CN106821390A - A kind of bow-backed alarm set and based reminding method based on muscle signals detection - Google Patents

A kind of bow-backed alarm set and based reminding method based on muscle signals detection Download PDF

Info

Publication number
CN106821390A
CN106821390A CN201710154080.XA CN201710154080A CN106821390A CN 106821390 A CN106821390 A CN 106821390A CN 201710154080 A CN201710154080 A CN 201710154080A CN 106821390 A CN106821390 A CN 106821390A
Authority
CN
China
Prior art keywords
muscle signals
bow
backed
signal
band
Prior art date
Application number
CN201710154080.XA
Other languages
Chinese (zh)
Inventor
王璐
葛科铎
王萌
谢能刚
Original Assignee
安徽工业大学
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 安徽工业大学 filed Critical 安徽工业大学
Priority to CN201710154080.XA priority Critical patent/CN106821390A/en
Publication of CN106821390A publication Critical patent/CN106821390A/en

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/103Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/1107Measuring contraction of parts of the body, e.g. organ, muscle
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/103Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/1118Determining activity level
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/45For evaluating or diagnosing the musculoskeletal system or teeth
    • A61B5/4519Muscles
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/48Other medical applications
    • A61B5/4803Speech analysis specially adapted for diagnostic purposes
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/68Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6813Specially adapted to be attached to a specific body part
    • A61B5/6823Trunk, e.g., chest, back, abdomen, hip
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/68Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/683Means for maintaining contact with the body
    • A61B5/6831Straps, bands or harnesses
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7225Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/72Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235Details of waveform analysis
    • A61B5/7264Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A61B5/7267Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/74Details of notification to user or communication with user or patient ; user input means
    • A61B5/7405Details of notification to user or communication with user or patient ; user input means using sound
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/74Details of notification to user or communication with user or patient ; user input means
    • A61B5/7455Details of notification to user or communication with user or patient ; user input means characterised by tactile indication, e.g. vibration or electrical stimulation
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Detecting, measuring or recording for diagnostic purposes; Identification of persons
    • A61B5/74Details of notification to user or communication with user or patient ; user input means
    • A61B5/746Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms
    • GPHYSICS
    • G08SIGNALLING
    • G08BSIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00Alarms responsive to a single specified undesired or abnormal operating condition and not elsewhere provided for
    • G08B21/18Status alarms
    • G08B21/24Reminder alarms, e.g. anti-loss alarms

Abstract

The invention discloses a kind of bow-backed alarm set and based reminding method based on muscle signals detection, belong to flesh sound detection technique field.Bow-backed alarm set of the invention, including the elastic suspenders positioned at human body back and the band positioned at waist, muscle signals acquisition module, signal processing module and signal performing module are provided with the elastic suspenders, the muscle signals acquisition module and signal performing module are electrically connected with signal processing module, muscle signals acquisition module is distributed in elastic suspenders, and correspondence human body latissimus dorsi region;The signal performing module is used to produce vibration or auditory tone cues to the bow-backed information for detecting.The present invention makes it form effectively cooperation when in use with muscle signals detection lug by the improvement to band structure, improves the degree of accuracy of bow-backed detection, and bow-backed attitude that can be in time to user is reminded.

Description

A kind of bow-backed alarm set and based reminding method based on muscle signals detection

Technical field

The present invention relates to flesh sound detection technique application field, more specifically to a kind of based on muscle signals detection Bow-backed alarm set and based reminding method.

Background technology

In recent years, China humpback number is on the increase.Especially teenager, due to not keeping waist to endure in study It is straight or straight and upright nonstandard, cause teenager's hunchback phenomenon serious long lasting for this state.Bow-backed deformity can often cause chest Wide deformity, the health problem, greatly influence breathing, circulation and digestion work(such as diaphragm activity is limited, internal organ are frightened, chest and abdomen are affixed Energy.Therefore, prevention hunchback seems particularly significant.

There are the bow-backed auxiliary equipment of some preventions, such as BABAKA in the market.This bow-backed correction mode of BABAKA is main It is to make waist straight and upright by auxiliary force, by forcing power by shoulder toward post-tensioning, while producing a downward pulling force, is unfavorable for bone Normal development, and wear and free all inconvenient.Additionally, similar bow-backed apparatus for correcting is all to carry out attitude by external force to rectify Just, user can not be made to form the consciousness of self, is cured the symptoms, not the disease, it is difficult to play long-term rectification effect.Therefore, It is particularly important how a kind of application effect good bow-backed apparatus for correcting is developed.

Find there is the technology for making it independently correct faulty sitting posture to user's prompting, such as side in the prior art by retrieval Case 1:China Patent No.:CN 2639967Y, the day for announcing:On September 8th, 2004, bow-backed attitude is prevented that application discloses a kind of Alarm, power supply is connected to each circuit by wire, and signal generator sends detection signal by signal transduction and attitude detection The signal receiver being connected to signal processor is input into, alarm is connected with signal processor;The neck with people can be hung Soft rope band on son is connected;Power supply uses dry cell, and signal generator is a light emitting diode;The signal receiver is One photosensitive component, the signal that photosensitive component will be detected is converted to electric signal and is sent to signal processor, by signal Processor is amplified treatment to it, produces an alarm signal to be input into alarm, and vibrations are produced so as to allow, and reaches prompting police The purpose shown.

And for example scheme 2:Chinese Patent Application No.:201410158064.4, the applying date:On April 16th, 2014, this application case The back apparatus for correcting of bending warning, including an overcoat for vest structure are disclosed, overcoat back is the stock of template, in set Some monitoring sites are set on frame, dc source, switch, alarm module are set, monitoring site is because human body is used when startup is used And bend, when some sites bend overlong time, alarm module sends prompt message, points out correction;Monitoring is set between site Pulling force monitoring modular, degree of crook is determined by determining pulling force;Dc source sets charging structure.

It is the bending that human body back is judged using signal detection for above two scheme, after entering signal transacting, has Alarming device sends signal, reminds user's correction attitude.This method can more cultivate oneself of user relative to external force correction I realizes in attitude correction, but for how to be accurately detected bow-backed information, is not given clearly in above two scheme Explain.

Either wearable device or suspension arrangement, only accurately detecting bow-backed information can just be reminded, if inspection Measurement information is forbidden, and still sends alarm signal under normal sitting position, it will cause user constantly to adjust attitude, so that alarm disappears Remove, this attitude is likely to be a kind of sitting posture of mistake.In the course of time, user does not correct attitude not only, and one is formed on the contrary Incorrect sitting posture is planted, is run counter to desire.Therefore, the accuracy of signal detection turns into the key factor of the bow-backed alarm set of restriction.

The content of the invention

1. the invention technical problem to be solved

It is an object of the invention to overcome the shortcomings of to be difficult to the bow-backed signal of accurate detection in the prior art, there is provided Yi Zhongji In the bow-backed alarm set and based reminding method of muscle signals detection.Technical scheme, by the improvement to band structure, It is formed effectively cooperation when in use with muscle signals detection lug, improve the degree of accuracy of bow-backed detection, can be in time to making The bow-backed attitude of user is reminded.

2. technical scheme

To reach above-mentioned purpose, the technical scheme that the present invention is provided is:

A kind of bow-backed alarm set based on muscle signals detection of the invention, including positioned at the elastic suspenders of human body back With the band positioned at waist, elastic suspenders top leads to two shoulder belts, and the other end of shoulder belt is connected with band, the elasticity back of the body Muscle signals acquisition module, signal processing module and signal performing module, the muscle signals acquisition module and letter are provided with band Number performing module is electrically connected with signal processing module, and muscle signals acquisition module is distributed in elastic suspenders, and correspondence human body Latissimus dorsi region;The signal performing module is used to produce vibration or auditory tone cues to the bow-backed information for detecting.

As further improvement of the present invention, flexible tie belt is connected between two described shoulder belts.

Used as further improvement of the present invention, the shoulder belt is with band for integral type elastic webbing or by being clasped Connection is connected by VELCRO.

As further improvement of the present invention, the band be located at abdominal region at be provided with chest cup, the chest cup top with It is the distance between following to be more than band width.

Used as further improvement of the present invention, described elastic suspenders both sides are arc, are formed and are located at elastic suspenders two The wing band of side, the wing band is fitted with human body waist.

Used as further improvement of the present invention, the lower end of the elastic suspenders protrudes from the lower end of band, and its spacing is poor It is 0.5~1cm.

Used as further improvement of the present invention, the muscle signals acquisition module includes being symmetricly set on elastic suspenders two The muscle signals detection lug of side, muscle signals detection lug has the contact surface fitted with human body back.

As further improvement of the present invention, the signal processing module include preposition modulate circuit plate, capture card and Processor;The preposition modulate circuit plate is used to carry out denoising and amplification to the muscle signals for gathering, and by the signal after treatment It is sent in supporting capture card;

The capture card is used to for the simulation muscle signals for gathering to be converted into the recognizable digital muscle signals of processor, and Digital muscle signals are sent to processor;

The processor carries out feature extraction to muscle signals, and then carries out pattern-recognition, judges whether user is bow-backed, When hunchback is judged as, processor sends execution information, the action of signal performing module to signal performing module.

As further improvement of the present invention, switch is provided with the band, the switch is electrically connected with power module, Connecting line is located inside band.

A kind of bow-backed based reminding method based on muscle signals detection of the present invention, carries out hunchback and carries using above-mentioned alarm set Wake up, its process is:

(1) user is worn by bow-backed alarm set, startup power supply module, closing of circuit, system starts;

(2) the muscle signals detection lug for being placed in back starts to gather the muscle signals of latissimus dorsi, when the action of wearer back When, the muscle signals of latissimus dorsi can produce change, and then sending the signal to supporting preposition modulate circuit plate carries out signal Make an uproar and enhanced processing;

(3) capture card is processed the signal treated by front end circuit, and being changed into signal can be recognized by computer Digital muscle signals;Digital muscle signals are sent to the processor, and system judges now whether wearer is bow-backed;

(4) when system judges bow-backed, processor sends execution information, signal performing module to signal performing module Action, sends alerting signal.

3. beneficial effect

The technical scheme provided using the present invention, compared with prior art, is had the advantages that:

(1) a kind of bow-backed alarm set based on muscle signals detection of the invention, sets flesh message in elastic suspenders Number acquisition module, signal processing module and signal performing module, signal acquisition is carried out using muscle signals acquisition module, then to this Whether muscle signals carry out feature extraction and pattern-recognition bow-backed to judge wearer, and accuracy in detection is high, by individual sex differernce It is small with external environment influence, bow-backed prompting can be carried out to user accurately and in time.

(2) a kind of bow-backed alarm set based on muscle signals detection of the invention, table flesh is symmetrical arranged in waist both sides Message detection lug, can fully ensure that muscle signals detection lug is contacted with human body back, significantly by the wing band structure of arc Improve the degree of accuracy of detection.

(3) a kind of bow-backed alarm set based on muscle signals detection of the invention, band is engaged with shoulder belt, from two Direction is positioned, it is to avoid detecting electrode position offsets;And tie belt is set between two shoulder belts, can have by tie belt Effect prevents shoulder belt from being released to both sides;When hunchback is detected, the motor at back produces vibrations, and information can be transmitted accurately.

Brief description of the drawings

Fig. 1 is the structural representation of bow-backed alarm set of the invention;

Fig. 2 is the system schematic of bow-backed alarm set of the invention;

Fig. 3 is connection button shoulder harness structure schematic diagram in the present invention;

Fig. 4 reminds flow chart for bow-backed in the present invention.

Label declaration in schematic diagram:11st, elastic suspenders;12nd, wing band;13rd, band;14th, chest cup;15th, shoulder belt;16th, draw Ligament;21st, power module;22nd, switch;23rd, vibrator;24th, muscle signals detection lug.

Specific embodiment

To further appreciate that present disclosure, the present invention is described in detail in conjunction with the accompanying drawings and embodiments.

Surface electromyogram signal be on superficial muscular EMG and nerve cord electrical activity skin surface comprehensive effect, in reality In, because surface electromyogram signal lag time is short and strong antijamming capability, using relatively broad.But surface electromyogram signal is still Some shortcomings part is so there is, because EMG is generally very faint, the myoelectric information that surface electrode is detected is one group of muscle group Recruitment information, it is impossible to reflect that human brain, to the movement instruction of a certain action, can influence its accuracy in detection completely.

Compared with electromyographic signal, muscle signals still have accuracy rate higher in contraction of muscle fatigue.Due to muscle signals Propagation characteristic, sensor need not be fixed on some specific positions, it might even be possible to be not directly contacted with skin, and flesh sound Signal can also be collected in the tip of motion muscle.Muscle signals are influenceed small by Body surface impedance, small by the interference of external electric field, With accuracy in detection higher.With the utilization of muscle signals be combined bow-backed detection means by the present invention, for improving hunchback The degree of accuracy of signal detection.

Embodiment 1

With reference to Fig. 1, a kind of bow-backed alarm set based on muscle signals detection of the present embodiment, including positioned at human body back Elastic suspenders 11 and the band 13 positioned at waist, the top of elastic suspenders 11 leads to two shoulder belts 15, the other end of shoulder belt 15 It is connected with band 13, shoulder belt 15 is integral type elastic webbing with band 13.Elasticity can be carried on the back by elastic shoulder belt 11 and band 13 Band is stably fixed to back, it is to avoid it is shifted over.Shoulder belt and band can also be by snapping fit attachment, or by VELCRO It is connected, is not specifically limited.Shoulder belt 15 and band 13 are integral type structure in the present embodiment, therefore, both have to There is preferable elasticity, cannot otherwise dress.

Accordingly, flexible tie belt 16 is connected between two shoulder belts 15, the tie belt 16 correspondence is in human body back. When wearer jackknifes, bends over, back elastic suspenders pull-up, shoulder belt 15 is easily released, and the tie belt 16 for setting can laterally draw Firmly shoulder belt, prevents it from releasing, so as to for normal use of elastic suspenders 11 provides safeguard, be conducive to improving accuracy in detection.

With reference to Fig. 2, muscle signals acquisition module, signal processing module and signal performing module are provided with elastic suspenders 11, The muscle signals acquisition module and signal performing module are electrically connected with signal processing module, the distribution of muscle signals acquisition module In elastic suspenders 11, and correspondence human body latissimus dorsi region;The signal performing module is used to produce the bow-backed information for detecting Raw vibration or auditory tone cues.For the power module of system power supply may be provided in elastic suspenders 11, may also be arranged on band, do not have There is concrete restriction.

Wherein, signal processing module includes preposition modulate circuit plate, capture card and processor;The preposition modulate circuit plate For carrying out denoising and amplification to the muscle signals for gathering, and the signal after treatment is sent in supporting capture card;

Capture card is used to for the simulation muscle signals for gathering to be converted into the digital muscle signals that processor can recognize that, and by number Word muscle signals are sent to processor;Processor carries out feature extraction to muscle signals, and then carries out pattern-recognition, judges to use Whether person is bow-backed, and when hunchback is judged as, processor sends execution information to signal performing module, and signal performing module is acted, Produce corresponding vibration or auditory tone cues.Processor can be using wavelet transformation and VPMCD algorithms to hunchback and non-hunchback Two states are identified, and can be also identified using other algorithms.Can be using LabVIEW and matlab design muscle signals The programs such as collection, feature extraction, pattern-recognition and vibrating motor control/sound prompting.Above-mentioned LabVIEW and matlab programs Flesh sound detection field routine techniques is, is not being repeated.

It is well known that human body is in hunchback, the most substantially, chest muscle folds angle and has substantially for chest and back change Change, back can swell.Traditional detection method is also favorably detected that the method is more directly to flexes with electrode Angle is detected that such as the number of patent application 201410158064.4 of scheme 2 is the detection of external physical change, by the change on plate Shape test point discriminates whether hunchback, and the program can realize testing goal, but the degree of accuracy is inadequate.

Hunchback for user is reminded, and is easily to realize using vibration or sound prompting, and Major Difficulties are How bow-backed information is accurately collected.Detected using muscle signals acquisition module in the present invention, institute's acquisition information is beaten completely The limitation of conventional external testing equipment is broken, bow-backed differentiation can have been carried out according to the form of muscle, stretching change, relative to biography System technology tool is had large improvement.

And when signal acquisition is carried out using muscle signals acquisition module, chest and muscle of back can all be changed, and Body encloses a change the most substantially under straight and upright attitude and hunchbacked attitude pectus, and the point is ideal signal acquisition point. But found in specific experiment, if designing bow-backed signal acquisition point at the position, it is difficult to combined with wearable device, and sex Influence of the difference to device structure is very big, once detecting electrode piece is arranged in wearable device, detecting electrode piece is difficult to and people Body is preferably fitted.Even if can fit, when human body is in bow-backed state, because chest muscle is gathered, wearable device is grand Rise, cause laminating degree poor, cannot now be detected, it is difficult to reach the purpose of bow-backed prompting.

For these reasons, using latissimus dorsi as test point, when human body bends over hunchback, back clothing is drawn the present invention Tightly, can preferably be fitted with back, if muscle signals acquisition module to be arranged on the corresponding region of latissimus dorsi, be more conducive to carry High detection accuracy.The present invention is arranged on muscle signals acquisition module in elastic suspenders, and band is connected with the elastic suspenders And shoulder belt, band makes elastic suspenders be attached to back, and shoulder belt can then prevent elastic suspenders from being glided during physical activity, and Can prevent elastic suspenders from rotating, it is ensured that muscle signals acquisition module is corresponding with human body latissimus dorsi region, it is ensured that muscle signals are gathered Module is in monitoring point.

Embodiment 2

A kind of bow-backed alarm set based on muscle signals detection of the present embodiment, its basic structure is same as Example 1, Its difference is:As shown in figure 3, band 13 is connected by junction button 16 in the present embodiment.Muscle signals acquisition module bag The muscle signals detection lug 24 for being symmetricly set on the both sides of elastic suspenders 11 is included, muscle signals detection lug 24 is built with piezoelectric acceleration Sensor, muscle signals detection lug 24 has the contact surface fitted with human body back, the detection of internal piezoelectric acceleration sensor Face can fit with latissimus dorsi, and when human motion, muscle signals detection lug 24 can collect corresponding signal.

The bottom both sides of elastic suspenders 11 in the present embodiment are arc, form the wing band 12 positioned at the both sides of elastic suspenders 11, Wing band 12 is fitted with human body waist.The wing band 12 of arc can be combined fully with organization of human body, and wing band 12 is elastic strip.

The middle part of elastic suspenders 11 is provided with circular hole.Because vertebra protrudes higher, in the human body back of a bow, it is easy to make elastic suspenders Moved on 11, and after circular hole is set, can engaged with the vertebra position arched, reduced elastic suspenders 11 and shift.

Further, band 13 is located at abdominal region and is provided with chest cup 14, the chest 14 tops of cup with it is the distance between following More than the width of band 13, not less than distance between upper and lower side, its difference is 0~3cm to distance between 14 left and right both sides of chest cup, this Embodiment can be preferably 1cm differences.The lower end of elastic suspenders 11 protrudes from the lower end of band 13, and its spacing difference is 0.8cm. Experiment finds when using, if the lower end that the lower end of elastic suspenders 11 protrudes from band 13 flushes, lower edge can undertake main drawing Power, and upside position can swell, and cause detection inaccurate;If the lower end prominent length of elastic suspenders 11 is larger, outside clothing is received The pull-up influence of thing, can tilt lower edge, it is easier to elastic suspenders is misplaced.Therefore, its length range is arranged on 0.5~ There is preferable effect in the range of 1cm.

What deserves to be explained is, for more preferable Detection results, it is necessary to set chest cup on band.For Detection results Speech, as long as setting band and shoulder belt, it is already possible to realize detection and localization purpose, but the effect of chest cup is also extremely important.Many institutes Known, when human body is bended over, spinal joint is opened, and upper body clothes can be pulled up, even if drawing oneself up again, clothes is also required to multiple manually Position.Similarly, because being provided with shoulder belt in the present embodiment, when the human body back of a bow, due to the effect of shoulder belt, elastic suspenders brace by Pull-up, because latissimus dorsi opens, does not influence detection, but after body is straight and upright, muscle normally bounces back, and elastic suspenders can be stayed At top position, now internal muscle signals acquisition module just can not accurately fit with human body, cause subsequent detection inaccurate Really.

For above mentioned problem, after chest cup is set, because chest cup area is larger, Resistance can be formed on front side of body Domain, make band when human body is bended over will not upward sliding, after body is straight and upright, under the resilient force of band, elastic suspenders Can together recover to original position with muscle of back, it is ensured that will not be misplaced when subsequently using, substantially increased under lasting use state The bow-backed signal detection degree of accuracy.

Embodiment 3

A kind of bow-backed alarm set based on muscle signals detection of the present embodiment, its abridged edition structure is same as Example 2, Its difference is:The muscle signals detection lug 24 is arranged on the both sides of elastic suspenders 11, and wing band arc transition zone, 1 piezoelectric acceleration sensor is provided only with each muscle signals detection lug 24, each piezoelectric acceleration sensor forms one Individual signal sampling channel.Because the collection degree of accuracy of muscle signals is higher, weaker muscle signals can be perceived, if intersecting Collection signal is unfavorable for carrying for accuracy in detection, it is necessary to complex processor, can cause the disorder of signal transacting on the contrary It is high.Using vibrator as signal performing module, vibrator is located at the middle part of elastic suspenders 11, may be located at circular hole top or under Portion, can also be set around circular hole.

Embodiment 4

A kind of bow-backed alarm set based on muscle signals detection of the present embodiment, its abridged edition structure is same as Example 3, Closer:Switch 22 is set on band 13, and the switch 22 can be located on chest cup 14, and switch 22 is electrically connected with power module 21 Connect, connecting line is located inside band 13, it is easy to which user manipulates.

Embodiment 5

Using the bow-backed alarm set in embodiment 1~4, the invention provides a kind of bow-backed based reminding method, with reference to Fig. 4, Its process is:

(1) user is worn by bow-backed alarm set, after pressing shift knob, startup power supply module, closing of circuit, system Start working.

(2) the muscle signals detection lug for being placed in back starts to gather the muscle signals of latissimus dorsi, when the action of wearer back When, the muscle signals of latissimus dorsi can produce change, and then sending the signal to supporting preposition modulate circuit plate carries out signal Make an uproar and enhanced processing;

(3) and then signal is transferred to capture card, capture card is processed the signal treated by front end circuit, makes letter Number it is changed into the digital muscle signals that can be recognized by computer;Digital muscle signals are sent to the processor, by feature extraction, After the treatment of the program such as Feature Dimension Reduction and pattern-recognition, processor is bow-backed with non-hunchback using wavelet transformation and the identification of VPMCD algorithms Two states;

(4) when system judges bow-backed, processor sends execution information, signal performing module to signal performing module Action, sends alerting signal.By taking vibrator as an example, processor send vibration to instruct to vibrator, and vibrating motor receives vibration and refers to Vibration is produced after order, early warning is carried out to wearer, wearer answers straight and upright back after receiving the prompting.

Above-mentioned VPMCD methods are built upon the mode identification method on the basis of variable prediction model, in party's law theory, no Correlation between generic middle characteristic variable is also different.In the pattern recognition problem of human body muscle signals, using one Group characteristic quantity X=[X1,X2,···,Xp] a kind of classification of description, in different gesture classifications, one of characteristic value XiReceive Further feature value XjThe influence of (j ≠ i) is also different.That they exist each other is probably man-to-man relation X1=f (X2), It is also likely to be one-to-many relation X1=f (X2,X3,···).In order to recognize different classes of motor pattern, it is necessary to utilize Training sample set up can between expression characteristic variable internal relation Mathematical Modeling, i.e. variable prediction model (Variable Predictive Model,VPM).Test sample is predicted by these forecast models again, and it is flat with every class predicated error Square and minimum discriminant function is classified.

Above-mentioned processor is bow-backed with non-bow-backed two states using wavelet transformation and the identification of VPMCD algorithms, and its method is as follows:

S1, right lateral bend four kind waists action collecting samples curved to the straight back of the body, the back of a bow, left side, obtain pre- using VPMCD algorithms Survey model VPMtraining;Forecast model VPMtrainingIt is the pre- program for first passing through test storage in system, as original With type of action;

S2, electrode slice carry out sEMG signal acquisitions to waist action;

S3, the sEMG signals to gathering carry out three layers of wavelet decomposition using wavelet basis function, obtain the wavelet systems of each frequency range Number.Shown in wavelet transformation implication formula such as formula (1):

Wherein:f(t)∈L2(R), L2(R) it is the function space of quadractically integrable function composition on R.ψ (t) is mother wavelet or mother Small echo, a is scale factor, and b is shift factor.

Again the signal is represented with the linear combination of different scale and different shifted wavelet bases;As shown in formula (2):

Wherein, the coefficient of each single item is known as wavelet coefficient, AI(t)、DiT () represents I grades of low frequency component and height respectively Frequency component, waI、wdiThe low-frequency wavelet coefficients and high-frequency wavelet coefficient of i-stage are represented respectively;

The wavelet coefficient modulus maximum of each frequency sub-band is used as characteristic vector after S4, extraction wavelet decomposition.So obtain every kind of The eigenvalue matrix of classification, can be using the internal association between cross-correlation analysis method checking characteristic value during experiment;

S5, with the forecast model VPM for having trainedtrainingClassification and Identification, various types of backs are carried out to eigenvalue matrix Action can be just determined by the data result of VPMCD graders.

It is bow-backed with non-bow-backed two states using wavelet transformation and the identification of VPMCD algorithms, have relative to other recognition methods There is accuracy higher, the method sets up variable prediction model by the method for parameter Estimation, it is to avoid ANN structures, type Select permeability and iterative process.In addition, in the selection of types of models and parameter, it depends on the minimum of predicted value and actual value Prediction sum squares, so as to avoid the dependence to priori, more objectivity.Therefore, the method has more preferable VPMCD Training Capabilities and classification performance, are a kind of effective, reliable muscle signals mode identification methods.

The present invention carries out signal acquisition using muscle signals acquisition module, feature extraction is carried out to the electric signal and pattern is known Whether not bow-backed to judge wearer, accuracy in detection is high, small by individual sex differernce and external environment influence, can in time, it is accurate Bow-backed prompting really is carried out to user.

Schematical above that the present invention and embodiments thereof are described, the description does not have restricted, institute in accompanying drawing What is shown is also one of embodiments of the present invention, and actual structure is not limited thereto.So, if the common skill of this area Art personnel enlightened by it, in the case where the invention objective is not departed from, is designed and the technical scheme without creative Similar frame mode and embodiment, all should belong to protection scope of the present invention.

Claims (10)

1. a kind of bow-backed alarm set based on muscle signals detection, including elastic suspenders (11) positioned at human body back and it is located at The band (13) of waist, it is characterised in that:Elastic suspenders (11) top leads to two shoulder belts (15), the other end of shoulder belt (15) It is connected with band (13), muscle signals acquisition module, signal processing module and signal is provided with the elastic suspenders (11) and is performed Module, the muscle signals acquisition module and signal performing module are electrically connected with signal processing module, muscle signals collection mould Block is distributed in elastic suspenders (11), and correspondence human body latissimus dorsi region;The signal performing module is used for the camel to detecting Breach of faith breath produces vibration or auditory tone cues.
2. it is according to claim 1 it is a kind of based on muscle signals detection bow-backed alarm set, it is characterised in that:Described Flexible tie belt (16) is connected between two shoulder belts (15).
3. it is according to claim 1 it is a kind of based on muscle signals detection bow-backed alarm set, it is characterised in that:The shoulder Band (15) is with band (13) for integral type elastic webbing or by snapping fit attachment or being connected by VELCRO.
4. according to claim 1 or 2 or 3 it is a kind of based on muscle signals detection bow-backed alarm set, it is characterised in that: The band (13) is provided with chest cup (14) at abdominal region, and chest cup (14) top is more than band with the distance between following (13) width.
5. it is according to claim 4 it is a kind of based on muscle signals detection bow-backed alarm set, it is characterised in that:Described Elastic suspenders (11) both sides are arc, form the wing band (12) positioned at elastic suspenders (11) both sides, the wing band (12) and human lumbar Fit in portion.
6. it is according to claim 5 it is a kind of based on muscle signals detection bow-backed alarm set, it is characterised in that:The bullet Property brace (11) lower end protrude from the lower end of band (13), its spacing difference is 0.5~1cm.
7. it is according to claim 5 it is a kind of based on muscle signals detection bow-backed alarm set, it is characterised in that:The flesh Sound signal acquisition module includes being symmetricly set on the muscle signals detection lug (24) of elastic suspenders (11) both sides, muscle signals detection Piece (24) is with the contact surface fitted with human body back.
8. according to claim 1 or 7 it is a kind of based on muscle signals detection bow-backed alarm set, it is characterised in that:Institute Stating signal processing module includes preposition modulate circuit plate, capture card and processor;The preposition modulate circuit plate is used for collection Muscle signals carry out denoising and amplification, and the signal after treatment is sent in supporting capture card;
The capture card is used to for the simulation muscle signals for gathering to be converted into the digital muscle signals that processor can recognize that, and by number Word muscle signals are sent to processor;
The processor carries out feature extraction to muscle signals, and then carries out pattern-recognition, judges whether user is bow-backed, when sentencing Break during for hunchback, processor sends execution information, the action of signal performing module to signal performing module.
9. it is according to claim 8 it is a kind of based on muscle signals detection bow-backed alarm set, it is characterised in that:The beam Switch (22) is provided with band (13), the switch (22) is electrically connected with power module (21), it is internal that connecting line is located at band (13).
10. a kind of bow-backed based reminding method based on muscle signals detection, is entered using the alarm set as described in claim 1~9 Row is bow-backed to be reminded, it is characterised in that its process is:
(1) user is worn by bow-backed alarm set, startup power supply module, closing of circuit, system starts;
(2) the muscle signals detection lug for being placed in back starts to gather the muscle signals of latissimus dorsi, when wearer back acts, the back of the body The muscle signals of platysma can produce change, and then sending the signal to supporting preposition modulate circuit plate carries out signal denoising and put Big treatment;
(3) capture card is processed the signal treated by front end circuit, signal is changed into the number that can be recognized by computer Word muscle signals;Digital muscle signals are sent to the processor, and system judges now whether wearer is bow-backed;
(4) when system judges bow-backed, processor sends execution information to signal performing module, and signal performing module is moved Make, send alerting signal.
CN201710154080.XA 2017-03-15 2017-03-15 A kind of bow-backed alarm set and based reminding method based on muscle signals detection CN106821390A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710154080.XA CN106821390A (en) 2017-03-15 2017-03-15 A kind of bow-backed alarm set and based reminding method based on muscle signals detection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710154080.XA CN106821390A (en) 2017-03-15 2017-03-15 A kind of bow-backed alarm set and based reminding method based on muscle signals detection

Publications (1)

Publication Number Publication Date
CN106821390A true CN106821390A (en) 2017-06-13

Family

ID=59144983

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710154080.XA CN106821390A (en) 2017-03-15 2017-03-15 A kind of bow-backed alarm set and based reminding method based on muscle signals detection

Country Status (1)

Country Link
CN (1) CN106821390A (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5827209A (en) * 1994-09-23 1998-10-27 Bcam International, Inc. Intelligent body support
CN2639967Y (en) * 2003-09-02 2004-09-08 刘正德 Posture alarming apparatus for preventing hunchback
KR100661305B1 (en) * 2005-11-09 2006-12-19 인제대학교 산학협력단 An economy in power consumption sensor circuit of a vertebra curve-symptom correction appointment surveillance device
CN2868245Y (en) * 2005-09-27 2007-02-14 胡建庄 Sitting posture prompting corrector for preventing myopia, hunchback and sleeping
AU2007238017A1 (en) * 2006-04-13 2007-10-25 Aecc Enterprises Limited Method for imaging the motion of joints
CN102426651A (en) * 2011-08-25 2012-04-25 武汉理工大学 Human body forearm surface electromyogram signal acquisition and pattern recognition system
CN103488995A (en) * 2013-08-31 2014-01-01 中山大学 Method for identifying rotation of neck
CN205751175U (en) * 2016-05-19 2016-11-30 重庆医药高等专科学校 A kind of pose reminding device
CN106725466A (en) * 2016-12-31 2017-05-31 安徽工业大学 A kind of bow-backed alarm set and based reminding method based on surface myoelectric technology

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5827209A (en) * 1994-09-23 1998-10-27 Bcam International, Inc. Intelligent body support
CN2639967Y (en) * 2003-09-02 2004-09-08 刘正德 Posture alarming apparatus for preventing hunchback
CN2868245Y (en) * 2005-09-27 2007-02-14 胡建庄 Sitting posture prompting corrector for preventing myopia, hunchback and sleeping
KR100661305B1 (en) * 2005-11-09 2006-12-19 인제대학교 산학협력단 An economy in power consumption sensor circuit of a vertebra curve-symptom correction appointment surveillance device
AU2007238017A1 (en) * 2006-04-13 2007-10-25 Aecc Enterprises Limited Method for imaging the motion of joints
CN102426651A (en) * 2011-08-25 2012-04-25 武汉理工大学 Human body forearm surface electromyogram signal acquisition and pattern recognition system
CN103488995A (en) * 2013-08-31 2014-01-01 中山大学 Method for identifying rotation of neck
CN205751175U (en) * 2016-05-19 2016-11-30 重庆医药高等专科学校 A kind of pose reminding device
CN106725466A (en) * 2016-12-31 2017-05-31 安徽工业大学 A kind of bow-backed alarm set and based reminding method based on surface myoelectric technology

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
张旭: "基于表面肌电信号的人体动作识别与交互", 《万方数据》 *
魏伟: "基于小波变换和PSO-SVM的表面肌动作模式分类", 《安徽工业大学学报(自然科学版)》 *

Similar Documents

Publication Publication Date Title
Atallah et al. Sensor positioning for activity recognition using wearable accelerometers
US8369936B2 (en) Wearable apparatus for measuring heart-related parameters and deriving human status parameters from sensed physiological and contextual parameters
Tapia et al. Real-time recognition of physical activities and their intensities using wireless accelerometers and a heart rate monitor
US6984208B2 (en) Method and apparatus for sensing body gesture, posture and movement
JP3733133B2 (en) Sleep state estimation device
Shany et al. Sensors-based wearable systems for monitoring of human movement and falls
Atallah et al. Sensor placement for activity detection using wearable accelerometers
US20070179396A1 (en) Method and System for Detecting and Classifying Facial Muscle Movements
Di Rienzo et al. MagIC system: A new textile-based wearable device for biological signal monitoring. Applicability in daily life and clinical setting
Mattmann et al. Recognizing upper body postures using textile strain sensors
JP2011120871A (en) Method and system for monitoring physiological and athletic performance characteristics of subject
JP3829197B2 (en) Auscultation education equipment
TWI393579B (en) The state of the muscle movement state analysis system, methods and computer program products
Cheng et al. Active capacitive sensing: Exploring a new wearable sensing modality for activity recognition
US20040167420A1 (en) Apparatus and method for analyzing motions using bio-impedance
EP2289410A1 (en) Noninvasive method and system for monitoring physiological characteristics
Kalantarian et al. Monitoring eating habits using a piezoelectric sensor-based necklace
JP5841320B2 (en) Non-invasive methods and systems for monitoring physiological characteristics and exercise performance
EP2068704A2 (en) Bio-mechanical sensor system
Khan et al. Accelerometer signal-based human activity recognition using augmented autoregressive model coefficients and artificial neural nets
Singh et al. Evaluation of Electrocardiogram for Biometric Authentication.
US10335080B2 (en) Biomechanical activity monitoring
Shen et al. Implementation of a one-lead ECG human identification system on a normal population
US8246555B2 (en) Method and system for monitoring sport related fitness by estimating muscle power and joint force of limbs
Bonato et al. Data mining of motor patterns recorded with wearable technology

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