CN107357419A - A kind of motion recognition system and method based on co-contraction rate - Google Patents
A kind of motion recognition system and method based on co-contraction rate Download PDFInfo
- Publication number
- CN107357419A CN107357419A CN201710441909.4A CN201710441909A CN107357419A CN 107357419 A CN107357419 A CN 107357419A CN 201710441909 A CN201710441909 A CN 201710441909A CN 107357419 A CN107357419 A CN 107357419A
- Authority
- CN
- China
- Prior art keywords
- muscle
- motion
- contraction rate
- agonistic
- opposing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
-
- 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
- G06F2218/02—Preprocessing
-
- 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)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Neurosurgery (AREA)
- General Health & Medical Sciences (AREA)
- Neurology (AREA)
- Health & Medical Sciences (AREA)
- Dermatology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (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 application is related to a kind of motion recognition system and method based on co-contraction rate, and the motion recognition system includes：Acquisition device obtains the physiology electromyographic signal that muscle release is moved in motion process；Wave filter is pre-processed using filtering algorithm to the physiology electromyographic signal；Processor extracts the co-contraction rate of the motion muscle by average myoelectricity algorithm from pretreated physiology electromyographic signal；The processor carries out feature recognition to identify action corresponding to the motion muscle according to the co-contraction rate.The application gathers the physiology electromyographic signal of muscle by way of noninvasive, objective, science, obtains the muscular co-contraction rate that muscle is moved under different motion, different motion is identified with this；The application takes full advantage of significant difference characteristic of the co-contraction rate under different motion pattern, and objective, easy mode realizes the action recognition of motion, and practical basis is provided for the athletic posture PRS of human body.
Description
Technical field
The application is related to technical field of intelligence, is specifically related to a kind of motion recognition system based on co-contraction rate and side
Method.
Background technology
Human motion gesture recognition is one very active research direction of computer vision field.Human motion identification master
To include detection, the tracking of movement human in image and video sequence, extract motion feature to be characterized to human body behavior, from
And human motion state can be understood with characteristic.In recent years, with the popularization of smart mobile phone and wearable device,
Human motion identification is widely used in fields such as child custody, H-NTLAs.Many movement recognition system masters
If the detection of human motion is realized by acceleration transducer acquisition body motion information.
Domestic and international researcher's original adoption pin electrode insertion muscle detection electromyogram, its interference is small, and polarization is good, easily knows
Not, but because it is a kind of invasive detection method, people is not easily accepted by so that the application is by a definite limitation.By long-term
Probe into, it has been found that surface electromyogram signal can equally collect the movable information of muscle, because surface electromyogram signal is one
The noninvasive detection method of kind, it is the biological electricity provided when the electrode slice pasted by muscle surface records neuron-muscular activity
Signal, simple to operate, people is easily accepted by, therefore surface electromyogram signal is in clinical medicine, biomedical engineering, artificial limb are bionical, pattern
The fields such as identification, sports are widely used.Compared with pin electrode detects, the application of surface electromyogram signal also in
Developing stage.It is closely related with nervimuscular basic research, and nervimuscular Physiologic Studies is the application of electromyographic signal
Solid foundation is provided, and the nervimuscular research that is detected as of surface electromyogram signal provides preferable detection method.People
During motion, muscle can be shunk body, discharge electric signal.Surface electromyogram signal discharges as human motion muscle
One of physiological signal, simple to operate, real-time and bio-imitability are good.In recent years, many people identified fortune using electromyographic signal power
Dynamic model formula, but may be only available for single-degree-of-freedom finger.
Having invented CN200880012887.9 and CN201310502947.8 is using acceleration transducer respectively and regards
Frequency sequence, the training pattern plus motor pattern is, it is necessary to constantly be trained, so as to carry out the identification of different actions.And it is somebody's turn to do
For technology for for childhood, the elderly and patient, long-term training and multinode motion sensor give them
Psychology and body bring certain burden, have some limitations.The present invention is based on co-contraction rate using simple to operate
Recognizing model of movement device portable, without heavy burden, human motion pattern is identified.
It using bone as lever, joint is hinge that the motion of human body, which is, contraction of muscle for power, and in the domination of nervous system
It is lower to coordinate to complete.Contraction of muscle is to play engine in human motion, provides power for human motion, muscle is during exercise
Very important effect is play, and the mechanical property of muscle is extremely complex, therefore, muscle research is attractive and challenge
The field of property.Exercise attitudes identification is based on dynamic state, and motion will necessarily cause the various organs of human body, joint, muscle
Deng change.Muscle is to occupy composition in tissue at most, and various motions can all drive the motion of different muscle to shrink, nothing
It is a kind of priority condition to recognizing model of movement to doubt.
Therefore, in the art, how simply, it is noninvasive painless, easy to operation from the various fortune of human body
Useful information is obtained in dynamic tissue to go to carry out motion identification, is the technical issues that need to address.
The content of the invention
Based on this, it is necessary in view of the above-mentioned problems, a kind of motion recognition system and method based on co-contraction rate are provided,
Can it is noninvasive it is painless, motion identification is simply and effectively carried out according to the physiology electromyographic signal of motion release, and identify it is accurate
Rate is high, beneficial to promoting the use of.
The embodiment of the present application provides a kind of motion recognition system based on co-contraction rate, the motion recognition system bag
Include：
Acquisition device, the physiology electromyographic signal of muscle release is moved in motion process for obtaining；
Wave filter, for being pre-processed using filtering algorithm to the physiology electromyographic signal；
Processor, for extracting the motion muscle from pretreated physiology electromyographic signal by average myoelectricity algorithm
Co-contraction rate；
The processor, it is additionally operable to carry out feature recognition according to the co-contraction rate to identify that the motion muscle is corresponding
Action.
The embodiment of the present application provides a kind of action identification method based on co-contraction rate, the action identification method simultaneously
Including：
Obtain the physiology electromyographic signal that muscle release is moved in motion process；
The physiology electromyographic signal is pre-processed using filtering algorithm；
The co-contraction of the motion muscle is extracted from pretreated physiology electromyographic signal by average myoelectricity algorithm
Rate；
Feature recognition is carried out according to the co-contraction rate to identify action corresponding to the motion muscle.
Above-mentioned motion recognition system and method based on co-contraction rate, obtained in motion process and moved by acquisition device
The physiology electromyographic signal of muscle release, is located in advance followed by wave filter using filtering algorithm to the physiology electromyographic signal
Reason, the co-contraction rate of the motion muscle is extracted from pretreated physiology electromyographic signal finally by average myoelectricity algorithm
And feature recognition is carried out to identify action corresponding to the motion muscle according to the co-contraction rate.The application by it is noninvasive,
Objective, science mode gathers the physiology electromyographic signal of muscle, obtains the muscular co-contraction that muscle is moved under different motion
Rate, different motion is identified with this；It is notable under different motion pattern that the application takes full advantage of co-contraction rate
Sex differernce characteristic, objective, easy mode realize the action recognition of motion, are carried for the athletic posture PRS of human body
Practical basis is supplied.
Brief description of the drawings
Fig. 1 is the module frame chart of the motion recognition system based on co-contraction rate in an embodiment；
Fig. 2 is the flow chart of the action identification method based on co-contraction rate in an embodiment；
Fig. 3 is the implementation process schematic diagram of the motion recognition system based on co-contraction rate in a concrete application example.
Embodiment
In one embodiment, a kind of motion recognition system based on co-contraction rate, as shown in figure 1, the action is known
Other system includes but is not limited to acquisition device 11, wave filter 12 and processor 13.
In the present embodiment, acquisition device 11 is used to obtain the physiology electromyographic signal that muscle release is moved in motion process,
The wave filter 12 is used to pre-process the physiology electromyographic signal using filtering algorithm, and the processor 13 is used to pass through
Average myoelectricity algorithm extracts the co-contraction rate of the motion muscle from pretreated physiology electromyographic signal, and is used for basis
The co-contraction rate carries out feature recognition to identify action corresponding to the motion muscle.
In a preferred embodiment, the acquisition device 11, which is specifically used for obtaining, moves the short of money of muscle release in motion process
The myoelectricity value of anti-flesh and the myoelectricity value of agonistic muscle.
Furthermore, the acquisition device 11 is specifically used for：
When lean forward motion when, the agonistic muscle of acquisition includes：Left and right side musculus obliquus externus abdominis, oblique/musculus trasversus abdomins, is obtained
The Opposing muscle taken includes：Left and right side erector spinae/multifidus；
When carrying out layback motion, the agonistic muscle of acquisition includes：Left and right side erector spinae/multifidus, the Opposing muscle bag of acquisition
Include：Left and right side musculus obliquus externus abdominis, oblique/musculus trasversus abdomins；
When carrying out left-leaning motion, the agonistic muscle of acquisition includes：Left side erector spinae/multifidus, musculus obliquus externus abdominis, in abdomen tiltedly
Flesh/musculus trasversus abdomins, the Opposing muscle of acquisition include：Right side erector spinae/multifidus, musculus obliquus externus abdominis, oblique/musculus trasversus abdomins；
When carrying out Right deviation motion, the agonistic muscle of acquisition includes：Right side erector spinae/multifidus, musculus obliquus externus abdominis, in abdomen tiltedly
Flesh/musculus trasversus abdomins, the Opposing muscle of acquisition include：Left side erector spinae/multifidus, musculus obliquus externus abdominis, oblique/musculus trasversus abdomins.
It should be noted that the acquisition device 11, the standard of age, height and Body Mass Index is met specifically for obtaining
The myoelectricity value of Opposing muscle and the myoelectricity value of agonistic muscle of the lumbar vertebrae muscle group of object when carrying out standard operation.
In a particular embodiment, the acquisition of acquisition device 11 meets the standard object of age, height and Body Mass Index
The myoelectricity value of the Opposing muscle of lumbar vertebrae muscle group and the myoelectricity value of agonistic muscle, specifically include using disposable electrode piece along meat fiber
Direction is pasted on waist erector spinae/multifidus of the standard object, musculus obliquus externus abdominis, and the lumbar vertebrae flesh of oblique/musculus trasversus abdomins
On meat group.Wherein, for raising accuracy, it is necessary to which the stickup part to the standard object is first entered using 60%-80% alcohol
Row wipes, it is preferable that is wiped using 75% alcohol.
It should be noted that the processor 13 is specifically used for extracting from physiology electromyographic signal by average myoelectricity algorithm
The myoelectricity value of Opposing muscle and the myoelectricity value of agonistic muscle of the motion muscle；According to the myoelectricity value of the Opposing muscle and agonistic muscle
Myoelectricity value calculates co-contraction rate, and calculation formula is including but not limited to as follows：
Wherein, CCR is co-contraction rate, AEMGOpposing muscleFor the myoelectricity value of Opposing muscle, AEMGAgonistic muscleFor the myoelectricity value of agonistic muscle.
It is worth noting that, the present embodiment is the stability of raising system and the degree of accuracy of identification, it can use and repeatedly obtain
The mode for taking and averaging, specifically, the myoelectricity value of the Opposing muscle and the myoelectricity value of agonistic muscle are respectively Opposing muscle
The average myoelectricity value of average myoelectricity value and agonistic muscle, the calculation formula that the processor 13 uses include：
Wherein, n is the natural number more than 0, | data [i] | it is the myoelectricity value of ith Opposing muscle, | data [j] | or jth time
The myoelectricity value of agonistic muscle.
Wherein, the processor 13 carries out feature recognition to identify that the motion muscle is corresponding according to the co-contraction rate
Action, be specifically including but not limited to following process：
When the co-contraction rate is approximately equal to 0.2986, it is motion of leaning forward to identify action corresponding to the motion muscle；
When the co-contraction rate is approximately equal to 0.7219, it is layback motion to identify action corresponding to the motion muscle；
When the co-contraction rate is approximately equal to 0.5255, it is left-leaning motion to identify action corresponding to the motion muscle；
When the co-contraction rate is approximately equal to 0.4723, identify that action corresponding to the motion muscle is moved for Right deviation；
Wherein, if being identified as motion of leaning forward, including feature：Agonistic muscle shrinkage water is flat to be noticeably greater than layback motion, "Left"-deviationist
The agonistic muscle that motion and Right deviation are moved shrinks level, and shrinks level more than the Opposing muscle of itself；If being identified as layback motion,
Then include feature：Agonistic muscle shrinkage water is flat to shrink level significantly less than Opposing muscle；If being identified as left-leaning motion or Right deviation motion,
Including feature：Agonist and an tagonist shrinks horizontal basically identical.
Wherein, the wave filter 12 is corresponded to using 35~500Hz bandpass filter for removing 220 volts of supply voltage
50Hz power frequency component interfering noises, and remove electrocardiosignal 0.25~35Hz band interference noises.
The application gathers the physiology electromyographic signal of muscle by way of noninvasive, objective, science, obtains under different motion
The muscular co-contraction rate of muscle is moved, different motion is identified with this；The application takes full advantage of co-contraction rate
Significant difference characteristic under different motion pattern, objective, easy mode realize the action recognition of motion, are human body
Athletic posture PRS provides practical basis.
Please referring next to Fig. 2, in one embodiment, there is provided a kind of action identification method based on co-contraction rate, institute
State action identification method and include but is not limited to following steps.
S201, obtain the physiology electromyographic signal that muscle release is moved in motion process.
In a preferred embodiment, the S201 can specifically include：Obtain and the short of money of muscle release is moved in motion process
The myoelectricity value of anti-flesh and the myoelectricity value of agonistic muscle.
Furthermore, the S201 can include following acquisition process：
When lean forward motion when, the agonistic muscle of acquisition includes：Left and right side musculus obliquus externus abdominis, oblique/musculus trasversus abdomins, is obtained
The Opposing muscle taken includes：Left and right side erector spinae/multifidus；
When carrying out layback motion, the agonistic muscle of acquisition includes：Left and right side erector spinae/multifidus, the Opposing muscle bag of acquisition
Include：Left and right side musculus obliquus externus abdominis, oblique/musculus trasversus abdomins；
When carrying out left-leaning motion, the agonistic muscle of acquisition includes：Left side erector spinae/multifidus, musculus obliquus externus abdominis, in abdomen tiltedly
Flesh/musculus trasversus abdomins, the Opposing muscle of acquisition include：Right side erector spinae/multifidus, musculus obliquus externus abdominis, oblique/musculus trasversus abdomins；
When carrying out Right deviation motion, the agonistic muscle of acquisition includes：Right side erector spinae/multifidus, musculus obliquus externus abdominis, in abdomen tiltedly
Flesh/musculus trasversus abdomins, the Opposing muscle of acquisition include：Left side erector spinae/multifidus, musculus obliquus externus abdominis, oblique/musculus trasversus abdomins.
It should be noted that the present embodiment can specifically obtain the standard object that meets age, height and Body Mass Index
The myoelectricity value of Opposing muscle and the myoelectricity value of agonistic muscle of the lumbar vertebrae muscle group when carrying out standard operation.
In a particular embodiment, the standard object can be pasted on along meat fiber direction using disposable electrode piece
Waist erector spinae/multifidus, on musculus obliquus externus abdominis, and the lumbar vertebrae muscle group of oblique/musculus trasversus abdomins.Wherein, to improve accuracy,
Need first to wipe using 60%-80% alcohol in place of the stickup to the standard object, it is preferable that using 75% wine
The row that progresses greatly wipes.
S202, the physiology electromyographic signal is pre-processed using filtering algorithm.
In the S202 of the present embodiment, 35~500Hz bandpass filter can be used, to remove 220 volts pairs of supply voltage
The 50Hz power frequency component interfering noises answered, and remove electrocardiosignal 0.25~35Hz band interference noises.
S203, the common of the motion muscle is extracted from pretreated physiology electromyographic signal by average myoelectricity algorithm
Shrinkage factor.
In S203, institute is extracted from pretreated physiology electromyographic signal by average myoelectricity algorithm described in the present embodiment
The co-contraction rate of motion muscle is stated, is specifically included：
Myoelectricity value and the master of the Opposing muscle of the motion muscle are extracted from physiology electromyographic signal by average myoelectricity algorithm
The myoelectricity value of dynamic flesh；
Co-contraction rate is calculated according to the myoelectricity value of the Opposing muscle and the myoelectricity value of agonistic muscle, calculation formula includes：
Wherein, CCR is co-contraction rate, AEMGOpposing muscleFor the myoelectricity value of Opposing muscle, AEMGAgonistic muscleFor the myoelectricity value of agonistic muscle.
The myoelectricity value of the Opposing muscle and the myoelectricity value of agonistic muscle are respectively the average myoelectricity value and agonistic muscle of Opposing muscle
Average myoelectricity value, calculation formula include：
Wherein, n is the natural number more than 0, | data [i] | it is the myoelectricity value of ith Opposing muscle, | data [j] | or jth time
The myoelectricity value of agonistic muscle.
S204, feature recognition is carried out according to the co-contraction rate to identify action corresponding to the motion muscle.
In S204, the present embodiment is specifically including but not limited to following process：
When the co-contraction rate is approximately equal to 0.2986, it is motion of leaning forward to identify action corresponding to the motion muscle；
When the co-contraction rate is approximately equal to 0.7219, it is layback motion to identify action corresponding to the motion muscle；
When the co-contraction rate is approximately equal to 0.5255, it is left-leaning motion to identify action corresponding to the motion muscle；
When the co-contraction rate is approximately equal to 0.4723, identify that action corresponding to the motion muscle is moved for Right deviation；
Wherein, if being identified as motion of leaning forward, including feature：Agonistic muscle shrinkage water is flat to be noticeably greater than layback motion, "Left"-deviationist
The agonistic muscle that motion and Right deviation are moved shrinks level, and shrinks level more than the Opposing muscle of itself；If being identified as layback motion,
Then include feature：Agonistic muscle shrinkage water is flat to shrink level significantly less than Opposing muscle；If being identified as left-leaning motion or Right deviation motion,
Including feature：Agonist and an tagonist shrinks horizontal basically identical.
Referring to Fig. 3, in a concrete application example, the motion recognition system based on co-contraction rate can include but
It is not limited to lumbar vertebrae muscle group electromyographic signal collection module, electromyographic signal pretreatment module, co-contraction rate characteristic extracting module, fortune
Dynamic gesture recognition module etc..
Understandable to be, lumbar vertebrae muscle group electromyographic signal collection module specifically needs to carry out volunteer recruiting, lumbar vertebrae flesh
Meat mass selection is selected and exercise attitudes conceptual design.
For example, because electromyographic signal is influenceed by factors such as age, fat, healthy population is recruited part and mainly sieved
It is selected in the volunteer that the indexs such as age, height, body weight are mutually matched.Effect of the waist in human body is to form a connecting link, just as bow and arrow
The back of a bow, shrink when pulling the bow, unfold when shooting arrows, its support upper body balance, maintain bodily movement of practising Wushu, coordinate the lower part of the body maintain steadily of centre of gravity.Together
When, because waist is the maximum position of human strength, is falling, the release action of explosive force is played in the action such as whip leg.Therefore, lumbar vertebrae flesh
Meat group plays conclusive effect during body mass motion.
As it was previously stated, the present embodiment can use disposable electrode piece, the waist crossed through 75% alcohol wipe is pasted on
On the polylith muscle groups such as portion's erector spinae/multifidus, musculus obliquus externus abdominis and oblique/musculus trasversus abdomins, pasted along meat fiber direction.Survey
Examination person lean forward, swing back, left-leaning and four kinds of exercise attitudes of Right deviation when, can pass through BIOPAC's (polygraph) first
Transmitter module is by reception device of the analog signal transmission caused by muscle to MP150 (how conductive physiograph), reception device
Analog-to-digital conversion is carried out, Muscle Simulation signal is converted into one-dimensional random voltage signal, lumbar vertebrae muscle group electromyographic signal is completed with this
The electromyographic signal collection of acquisition module.
Electromyographic signal pretreatment module mainly includes carrying out at bandpass filtering and the filtering process of power frequency denoising and standardization
Reason.Understandable to be, because effective frequency range of electromyographic signal is between 10-500Hz, power frequency produced by China voltage 220v is done
Disturb 50Hz has large effect to electromyographic signal, and power frequency component need to be filtered for this.Because upper half of human body is by the shadow of heartbeat
Ring, the main frequency range of electrocardiosignal is that 0.25-35Hz has certain interference to electric signal caused by muscular movement, is collected for this
Electromyographic signal carried out 35-500Hz bandpass filtering treatment.
It should be noted that when co-contraction rate refers to that human body is moved, the average myoelectricity value of Opposing muscle and agonistic muscle and
The ratio of the average myoelectricity total value of Opposing muscle, what it reflected is the harmony of muscle, wherein, the present embodiment co-contraction rate feature carries
Modulus block uses the calculation that above-described embodiment refers to：
Equally, average myoelectricity value is the average value of muscle discharge capacity in a period of time, using the calculating side of above-described embodiment
Formula：
Wherein, n is the natural number more than 0, | data [i] | it is the myoelectricity value of ith Opposing muscle, | data [j] | or jth time
The myoelectricity value of agonistic muscle.
The concrete operating principle of the exercise attitudes identification module of the present embodiment is as follows.It is main when human body lean forward motion
Dynamic flesh is：Left and right side musculus obliquus externus abdominis and oblique/musculus trasversus abdomins, Opposing muscle are：Left and right side erector spinae/multifidus；Work as progress
During layback motion, agonistic muscle is left and right side erector spinae/multifidus, and Opposing muscle is：Left and right side musculus obliquus externus abdominis and oblique/abdomen
Transversus；During "Left"-deviationist motion, agonistic muscle is：Left side erector spinae/multifidus, musculus obliquus externus abdominis and oblique/musculus trasversus abdomins, Opposing muscle
For：Right side erector spinae/multifidus, musculus obliquus externus abdominis and oblique/musculus trasversus abdomins；When Right deviation is moved, agonistic muscle is：Erect ridge in right side
Flesh/multifidus, musculus obliquus externus abdominis and oblique/musculus trasversus abdomins, Opposing muscle are：In left side erector spinae/multifidus, musculus obliquus externus abdominis and abdomen
Oblique/musculus trasversus abdomins.
Under different motion pattern, the average myoelectricity value difference opposite sex relatively numerical value of the agonist and an tagonist of the application is as follows：
Lean forward motion when：The average myoelectricity value of agonistic muscle is about 0.45, and the average myoelectricity value of Opposing muscle is about 0.19；Layback
During motion：The average myoelectricity value of agonistic muscle is about 0.14, and the average myoelectricity value of Opposing muscle is about 0.39；During "Left"-deviationist motion：Actively
The average myoelectricity value of flesh is about 0.27, and the average myoelectricity value of Opposing muscle is about 0.3；When Right deviation is moved：The average myoelectricity of agonistic muscle
Value about 0.29, the average myoelectricity value of Opposing muscle is about 0.26.Can calculate accordingly be derived by each lower agonistic muscle of motion it
Between, the otherness between Opposing muscle.
It is not difficult to find out, leaning forward, swinging back, in left-leaning and Right deviation motor pattern, the agonistic muscle shrinkage water HUD for motion of leaning forward writes
Agonistic muscle more than layback, "Left"-deviationist and Right deviation motor pattern shrinks level, and shrinks level more than itself Opposing muscle；Before explanation
Incline motion when, agonistic muscle muscle group coordinates the human body motion that lean forward and accounts for main function.In layback motion, agonistic muscle shrinks horizontal
Level is shunk significantly less than Opposing muscle, illustrates that Opposing muscle mainly coordinates human body and carries out layback campaign.It is main in the motion of left-leaning and Right deviation
Dynamic flesh and Opposing muscle contraction is horizontal basically identical, and common human body of coordinating carries out "Left"-deviationist and Right deviation motion.Agonist and an tagonist exists
Lean forward, swing back, between left-leaning and four kinds of motor patterns of Right deviation, there is significant otherness, each p under the present embodiment<0.05.
Co-contraction rate lean forward, swing back, the otherness between left-leaning and four kinds of motor patterns of Right deviation it is as follows：
Co-contraction rate is about 0.2986 when leaning forward, and co-contraction rate is about 0.7219 during layback, co-contraction when left-leaning
Rate is about 0.5255, and co-contraction rate is about 0.4723 during Right deviation.Otherness between every motion is calculated accordingly about
For 0.000*, i.e. P=0.000*.
Therefore, it is not difficult to find out, four kinds of motor patterns can be distinguished, the otherness of characteristic parameter mainly under solution pattern,
If characteristic parameter is under four kinds of patterns, there was no significant difference, shows that four kinds of pattern-recognitions do not go out.By comparing each pattern
Co-contraction rate, the co-contraction rate for the motion that finds to lean forward is respectively present system with layback motion, left-leaning motion and Right deviation motion
Meter learns significant difference (p<0.05)；The co-contraction rate of motion of swinging back and motion of leaning forward, the motion of left-leaning and Right deviation are also deposited respectively
In the significance difference opposite sex (p<0.05)；"Left"-deviationist motion co-contraction rate with lean forward, swing back and Right deviation motor pattern be respectively present it is aobvious
Otherness (the p of work<0.05)；The co-contraction rate of Right deviation motion and lean forward, swing back and left-leaning motor pattern also exist it is notable
Otherness (p<0.05), with this, lean forward, swing back, the co-contraction rate between left-leaning and four kinds of motor patterns of Right deviation is present substantially
Difference, can identify motor pattern using co-contraction rate.
The application gathers the physiology electromyographic signal of muscle by way of noninvasive, objective, science, obtains under different motion
The muscular co-contraction rate of muscle is moved, different motion is identified with this；The application takes full advantage of co-contraction rate
Significant difference characteristic under different motion pattern, objective, easy mode realize the action recognition of motion, are human body
Athletic posture PRS provides practical basis.
Embodiments herein is the foregoing is only, not thereby limits the scope of the claims of the application, it is every to utilize this Shen
Please the equivalent structure made of specification and accompanying drawing content or equivalent flow conversion, or be directly or indirectly used in other related skills
Art field, is similarly included in the scope of patent protection of the application.
Claims (12)
1. a kind of motion recognition system based on co-contraction rate, it is characterised in that the motion recognition system includes：
Acquisition device, the physiology electromyographic signal of muscle release is moved in motion process for obtaining；
Wave filter, for being pre-processed using filtering algorithm to the physiology electromyographic signal；
Processor, for extracting being total to for the motion muscle from pretreated physiology electromyographic signal by average myoelectricity algorithm
Same shrinkage factor；
The processor, it is additionally operable to be moved corresponding to the motion muscle to identify according to co-contraction rate progress feature recognition
Make.
2. motion recognition system according to claim 1, it is characterised in that the processor is specifically used for：
The myoelectricity value and agonistic muscle of the Opposing muscle of the motion muscle are extracted from physiology electromyographic signal by average myoelectricity algorithm
Myoelectricity value；
Co-contraction rate is calculated according to the myoelectricity value of the Opposing muscle and the myoelectricity value of agonistic muscle, calculation formula includes：
Wherein, CCR is co-contraction rate, AEMGOpposing muscleFor the myoelectricity value of Opposing muscle, AEMGAgonistic muscleFor the myoelectricity value of agonistic muscle.
3. motion recognition system according to claim 2, it is characterised in that the myoelectricity value of the Opposing muscle and agonistic muscle
Myoelectricity value is respectively the average myoelectricity value of Opposing muscle and the average myoelectricity value of agonistic muscle, the calculation formula bag that the processor uses
Include：
Wherein, n is the natural number more than 0, | data [i] | it is the myoelectricity value of ith Opposing muscle, | data [j] | or jth time is actively
The myoelectricity value of flesh.
4. motion recognition system according to claim 2, it is characterised in that the acquisition device is specifically used for：
Obtain the myoelectricity value of Opposing muscle and the myoelectricity value of agonistic muscle that muscle release is moved in motion process.
5. motion recognition system according to claim 4, it is characterised in that the acquisition device is specifically used for：
When lean forward motion when, the agonistic muscle of acquisition includes：Left and right side musculus obliquus externus abdominis, oblique/musculus trasversus abdomins, acquisition
Opposing muscle includes：Left and right side erector spinae/multifidus；
When carrying out layback motion, the agonistic muscle of acquisition includes：Left and right side erector spinae/multifidus, the Opposing muscle of acquisition include：
Left and right side musculus obliquus externus abdominis, oblique/musculus trasversus abdomins；
When carrying out left-leaning motion, the agonistic muscle of acquisition includes：Left side erector spinae/multifidus, musculus obliquus externus abdominis, oblique/abdomen
Transversus, the Opposing muscle of acquisition include：Right side erector spinae/multifidus, musculus obliquus externus abdominis, oblique/musculus trasversus abdomins；
When carrying out Right deviation motion, the agonistic muscle of acquisition includes：Right side erector spinae/multifidus, musculus obliquus externus abdominis, oblique/abdomen
Transversus, the Opposing muscle of acquisition include：Left side erector spinae/multifidus, musculus obliquus externus abdominis, oblique/musculus trasversus abdomins.
6. motion recognition system according to claim 5, it is characterised in that the processor is according to the co-contraction rate
Carry out feature recognition and acted with identifying corresponding to the motion muscle, specifically included：
When the co-contraction rate is approximately equal to 0.2986, it is motion of leaning forward to identify action corresponding to the motion muscle；
When the co-contraction rate is approximately equal to 0.7219, it is layback motion to identify action corresponding to the motion muscle；
When the co-contraction rate is approximately equal to 0.5255, it is left-leaning motion to identify action corresponding to the motion muscle；
When the co-contraction rate is approximately equal to 0.4723, identify that action corresponding to the motion muscle is moved for Right deviation；
Wherein, if being identified as motion of leaning forward, including feature：Agonistic muscle shrinkage water is flat to be noticeably greater than layback motion, left-leaning motion
Shunk with the agonistic muscle of Right deviation motion horizontal and horizontal more than the Opposing muscle contraction of itself；If being identified as layback motion, wrap
Include feature：Agonistic muscle shrinkage water is flat to shrink level significantly less than Opposing muscle；If left-leaning motion or Right deviation motion are identified as, including
Feature：Agonist and an tagonist shrinks horizontal basically identical.
7. motion recognition system according to claim 4, it is characterised in that the acquisition device, be specifically used for：
Obtain Opposing muscle of the lumbar vertebrae muscle group for the standard object for meeting age, height and Body Mass Index when carrying out standard operation
Myoelectricity value and agonistic muscle myoelectricity value.
8. motion recognition system according to claim 7, it is characterised in that the acquisition device, which obtains, meets age, body
The myoelectricity value of Opposing muscle and the myoelectricity value of agonistic muscle of the lumbar vertebrae muscle group of high and Body Mass Index standard object, are specifically included：
Waist erector spinae/multifidus of the standard object is pasted on along meat fiber direction using disposable electrode piece, outside abdomen
On oblique, and the lumbar vertebrae muscle group of oblique/musculus trasversus abdomins.
9. motion recognition system according to claim 1, it is characterised in that the wave filter uses 35~500Hz band
Bandpass filter, for removing 50Hz power frequency component interfering noises corresponding to 220 volts of supply voltage, and remove electrocardiosignal 0.25
~35Hz band interference noises.
10. a kind of action identification method based on co-contraction rate, it is characterised in that the action identification method includes：
Obtain the physiology electromyographic signal that muscle release is moved in motion process；
The physiology electromyographic signal is pre-processed using filtering algorithm；
The co-contraction rate of the motion muscle is extracted from pretreated physiology electromyographic signal by average myoelectricity algorithm；
Feature recognition is carried out according to the co-contraction rate to identify action corresponding to the motion muscle.
11. action identification method according to claim 10, it is characterised in that it is described by average myoelectricity algorithm from pre-
The co-contraction rate of the motion muscle is extracted in physiology electromyographic signal after reason, is specifically included：
The myoelectricity value and agonistic muscle of the Opposing muscle of the motion muscle are extracted from physiology electromyographic signal by average myoelectricity algorithm
Myoelectricity value；
Co-contraction rate is calculated according to the myoelectricity value of the Opposing muscle and the myoelectricity value of agonistic muscle, calculation formula includes：
Wherein, CCR is co-contraction rate, AEMGOpposing muscleFor the myoelectricity value of Opposing muscle, AEMGAgonistic muscleFor the myoelectricity value of agonistic muscle.
12. action identification method according to claim 11, it is characterised in that the myoelectricity value and agonistic muscle of the Opposing muscle
Myoelectricity value be respectively the average myoelectricity value of Opposing muscle and the average myoelectricity value of agonistic muscle, calculation formula includes：
Wherein, n is the natural number more than 0, | data [i] | it is the myoelectricity value of ith Opposing muscle, | data [j] | or jth time is actively
The myoelectricity value of flesh.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710441909.4A CN107357419B (en) | 2017-06-13 | 2017-06-13 | Action recognition system and method based on common shrinkage rate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710441909.4A CN107357419B (en) | 2017-06-13 | 2017-06-13 | Action recognition system and method based on common shrinkage rate |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107357419A true CN107357419A (en) | 2017-11-17 |
| CN107357419B CN107357419B (en) | 2020-05-19 |
Family
ID=60273885
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710441909.4A Active CN107357419B (en) | 2017-06-13 | 2017-06-13 | Action recognition system and method based on common shrinkage rate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107357419B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108198623A (en) * | 2017-12-15 | 2018-06-22 | 东软集团股份有限公司 | Human body condition detection method, device, storage medium and electronic equipment |
| CN110232976A (en) * | 2019-07-01 | 2019-09-13 | 上海电机学院 | A kind of Activity recognition method based on the measurement of waist shoulder surface myoelectric |
| CN110859619A (en) * | 2018-08-28 | 2020-03-06 | 易适康连（上海）科技有限公司 | Device, system and method for acquiring abdominal electromyogram data |
| CN113288532A (en) * | 2021-05-31 | 2021-08-24 | 北京京东乾石科技有限公司 | Myoelectric control method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104107134A (en) * | 2013-12-10 | 2014-10-22 | 中山大学 | Myoelectricity feedback based upper limb training method and system |
| US20150017623A1 (en) * | 2012-01-04 | 2015-01-15 | Gabriele Ceruti | Method and apparatus for neuromotor rehabilitation using interactive setting systems |
| CN105148404A (en) * | 2015-09-16 | 2015-12-16 | 燕铁斌 | Electrical stimulation training system and training method based on health-side limb electromyographic signal trigger |
-
2017
- 2017-06-13 CN CN201710441909.4A patent/CN107357419B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150017623A1 (en) * | 2012-01-04 | 2015-01-15 | Gabriele Ceruti | Method and apparatus for neuromotor rehabilitation using interactive setting systems |
| CN104107134A (en) * | 2013-12-10 | 2014-10-22 | 中山大学 | Myoelectricity feedback based upper limb training method and system |
| CN105148404A (en) * | 2015-09-16 | 2015-12-16 | 燕铁斌 | Electrical stimulation training system and training method based on health-side limb electromyographic signal trigger |
Non-Patent Citations (1)
| Title |
|---|
| 魏勇，刘宇: "肌肉共同收缩研究进展", 《中国体育科技》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108198623A (en) * | 2017-12-15 | 2018-06-22 | 东软集团股份有限公司 | Human body condition detection method, device, storage medium and electronic equipment |
| CN110859619A (en) * | 2018-08-28 | 2020-03-06 | 易适康连（上海）科技有限公司 | Device, system and method for acquiring abdominal electromyogram data |
| CN110859619B (en) * | 2018-08-28 | 2023-02-14 | 易适康连（上海）科技有限公司 | Device, system and method for acquiring abdominal electromyogram data |
| CN110232976A (en) * | 2019-07-01 | 2019-09-13 | 上海电机学院 | A kind of Activity recognition method based on the measurement of waist shoulder surface myoelectric |
| CN113288532A (en) * | 2021-05-31 | 2021-08-24 | 北京京东乾石科技有限公司 | Myoelectric control method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107357419B (en) | 2020-05-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Celik et al. | Gait analysis in neurological populations: Progression in the use of wearables | |
| US10335080B2 (en) | Biomechanical activity monitoring | |
| CN107357419A (en) | A kind of motion recognition system and method based on co-contraction rate | |
| Godfrey et al. | Direct measurement of human movement by accelerometry | |
| CN103584919B (en) | Multi-modal bio signal synchronous detecting system and method | |
| CN109288651A (en) | Personalized upper-limbs rehabilitation training robot system and its recovery training method | |
| CN107832686A (en) | Merge the lower limb motion mode recognition methods of surface myoelectric and acceleration signal | |
| Ma et al. | sEMG-based trunk compensation detection in rehabilitation training | |
| CN105595995B (en) | Physiological data detection system, detection device, terminal device, and data analysis method | |
| CN103440498A (en) | Surface electromyogram signal identification method based on LDA algorithm | |
| CN113066260B (en) | Parkinson's disease early warning system based on daily behavior analysis | |
| CN110464349A (en) | A kind of upper extremity exercise function score method based on hidden Semi-Markov Process | |
| US20190320944A1 (en) | Biomechanical activity monitoring | |
| CN109567812A (en) | Gait analysis system based on Intelligent insole | |
| CN106821308A (en) | The method and apparatus of contraction movement assessment under a kind of skeletal muscle stimulating | |
| Dorofeev et al. | The assessment of gait features according to the data of a portable acceleration sensor in an intelligent monitoring system | |
| CN113713333B (en) | Dynamic virtual induction method and system for lower limb rehabilitation full training process | |
| Gong et al. | Design and implementation of wearable dynamic electrocardiograph real-time monitoring terminal | |
| CN116966056A (en) | Upper limb rehabilitation evaluation system for cerebral apoplexy patient | |
| Moreside et al. | How do elliptical machines differ from walking: a study of torso motion and muscle activity | |
| CN110215202A (en) | The pre- measuring/correlation method in Cardiac RR interval based on gait nonlinear characteristic | |
| Zhang et al. | Walking rehabilitation evaluation based on gait analysis | |
| CN114010453A (en) | Feedback wearing system based on motion signal and preprocessing method | |
| CN102125724A (en) | Control method for controlling open-loop functional electric stimulation of hand functions through acoustic muscle signals | |
| Celić et al. | WBAN for physical activity monitoring in health care and wellness |
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 |