CN109446957A - One kind being based on EMG signal recognition methods - Google Patents

One kind being based on EMG signal recognition methods Download PDF

Info

Publication number
CN109446957A
CN109446957A CN201811214783.8A CN201811214783A CN109446957A CN 109446957 A CN109446957 A CN 109446957A CN 201811214783 A CN201811214783 A CN 201811214783A CN 109446957 A CN109446957 A CN 109446957A
Authority
CN
China
Prior art keywords
signal
acquisition module
wavelet
coefficient
scale
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201811214783.8A
Other languages
Chinese (zh)
Other versions
CN109446957B (en
Inventor
周曦
杨学锐
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.)
Guangzhou Yuncong Artificial Intelligence Technology Co Ltd
Original Assignee
Guangzhou Yuncong Artificial Intelligence 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 Guangzhou Yuncong Artificial Intelligence Technology Co Ltd filed Critical Guangzhou Yuncong Artificial Intelligence Technology Co Ltd
Priority to CN201811214783.8A priority Critical patent/CN109446957B/en
Publication of CN109446957A publication Critical patent/CN109446957A/en
Application granted granted Critical
Publication of CN109446957B publication Critical patent/CN109446957B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/02—Preprocessing
    • G06F2218/04—Denoising
    • G06F2218/06—Denoising by applying a scale-space analysis, e.g. using wavelet analysis
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2203/00—Indexing scheme relating to G06F3/00 - G06F3/048
    • G06F2203/01—Indexing scheme relating to G06F3/01
    • G06F2203/011—Emotion or mood input determined on the basis of sensed human body parameters such as pulse, heart rate or beat, temperature of skin, facial expressions, iris, voice pitch, brain activity patterns
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Dermatology (AREA)
  • General Health & Medical Sciences (AREA)
  • Neurology (AREA)
  • Neurosurgery (AREA)
  • Human Computer Interaction (AREA)
  • Image Analysis (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

One kind being based on EMG signal recognition methods, using following steps, step 1: acquisition module being mounted on human arm, acquisition module acquisition human arm muscle electric signal carries out and data are normalized；Step 2: collected signal is transferred to the receiving module of server-side by acquisition module by signal transmission module；Step 3: after receiving module receives signal, server-side by wavelet transformation is filtered noise reduction to signal is received, the signal that obtains that treated.Deep learning processing and virtual reality technology of the invention based on muscle electric signal, realize a set of novel man-machine interactive system.Compared to other human-computer interaction means, the present invention is at low cost, and extracting and processing for EMG signal is more at low cost than the method for view-based access control model.

Description

One kind being based on EMG signal recognition methods
Technical field
The present invention relates to field of signal identification, and in particular to one kind is based on EMG signal recognition methods.
Background technique
With the development of internet technology, deep learning and signal processing technology are more and more in people's real life It applies in different fields.In terms of medicine, intelligent diagnostics, medical image processing has been achieved for certain effect.But Field of medical rehabilitation, related application are also fairly simple.
When human motion, it is excited that the nerve signal that brain generates can make different muscle groups, to generate surface muscle electricity Signal has important application in the fields such as smart home and rehabilitation in conjunction with deep learning identification technology.Virtual reality is in recent years It is quick to develop, it is applied to rehabilitation field, natural interactive mode, the help Rehabilitation of qualitative, quantitative can be passed through.At present The motion capture and identification of virtual reality rehabilitation are based primarily upon computer vision technique, because the reason of perceptual masking effect, tool There is the disadvantages of algorithm is complicated, and accuracy is low, at high cost.
To solve the above-mentioned problems, this patent is by acquisition human body surface myoelectric signal, and processing and identification signal institute are right The gesture information answered, combines artificial intelligence and virtual reality technology, and design is based on virtual reality healing hand function training mission System.
Summary of the invention
In view of the deficiencies of the prior art, the present invention proposes one kind to be based on EMG signal recognition methods, and specific technical solution is such as Under:
One kind being based on EMG signal recognition methods, it is characterised in that:
Using following steps,
Step 1: acquisition module being mounted on human arm, acquisition module acquires human arm muscle electric signal and carries out simultaneously Data are normalized；
Step 2: collected signal is transferred to the receiving module of server-side by acquisition module by signal transmission module；
Step 3: after receiving module receives signal, server-side is filtered noise reduction by wavelet transformation to reception signal, obtains To treated signal, wavelet transformation is filtered the concrete mode of noise reduction are as follows:
To reception signal f (t) ∈ L2(R) it is indicated with following formula are as follows:
The right first item is projection of the f (t) in scale space in formula, and Section 2 is projection of the f (t) in wavelet space；
J is that any scale starts, cj,kFor scale coefficient, dj,kFor wavelet coefficient；
Scale coefficient C is obtained as followsj,kWith wavelet coefficient Dj,k:
cj,k=< f (x), φj,k(x) >=∫ f (x) φj,k(x)dx
dj,k=< f (x), φj,k(x) >=∫ f (x) φj,k(x)dx
By scale coefficient cj,kWith wavelet coefficient dj,kCollectively form two-dimentional time-frequency spectrum；
Step 4: the feature in two-dimentional time-frequency spectrum being extracted using depth characteristic fusion convolutional neural networks；
Step 5: obtaining gesture classification result using trained classification output layer.
Further: acquisition module uses eight channel muscle electrical signal collection armlets, by being worn on human arm, eight The voltage change of one week eight position of arm is acquired by a sensor with the sample rate of 400Hz.
The invention has the benefit that the present invention is based on the processing of the deep learning of muscle electric signal and virtual reality technology, Realize a set of novel man-machine interactive system.Compared to other human-computer interaction means, the present invention is at low cost, EMG signal It extracts and processes more at low cost than the method for view-based access control model；
User experience is good, oneself training only can need to wear without purchasing special equipment or hospital being gone to use at home Bracelet；
The method of scene strong robustness, view-based access control model is limited to environment and light, for dark scene, muscle point signal Identification can be extracted；
Detailed description of the invention
Fig. 1 is work flow diagram of the invention.
Specific embodiment
The preferred embodiments of the present invention will be described in detail with reference to the accompanying drawing, so that advantages and features of the invention energy It is easier to be readily appreciated by one skilled in the art, so as to make a clearer definition of the protection scope of the present invention.
It is a kind of as shown in Figure 1: to be based on EMG signal recognition methods, it is characterised in that:
Using following steps,
Step 1: acquisition module being mounted on human arm, acquisition module uses eight channel muscle electrical signal collection arms Ring, by being worn on human arm, eight sensors can be by voltage change the adopting with 400Hz of one week eight position of arm Sample rate is acquired human arm muscle electric signal, and and data is normalized；
Step 2: collected signal is transferred to the receiving module of server-side by acquisition module by signal transmission module；
Step 3: after receiving module receives signal, server-side is filtered noise reduction by wavelet transformation to reception signal, obtains To treated signal, wavelet transformation is filtered the concrete mode of noise reduction are as follows:
To reception signal f (t) ∈ L2(R) it is indicated with following formula are as follows:
The right first item is projection of the f (t) in scale space in formula, and Section 2 is projection of the f (t) in wavelet space；
J is that any scale starts, Cj,kFor scale coefficient, Dj,kFor wavelet coefficient；
Scale coefficient C is obtained as followsj,kWith wavelet coefficient Dj,k:
cj,k=< f (x), φj,k(x) >=∫ f (x) φj,k(x)dx
dj,k=< f (x), φj,k(x) >=∫ f (x) φj,k(x)dx
By scale coefficient Cj,kWith wavelet coefficient Dj,kCollectively form two-dimentional time-frequency spectrum；
Step 4: the feature in two-dimentional time-frequency spectrum being extracted using depth characteristic fusion convolutional neural networks；
Step 5: obtaining gesture classification result using trained classification output layer.

Claims (2)

1. one kind is based on EMG signal recognition methods, it is characterised in that:
Using following steps,
Step 1: acquisition module being mounted on human arm, acquisition module acquires human arm muscle electric signal and carries out simultaneously logarithm According to being normalized；
Step 2: collected signal is transferred to the receiving module of server-side by acquisition module by signal transmission module；
Step 3: after receiving module receives signal, server-side is filtered noise reduction by wavelet transformation to reception signal, obtains everywhere Signal after reason, wavelet transformation are filtered the concrete mode of noise reduction are as follows:
To reception signal f (t) ∈ L2(R) it is indicated with following formula are as follows:
The right first item is projection of the f (t) in scale space in formula, and Section 2 is projection of the f (t) in wavelet space；
J is that any scale starts, cj,kFor scale coefficient, dj,kFor wavelet coefficient；
Scale coefficient C is obtained as followsj,kWith wavelet coefficient Dj,k:
cj,k=< f (x), φj,k(x) >=∫ f (x) φj,k(x)dx
dj,k=< f (x), φj,k(x) >=∫ f (x) φj,k(x)dx
By scale coefficient cj,kWith wavelet coefficient dj,kCollectively form two-dimentional time-frequency spectrum；
Step 4: the feature in two-dimentional time-frequency spectrum being extracted using depth characteristic fusion convolutional neural networks；
Step 5: obtaining gesture classification result using trained classification output layer.
2. a kind of according to claim 1 be based on EMG signal recognition methods, it is characterised in that: acquisition module uses eight channels Muscle electrical signal collection armlet, by being worn on human arm, eight sensors become the voltage of one week eight position of arm Change is acquired with the sample rate of 400Hz.
CN201811214783.8A 2018-10-18 2018-10-18 EMG signal identification method Active CN109446957B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811214783.8A CN109446957B (en) 2018-10-18 2018-10-18 EMG signal identification method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811214783.8A CN109446957B (en) 2018-10-18 2018-10-18 EMG signal identification method

Publications (2)

Publication Number Publication Date
CN109446957A true CN109446957A (en) 2019-03-08
CN109446957B CN109446957B (en) 2021-01-01

Family

ID=65547154

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811214783.8A Active CN109446957B (en) 2018-10-18 2018-10-18 EMG signal identification method

Country Status (1)

Country Link
CN (1) CN109446957B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109924977A (en) * 2019-03-21 2019-06-25 西安交通大学 A kind of surface electromyogram signal classification method based on CNN and LSTM
CN110414619A (en) * 2019-08-05 2019-11-05 重庆工商职业学院 One kind being based on EMG signal recognition methods
CN110738140A (en) * 2019-09-26 2020-01-31 深圳大学 action recognition method and system based on forearm electromyogram signals

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080015452A1 (en) * 2006-06-30 2008-01-17 Ricci Carlos A Method of processing electrocardiogram waveform
CN101587546A (en) * 2009-06-12 2009-11-25 浙江大学 Identification method of lower limb action pattern based on supporting vector multivariate classification
CN101832471A (en) * 2010-04-19 2010-09-15 哈尔滨工程大学 Signal identification and classification method
CN101859377A (en) * 2010-06-08 2010-10-13 杭州电子科技大学 Electromyographic signal classification method based on multi-kernel support vector machine
US20110077538A1 (en) * 2009-09-11 2011-03-31 Agency For Science, Technology And Research Electrocardiogram signal processing system
US20140016840A1 (en) * 2008-06-30 2014-01-16 Nellcor Puritan Bennett Ireland Systems and methods for ridge selection in scalograms of signals
CN105320969A (en) * 2015-11-20 2016-02-10 北京理工大学 A heart rate variability feature classification method based on multi-scale Renyi entropy
CN105997064A (en) * 2016-05-17 2016-10-12 成都奥特为科技有限公司 Method for identifying human lower limb surface EMG signals (electromyographic signals)
EP3171203A1 (en) * 2015-11-18 2017-05-24 CGG Services SAS Adaptive ensemble-based method and device for highly-nonlinear problems
CN107273798A (en) * 2017-05-11 2017-10-20 华南理工大学 A kind of gesture identification method based on surface electromyogram signal

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080015452A1 (en) * 2006-06-30 2008-01-17 Ricci Carlos A Method of processing electrocardiogram waveform
US20140016840A1 (en) * 2008-06-30 2014-01-16 Nellcor Puritan Bennett Ireland Systems and methods for ridge selection in scalograms of signals
CN101587546A (en) * 2009-06-12 2009-11-25 浙江大学 Identification method of lower limb action pattern based on supporting vector multivariate classification
US20110077538A1 (en) * 2009-09-11 2011-03-31 Agency For Science, Technology And Research Electrocardiogram signal processing system
CN101832471A (en) * 2010-04-19 2010-09-15 哈尔滨工程大学 Signal identification and classification method
CN101859377A (en) * 2010-06-08 2010-10-13 杭州电子科技大学 Electromyographic signal classification method based on multi-kernel support vector machine
EP3171203A1 (en) * 2015-11-18 2017-05-24 CGG Services SAS Adaptive ensemble-based method and device for highly-nonlinear problems
CN105320969A (en) * 2015-11-20 2016-02-10 北京理工大学 A heart rate variability feature classification method based on multi-scale Renyi entropy
CN105997064A (en) * 2016-05-17 2016-10-12 成都奥特为科技有限公司 Method for identifying human lower limb surface EMG signals (electromyographic signals)
CN107273798A (en) * 2017-05-11 2017-10-20 华南理工大学 A kind of gesture identification method based on surface electromyogram signal

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
KIKUO ASAI 等: ""Finger motion estimation based on frequency conversion of EMG signals and image recognition using convolutional neural network"", 《2017 17TH INTERNATIONAL CONFERENCE ON CONTROL, AUTOMATION AND SYSTEMS (ICCAS)》 *
李林伟: ""智能轮椅人机接口中表面肌电信号动作识别"", 《中国优秀硕士学位论文全文数据库-医药卫生科技辑》 *
杨亚慧 等: ""基于卷积神经网络的表面肌电信号手势识别"", 《微型机与应用》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109924977A (en) * 2019-03-21 2019-06-25 西安交通大学 A kind of surface electromyogram signal classification method based on CNN and LSTM
CN110414619A (en) * 2019-08-05 2019-11-05 重庆工商职业学院 One kind being based on EMG signal recognition methods
CN110414619B (en) * 2019-08-05 2023-04-07 重庆工商职业学院 EMG signal identification method
CN110738140A (en) * 2019-09-26 2020-01-31 深圳大学 action recognition method and system based on forearm electromyogram signals
CN110738140B (en) * 2019-09-26 2023-09-12 深圳大学 Action recognition method and system based on forearm electromyographic signals

Also Published As

Publication number Publication date
CN109446957B (en) 2021-01-01

Similar Documents

Publication Publication Date Title
Gurbuz et al. American sign language recognition using rf sensing
Bijalwan et al. HDL-PSR: Modelling spatio-temporal features using hybrid deep learning approach for post-stroke rehabilitation
Fakhrulddin et al. Convolutional neural networks (CNN) based human fall detection on body sensor networks (BSN) sensor data
US20200275895A1 (en) Methods and apparatus for unsupervised one-shot machine learning for classification of human gestures and estimation of applied forces
Nakano et al. Effect of dynamic feature for human activity recognition using smartphone sensors
CN109993068B (en) Non-contact human emotion recognition method based on heart rate and facial features
CN110688910B (en) Method for realizing wearable human body basic gesture recognition
CN109446957A (en) One kind being based on EMG signal recognition methods
CN111428583B (en) Visual compensation method based on neural network and touch lattice
CN115590515A (en) Emotion recognition method and system based on generative self-supervision learning and electroencephalogram signals
Hirsch et al. Hands-free gesture control with a capacitive textile neckband
CN114998983A (en) Limb rehabilitation method based on augmented reality technology and posture recognition technology
Thakur et al. Vision based computer mouse control using hand gestures
Zhang et al. Real-time activity and fall risk detection for aging population using deep learning
Makarov et al. Russian sign language dactyl recognition
Suppiah et al. Fuzzy inference system (FIS)-long short-term memory (LSTM) network for electromyography (EMG) signal analysis
Zea et al. Real-time hand gesture recognition: a long short-term memory approach with electromyography
Tripathi et al. SurfMyoAiR: A surface electromyography-based framework for airwriting recognition
Hu et al. Towards a multimodal human activity dataset for healthcare
CN111259699A (en) Human body action recognition and prediction method and device
CN109634419A (en) The recognition methods of healing robot motion intention and its computer readable storage medium
CN109498362A (en) A kind of hemiplegic patient&#39;s hand movement function device for healing and training and model training method
Rajendran et al. Emergency gesture communication by patients, elderly and differently abled with care takers using wearable data gloves
Chen et al. Hand gestures recognition from multi-channel forearm EMG signals
Suganya et al. Design Of a Communication aid for physically challenged

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant