based on dynamic threshold and EasyTL |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20181130 |