CN107684430A - Correcting device and its application method are detected based on Curie modules human body attitude - Google Patents
Correcting device and its application method are detected based on Curie modules human body attitude Download PDFInfo
- Publication number
- CN107684430A CN107684430A CN201710911992.7A CN201710911992A CN107684430A CN 107684430 A CN107684430 A CN 107684430A CN 201710911992 A CN201710911992 A CN 201710911992A CN 107684430 A CN107684430 A CN 107684430A
- Authority
- CN
- China
- Prior art keywords
- state
- posture
- curie
- circulation
- data
- 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.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/1116—Determining posture transitions
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/7455—Details of notification to user or communication with user or patient ; user input means characterised by tactile indication, e.g. vibration or electrical stimulation
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Physiology (AREA)
- Artificial Intelligence (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Dentistry (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention is related to human body attitude monitoring technical field based on Curie modules human body attitude detection correcting device and its application method.The present invention is corrected gesture module (2) and posture wrong identification microswitch (3) composition, connected in a manner of electric signal, App (4) real-time display is detected by human body attitude by main controller (1)；Wherein, gesture module (2) is corrected, microswitch (3) is connected with main controller (1)；Mobile phone human body attitude detection App (4) carries bluetooth by main controller (1), can the change of Real time dynamic display human body attitude.Apparatus of the present invention：With learning functionality, user's individuation data is gathered, dress ornament can be freely defined, freely define design, be adapted to various crowd demands.There is device interconnection to realize, and user oneself data can contrast with more personal datas, and data upload high in the clouds and obtain more precisely correcting analysis.Device is health industry, and correcting the market demand specifically for human posture provides technology material base.
Description
Technical field
The present invention relates to human body attitude monitoring technical field, refers specifically to a kind of human body based on Curie module intelligence learnings
Attitude detection correcting device and its application method.
Background technology
Modern does not often pay attention to station, the specification of sitting posture, and student, office's white collar all less pay attention to sitting posture.Body is twisted,
The sitting posture of back of a bow extension head, the problems such as neck lumbar muscle strain, spinal curvature can be caused.Also many young men working when curved the back of the body, stretch
Neck and see display, before being sitting in computer screen for a long time, this can cause cervical vertebra, shoulder anteflexion, oppress spinal nerve, induce serious
Waist, the back of the body, cervical vertebra ache.In addition, such as askew body, which is stood, neck is bent containing chest, both shoulders droop can cause lumbar vertebrae both sides discontinuity
Cause pain in back and loin, result even in the cervical vertebra/lumbar diseases such as cervical vertebrae hyperosteogeny, disc degeneration, long-time holding station looks are easy
Fatigue.
Under information age background, people often learn by oneself dancing by the way of by video, books or carry out body
Training.Many young white collars can use KEEP softwares to carry out body-building, but be entered during self-fitness according to the voice message in system
Row improves posture, and which is not prompted by user's real time status, is simply carried out systematization prompting, is not accomplished and user
Real-time interactive is simultaneously corrected.This mode can not accomplish real-time interactive, and action norm is difficult to ensure that.Body coach carries out one-to-many religion
Class hour, it is difficult to ensure the specification of the fine movement of each individual.
At the same time, the innovative point of human posture's detection means based on acceleration transducer emerges in an endless stream with product, full
The foot demand of every aspect in life --- robot, UAV Attitude detection, inertial navigation, virtual reality device, old man
Falling detection device etc..
Therewith, the research of human posture's correcting device also reaches its maturity, most based on acceleration transducer and gyroscope
Principle, commercially circulated in a manner of commercialization.Several applications generally used are as follows：Posture corrective belt, supra-aural sitting posture reminder,
Sitting posture correction function instrument.Its feature is：Current most of postures correction products concentrate on station, sitting posture correction function, and majority and belong to pressure
Correct, do not account for individual difference.Including the use of Microsoft Kinect, i.e. Xbox sensors identify appearance by catching bone node
Gesture whether standard.But this mode has no idea to realize that real-time interactive is corrected, only know at that time correct set whether.Its defect
It is that device is larger, wearable product can not be formed.All this kind, it is a kind of by way of device intelligence study, it is familiar with using
The personal physiology body data of person, carry out effectively adjusting and correcting the more personalized technique device for sitting stance, turning into market needs
The focus asked.
The content of the invention
It is an object of the invention to overcome missing existing for prior art and deficiency, propose that one kind is learnt by device intelligence
Mode, the action of real-time combined standard and it is personalized is familiar with user individual's sitting custom, physiology body data, carry out effective
Regulation and posture correcting device.
A kind of Curie modules human body attitude that is based on of the invention detects correcting device, and its feature is, by main controller, corrects
Gesture module and posture wrong identification microswitch composition, are connected in a manner of electric signal, and detect App by human body attitude shows in real time
Show；
Wherein, gesture module, microswitch and main controller is corrected to connect；Main controller carries bluetooth, can be in the people of mobile phone
Body attitude detection App Real time dynamic display attitudes vibrations.
The main controller for support neural network algorithm the wearable module Curie Nano of Curie, or include Curie can
Dress the development board Arduino101/Genuino101 of gesture recognition module.
The correction gesture module is vibrating sensor, such as vibrating reed.
The microswitch, such as touch a little or touch-switch.
The human body attitude detection App is user interface, wherein, one is homepage, and display current state makes with this
Continuous state changes；Another is the historical record page, the correction state change and posture accuracy rate that display history uses；
Local data is stored among Flash, and legacy data in Flash can be removed by the C keys of interactive interface.
A kind of application method based on Curie modules human body attitude detection correcting device of the present invention, it is based on
The Curie modules posture study of Arduino1.6.12 environment, and following three kinds of serial ports monitoring data are set：
Setting normal condition is motion category 1；
Setting fair state is motion category 2；
It is motion category 3 to set poor state；
When main controller is identified as normal condition, the normal remains stationary of vibrating reed of gesture module is corrected；
When main controller is identified as fair state, the vibrating reed for correcting gesture module slightly vibrates；
It is to correct the vibrating reed judder of gesture module when main controller is identified as poor state；
Wherein, including sitting posture correction function, stance walk appearance correction, and body-building posture is corrected, and continuous action is corrected.
Innovative point of the present invention is to gather user's individuation data；Device has learning functionality, and usage time is longer, more
Adapt to personal.At the same time, device also has interconnection consciousness, and user oneself data can contrast with more personal datas, in data
High in the clouds is passed to obtain more precisely correcting analysis.Data are uploaded to mobile phone by bluetooth, data can be carried out using neuroid
Optimization etc..Commercialization is realized by device, big data can be utilized to realize more accurately posture analysis, by a large number of users appearance
The data analysis of gesture information, a large amount of useful commercial market information can be excavated.
Brief description of the drawings
Fig. 1 detects correcting device structural schematic block diagram for the present invention based on Curie modules human body attitude；
Fig. 2 is the sitting posture program flow diagram of correcting device embodiment of the present invention；
Fig. 3 is the sitting posture apparatus structure schematic diagram of correcting device embodiment of the present invention；
Fig. 4 is the sitting posture installation optimization structural representation of correcting device embodiment of the present invention；
Fig. 5 is that the sitting posture of correcting device embodiment of the present invention learns human body attitude schematic diagram；
Fig. 6 is that the stance of correcting device embodiment of the present invention walks appearance study human body attitude schematic diagram；
Fig. 7 is the body-building operation program flow chart of correcting device embodiment of the present invention；
Fig. 8 is the body-building movement device structural representation of correcting device embodiment of the present invention；
Fig. 9 is that the body-building movement device of correcting device embodiment of the present invention optimizes structural representation；
Figure 10 is the body-building action learning human body attitude schematic diagram of correcting device embodiment of the present invention；
Figure 11 is correcting device embodiment continuous action correcting device frontal design figure of the present invention；
Figure 12 is correcting device embodiment continuous action correcting device back side design drawing of the present invention；
Figure 13 is that correcting device embodiment human body attitude of the present invention detects App display interfaces；
Figure 14 is that correcting device embodiment human body attitude of the present invention detects App homepages display example；
Figure 15 is that correcting device embodiment human body attitude of the present invention detects App historical records page display example.
Embodiment
Below in conjunction with drawings and examples, the invention will be further described.
Embodiment, the product of Curie modules is used using T-shirt as correcting device structure (as shown in Figure 1)
Arduino101/Genuino101, the study correcting device of standard operation is CurieNano modules.
The main controller 1 used in embodiment is that the Genuino 101 released in recent years learns development board.
It is based on Intel Curie modules, is integrated with low-power consumption bluetooth (BLE) and six axle attitude transducer (IMU) functions,
Machine learning operation is carried out by Curie.
Neural network algorithm has been used inside it, optimal solution can have been obtained by autonomous learning.
A. sitting posture correction function
The sitting posture correction function, using development board Arduino101, it is measured and entangles by the posture of following three kinds of different conditions
Just：
First is normal condition, and now vibrating reed does not vibrate, and user's back is straight and upright；The state by doctor or can entangle
The director of right sitting position provides initially training data, it is ensured that the setting of first state is correct.
Second is fair state, and now vibrating reed slightly vibrates, and user's back is slightly curved, in unconscious absent-mindedness, Ke Yiyou
Itself or people at one's side provide training auxiliary and obtain data, it is ensured that in the range of being arranged on of second state.
Third, is poor state, and now vibrating reed judder, user's back degree of crook is big, in complete absent-mindedness, is changed
State individual can voluntarily train, and scope is big, and accuracy requirement is low.
When user present second, the third two states posture when, by correct gesture module vibrating reed vibrating alert its
Correcting sitting postures.
Its program circuit (as shown in Figure 2)：
After gesture recognition starts, by including to obtain Curie 6 axle accelerations of wearable gesture recognition module acquisition in main controller
Sensing data；
If receiving study instruction, learn corresponding sitting posture according to learning to instruct, continue to obtain 6 axle accelerations biography after study
Sensor data, into circulation；
If not receiving study instruction, the 6 axle acceleration sensor data acquired are passed through into neutral net point
Class；
If being identified as standard sitting posture, continue to obtain 6 axle acceleration sensor data, into circulation；
If being identified as nonstandard sitting posture, continue to identify；
If being identified as second state sitting posture, the vibration of vibrating reed low-intensity；
If there is microswitch signal in 5 seconds, learn again using this action data as correct sitting posture, then continue to
6 axle acceleration sensor data are obtained, into circulation；
If continued to obtain 6 axle acceleration sensor data, into circulation without microswitch signal in 5 seconds；
If it is not second state sitting posture to be identified as, vibrating reed high-intensity oscillation；
If there is microswitch signal in 5 seconds, learn again using this action data as correct sitting posture, then continue to
6 axle acceleration sensor data are obtained, into circulation；
If continued to obtain 6 axle acceleration sensor data, into circulation without microswitch signal in 5 seconds；
Finally, deenergization, jumps out circulation, and correction terminates.
Sitting posture correction function can be thus set to obtain most true scope.
Curie is trained to carry out by the accelerometer built in Curie, gyroscope and other external sensing datas
Pattern classification.Equipment can be classified in real time to new pattern simultaneously, and report new state or abnormality.
The training data of backup hardware inside neurons or training data is transplanted to others when uploading to high in the clouds
On Curie modules, huge database will be formed.
Action recognition sensor uses Curie Nano modules, and the sensor is wrong to the gesture recognition of correcting device
By mistake, new pattern is classified in real time, and report new state or abnormality.
Sitting posture correction function apparatus structure (as shown in Figure 3).
In order to be designed to finished product attractive in appearance, trace arrangements inside clothes, Curie or CurieNano modules are used
Carry out gesture recognition.
It is preferred that sitting posture correction function device (as shown in Figure 4), uses Arduino101 development boards, touch-switch generation is used
For touching a little, trace arrangements are outside clothing.Test excellent.
Sitting posture study human body attitude (as shown in Figure 5).
B. stance walks appearance correction
The stance walks appearance correction, using development board Arduino101, is surveyed by the posture of following three kinds of different conditions
It is fixed to correct：
First is normal condition, and now vibrating reed does not vibrate, and user's back is straight and upright；The state by doctor or can entangle
The director that positive stance walks appearance provides initially training data, it is ensured that the setting of state first is correct；
Second is fair state, and now vibrating reed slightly vibrates, and user's back is slightly curved, in unconscious absent-mindedness, Ke Yiyou
Itself or people at one's side provide training auxiliary and obtain data, it is ensured that in the range of being arranged on of state second；
Third, is poor state, and now vibrating reed judder, user's back degree of crook is big, in complete absent-mindedness, is changed
State individual can voluntarily train, and scope is big, and accuracy requirement is low；
When user present second, the third two states posture when, by correct gesture module vibrating reed vibrating alert its
Correct stance and walk appearance；
If any identification mistake, it is normal condition that can press microswitch by wrong state recognition is sorted out；
Stance, walk three kinds of states (as shown in Figure 6) that appearance corrects state
Its program circuit is as follows：
After gesture recognition starts, by including to obtain Curie 6 axle accelerations of wearable gesture recognition module acquisition in main controller
Sensing data；
If receiving study instruction, appearance is walked according to the corresponding stance of instruction study is learnt, continues to obtain the acceleration of 6 axles after study
Sensing data is spent, into circulation；
If not receiving study instruction, the 6 axle acceleration sensor data acquired are passed through into neutral net point
Class；
If the standard stance of being identified as walks appearance, continue to obtain 6 axle acceleration sensor data, into circulation；
If being identified as nonstandard stance walks appearance, continue to identify；
If being identified as second state stance walks appearance, the vibration of vibrating reed low-intensity；
If there is microswitch signal in 5 seconds, this action data is walked appearance as correct stance and learnt again, then
Continue to obtain 6 axle acceleration sensor data, into circulation；
If continued to obtain 6 axle acceleration sensor data, into circulation without microswitch signal in 5 seconds；
If it is not that second state stance walks appearance, vibrating reed high-intensity oscillation to be identified as；
If there is microswitch signal in 5 seconds, this action data is walked appearance as correct stance and learnt again, then
Continue to obtain 6 axle acceleration sensor data, into circulation；
If continued to obtain 6 axle acceleration sensor data, into circulation without microswitch signal in 5 seconds；
Finally, deenergization, jumps out circulation, and correction terminates.
C. body-building posture is corrected
The body-building posture is corrected, such as the correction of push-up body-building action, uses Curie or CurieNano modules
It is more compact, posture is more delicately identified, is adapted to wearable.Gesture recognition mistake is corrected using microswitch；
Wherein, microswitch 3, such as touch a little, be more comfortable, be adapted to human body.
Body-building posture corrects flow (as shown in Figure 7)：
After gesture recognition starts, by including to obtain Curie 6 axle accelerations of wearable gesture recognition module acquisition in main controller
Sensing data；
If being identified as standard state, continue to obtain 6 axle acceleration sensor data, into circulation；
If being identified as poor state, vibrating reed vibration 0.5s, continue to obtain 6 axle acceleration sensor data, enter
Circulation；
Finally, deenergization, jumps out circulation, and correction terminates.
Body-building posture correcting device structure (as Figure 8,9)
The present invention using T-shirt as exemplary device, in order to be designed to finished product attractive in appearance, will trace arrangements inside clothes,
Gesture recognition is carried out using Curie or CurieNano modules.
Two states (as shown in Figure 10) when body-building posture learns
First is specification condition, and now vibrating reed does not vibrate, user's action norm.The state can be by doctor or correction
The director of sitting posture provides initially training data, it is ensured that the setting of state 1 is correct.
Second is state lack of standardization, and now vibrating reed slightly vibrates, and user's action is lack of standardization, can be with relaxed state
People by itself or at one's side provides training auxiliary and obtains data, it is ensured that in the range of being arranged on of state second.
From two kinds of posture is obtained, data are perfect during, improve the accuracy of gesture recognition.Pass through
Vibration module is connected, when second, the third two kinds of postures is presented in user, it will vibration walks appearance to remind it to correct station.
If any identification mistake, touch-switch can be pressed, is specification condition by wrong state recognition is sorted out.
D. continuous action is corrected
Continuous action correcting device is conceived to the correction of series of actions, and the correcting device is applied to setting-up exercises to music, and body-building is moved
Work or dance self-learning, it is applied widely.
Continuous action correcting device structure (as shown in accompanying drawing 11,12)
The vibrating reed of continuous action correcting device and the number of identifier are depending on demands of individuals and apparatus function.This hair
It is bright to be acted with bowing as example.
Its program circuit is as follows：
After gesture recognition starts, the data of last time study are obtained from Flash, initialize Bluetooth function；
By including to obtain Curie 6 axle acceleration sensor data of wearable gesture recognition module acquisition in main controller；
If receiving study instruction, learn corresponding continuous action according to learning to instruct, continue to obtain the acceleration of 6 axles after study
Sensing data is spent, into circulation；
If not receiving study instruction, the 6 axle acceleration sensor data acquired are passed through into neutral net point
Class；
If identification obtains current pose as the compulsory exercise after a upper posture, continue to obtain 6 axle accelerations sensing
Device data, into circulation；
If it is not the compulsory exercise after a upper posture that identification, which obtains current pose, vibrating reed vibration 0.5s, continue
6 axle acceleration sensor data are obtained, into circulation；
Finally, deenergization, jumps out circulation, and correction terminates.
E. human body attitude detection App display interfaces (as shown in Figure 13)
Detect that App graphic interfaces show to obtain real-time status, that historic state records is accurate with posture by human body attitude
Rate, has interactive, and user can be made to become apparent from visual understanding to oneself current posture state and history service condition,
Local data is stored among Flash, and legacy data in Flash is removed by the C keys of interactive interface.
Human body attitude detection App is broadly divided into two pages, and one is homepage, and display current state uses with this
Continuous state changes.Five kinds of states of A, B, C, D, E are provided that at present to realize correction function that this problem is shown.Another
For the historical record page, correction state change and posture accuracy rate that history n times use, the appearance that the page is shown can be at most shown
State accuracy rate is that the track registration homepage that state is corrected for continuous action is shown, shape when example is judges that appearance is sat in station
State.A condition is normal condition, and B state is general state, and C-state is poor state.A, B, C, D, E are respectively corresponding five
The state of operating position section.
Example (as shown in Figure 14)
Left figure is that homepage is shown, state when example is judges that appearance is sat in station.A condition is normal condition, B state one
As state, C-state is poor state.
Right figure be the historical record page show, display nearest ten times correct states, and with this judge with standard state or
The track registration that state is corrected for continuous action is shown in the coincidence factor of track, the page, is for the first time 54%, second
For 83%.A, B, C, D, E are respectively corresponding five continuous action position sections.This state simulation is at dance movement wrist
One eight repetition clapped and motion process.(as shown in Figure 15)
In summary, it is of the invention that correcting device and its application method are detected based on Curie modules human body attitude, there is master
Want feature as follows：
Firstth, the principle of acceleration transducer and gyroscope is not based only on, but is used and supported neural network algorithm
The wearable modules of Curie, neural network algorithm is supported, replace programming with training, there is self-learning function.
Secondth, using T-shirt as embodiment (device), for station, sitting posture correction function, walk appearance correction, body-building posture correct with-even
Continuous action correction is corrected.Device characteristic is can be real-time by acousto-optic or vibrating effect and people's polyesthesia (tactile vision)
It is interactive.It is the personalized posture correcting device of a action of combined standard in real time and individual physiological custom.Simultaneously can be by upper
The data measured are passed to internet, form big database.
3rd, the present invention has used neural network algorithm in the intelligence learning characteristic of automatic field, and makes human posture
Correcting device is more personalized, is generally applicable to different crowd, obtains more accurately and effectively correcting.And examined by human body attitude
Survey App graphic interfaces to show, obtain the posture classification of user under current state, this posture Classification Change process used
And the record of history service condition and posture accuracy rate, classified by A, B, C-state.
4th, device can freely define dress ornament, carry out self-definition design, be adapted to various crowd demands, and application prospect is wide
It is general.For health industry, correct the market demand specifically for human posture and technology material base is provided.
Claims (7)
1. one kind is based on Curie modules human body attitude detection correcting device, it is characterised in that by main controller (1), corrects posture mould
Block (2) and gesture recognition mistake microswitch (3) composition, are connected in a manner of electric signal, and it is real to detect App (4) by human body attitude
When show；
Wherein, gesture module (2) is corrected, microswitch (3) is connected with main controller (1)；Mobile phone human body attitude detection App (4) is logical
Cross main controller (1) and carry bluetooth, can Real time dynamic display attitudes vibration.
It is 2. as claimed in claim 1 a kind of based on Curie modules human body attitude detection correcting device, it is characterised in that described
Main controller (1) is the wearable module Curie Nano of Curie for supporting neural network algorithm, or includes the wearable postures of Curie
The development board Arduino101/Genuino101 of identification module；
The correction gesture module (2) is vibrating sensor, such as vibrating reed；
The microswitch (3) is touches a little or touch-switch；
The human body attitude detection App (4) is user interface, wherein, one is homepage, display current state and this use
Continuous state change；Another is the historical record page, the correction state change and posture accuracy rate that display history uses；This
Ground data storage can remove legacy data in Flash among Flash, by the C keys of interactive interface.
3. a kind of application method based on Curie modules human body attitude detection correcting device, it is characterised in that it is based on
The Curie modules posture study of Arduino1.6.12 environment, and following three kinds of serial ports monitoring data are set：
Setting normal condition is motion category 1；
Setting fair state is motion category 2；
It is motion category 3 to set poor state；
When main controller is identified as normal condition, the vibrating reed for correcting gesture module does not vibrate；
When main controller is identified as fair state, the vibrating reed for correcting gesture module slightly vibrates；
It is to correct the vibrating reed judder of gesture module when main controller is identified as poor state；
Wherein, including sitting posture correction function, stance walk appearance correction, and body-building posture is corrected, and continuous action is corrected.
4. a kind of application method based on Curie modules human body attitude detection correcting device as claimed in claim 3, its feature
It is, the sitting posture correction function, using development board Arduino101, correction is measured by the posture of following three kinds of different conditions：
First is normal condition, and now vibrating reed does not vibrate, and user's back is straight and upright；The state can be sat by doctor or correction
The director of appearance provides initially training data, it is ensured that the setting of state first is correct；
Second is fair state, and now vibrating reed slightly vibrates, and user's back is slightly curved, can be by itself in unconscious absent-mindedness
Or people at one's side provides training auxiliary and obtains data, it is ensured that in the range of being arranged on of state second；
Third, is poor state, and now vibrating reed judder, user's back degree of crook is big, in complete absent-mindedness, changes state
Individual can voluntarily train, and scope is big, and accuracy requirement is low；
When user present second, the third two states posture when, by correct gesture module vibrating reed vibrating alert its correction
Sitting posture；
Its program circuit is as follows：
After gesture recognition starts, by including to obtain Curie wearable gesture recognition module acquisition 6 axle accelerations sensings in main controller
Device data；
If receiving study instruction, the corresponding sitting posture of study is instructed according to study, continues to obtain 6 axle acceleration sensors after study
Data, into circulation；
If not receiving study instruction, the 6 axle acceleration sensor data acquired are passed through into neural network classification；
If being identified as standard sitting posture, continue to obtain 6 axle acceleration sensor data, into circulation；
If it is not standard sitting posture to be identified as, continue to identify；
If being identified as medium sitting posture, the vibration of vibrating reed low-intensity；
If there is microswitch signal in 5 seconds, learn again using secondary action data as correct sitting posture, then continue to obtain
6 axle acceleration sensor data, into circulation；
If continued to obtain 6 axle acceleration sensor data, into circulation without microswitch signal in 5 seconds；
If it is not medium sitting posture to be identified as, vibrating reed high-intensity oscillation；
If there is microswitch signal in 5 seconds, learn again using secondary action data as correct sitting posture, then continue to obtain
6 axle acceleration sensor data, into circulation；
If continued to obtain 6 axle acceleration sensor data, into circulation without microswitch signal in 5 seconds；
Power supply, which disconnects, then jumps out circulation, and correction terminates.
5. a kind of application method based on Curie modules human body attitude detection correcting device as claimed in claim 3, its feature
It is, the stance walks appearance correction, using development board Arduino101, is measured and entangles by the posture of following three kinds of different conditions
Just：
First is normal condition, and now vibrating reed does not vibrate, and user's back is straight and upright；The state can be stood by doctor or correction
The director for walking appearance provides initially training data, it is ensured that the setting of state first is correct；
Second is fair state, and now vibrating reed slightly vibrates, and user's back is slightly curved, can be by itself in unconscious absent-mindedness
Or people at one's side provides training auxiliary and obtains data, it is ensured that in the range of being arranged on of state second；
Third, is poor state, and now vibrating reed judder, user's back degree of crook is big, in complete absent-mindedness, changes state
Individual can voluntarily train, and scope is big, and accuracy requirement is low；
When user present second, the third two states posture when, by correct gesture module vibrating reed vibrating alert its correction
Stance walks appearance；
If any identification mistake, it is normal condition that can press microswitch by wrong state recognition is sorted out；
Its program circuit is as follows：
After gesture recognition starts, by including to obtain Curie wearable gesture recognition module acquisition 6 axle accelerations sensings in main controller
Device data；
If receiving study instruction, appearance is walked according to the corresponding stance of instruction study is learnt, continues to obtain 6 axle accelerations biography after study
Sensor data, into circulation；
If not receiving study instruction, the 6 axle acceleration sensor data acquired are passed through into neural network classification；
If the standard stance of being identified as walks appearance, continue to obtain 6 axle acceleration sensor data, into circulation；
If it is not that standard stance walks appearance to be identified as, continue to identify；
If being identified as medium stance walks appearance, the vibration of vibrating reed low-intensity；
If there is microswitch signal in 5 seconds, secondary action data is walked appearance as correct stance and learnt again, is then continued to
6 axle acceleration sensor data are obtained, into circulation；
If continued to obtain 6 axle acceleration sensor data, into circulation without microswitch signal in 5 seconds；
If it is not that medium stance walks appearance, vibrating reed high-intensity oscillation to be identified as；
If there is microswitch signal in 5 seconds, secondary action data is walked appearance as correct stance and learnt again, is then continued to
6 axle acceleration sensor data are obtained, into circulation；
If continued to obtain 6 axle acceleration sensor data, into circulation without microswitch signal in 5 seconds；
Finally, deenergization, jumps out circulation, and correction terminates.
6. a kind of application method based on Curie modules human body attitude detection correcting device as claimed in claim 3, its feature
It is, the body-building posture is corrected carries out gesture recognition using Curie or CurieNano modules, uses development board
Arduino101, gesture recognition mistake is corrected using microswitch；
First is normal condition, and now vibrating reed does not vibrate, user's action norm；The state provides data classification by setting person,
Ensure that the setting of state first is correct；
Second is poor state, and now vibrating reed slightly vibrates, and user's action is lack of standardization；
When user present second state posture when, by correct gesture module vibrating reed vibrating alert its entangle body-building posture；
If any identification mistake, microswitch can be pressed, is specification condition by wrong state recognition is sorted out；
Its program circuit is as follows：
After gesture recognition starts, by including to obtain Curie wearable gesture recognition module acquisition 6 axle accelerations sensings in main controller
Device data；
If being identified as standard state, continue to obtain 6 axle acceleration sensor data, into circulation；
If it is not standard state to be identified as, vibrating reed vibration 0.5s, continues to obtain 6 axle acceleration sensor data, enter
Circulation；
Finally, deenergization, jumps out circulation, and correction terminates.
7. a kind of application method based on Curie modules human body attitude detection correcting device as claimed in claim 3, its feature
It is, the continuous action correcting device is using the vibrating reed of Curie Nano modules and the number of main controller according to demands of individuals
Depending on product function；
Its program circuit is as follows：
After gesture recognition starts, the data of last time study are obtained from Flash, initialize Bluetooth function；
By including to obtain Curie 6 axle acceleration sensor data of wearable gesture recognition module acquisition in main controller；
If receiving study instruction, learn corresponding continuous action according to learning to instruct, continue to obtain 6 axle accelerations biography after study
Sensor data, into circulation；
If not receiving study instruction, the 6 axle acceleration sensor data acquired are passed through into neural network classification；
If it is for the compulsory exercise after a upper posture that identification, which obtains current pose, continue to obtain 6 axle acceleration sensor numbers
According into circulation；
It is not to vibrate 0.5s for the compulsory exercise after a upper posture, vibrating reed, continue acquisition 6 that if identification, which obtains current pose,
Axle acceleration sensor data, into circulation；
Finally, deenergization, jumps out circulation, and correction terminates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710911992.7A CN107684430A (en) | 2017-09-29 | 2017-09-29 | Correcting device and its application method are detected based on Curie modules human body attitude |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710911992.7A CN107684430A (en) | 2017-09-29 | 2017-09-29 | Correcting device and its application method are detected based on Curie modules human body attitude |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107684430A true CN107684430A (en) | 2018-02-13 |
Family
ID=61153862
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710911992.7A Pending CN107684430A (en) | 2017-09-29 | 2017-09-29 | Correcting device and its application method are detected based on Curie modules human body attitude |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107684430A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108542568A (en) * | 2018-04-20 | 2018-09-18 | 上海澄潭网络科技有限公司 | A kind of method and apparatus for adjusting back support device |
| CN108710901A (en) * | 2018-05-08 | 2018-10-26 | 广州市新苗科技有限公司 | A kind of deformity of spine screening system and method based on deep learning |
| CN109620244A (en) * | 2018-12-07 | 2019-04-16 | 吉林大学 | The Infants With Abnormal behavioral value method of confrontation network and SVM is generated based on condition |
| CN110432910A (en) * | 2019-09-18 | 2019-11-12 | 杨焕琦 | Posture correction system and wearable treatment clothing |
-
2017
- 2017-09-29 CN CN201710911992.7A patent/CN107684430A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108542568A (en) * | 2018-04-20 | 2018-09-18 | 上海澄潭网络科技有限公司 | A kind of method and apparatus for adjusting back support device |
| CN108710901A (en) * | 2018-05-08 | 2018-10-26 | 广州市新苗科技有限公司 | A kind of deformity of spine screening system and method based on deep learning |
| CN108710901B (en) * | 2018-05-08 | 2022-03-01 | 广州市新苗科技有限公司 | Deep learning-based spine deformity screening system and method |
| CN109620244A (en) * | 2018-12-07 | 2019-04-16 | 吉林大学 | The Infants With Abnormal behavioral value method of confrontation network and SVM is generated based on condition |
| CN110432910A (en) * | 2019-09-18 | 2019-11-12 | 杨焕琦 | Posture correction system and wearable treatment clothing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Interactive wearable systems for upper body rehabilitation: a systematic review | |
| US11037369B2 (en) | Virtual or augmented reality rehabilitation | |
| CN107684430A (en) | Correcting device and its application method are detected based on Curie modules human body attitude | |
| TWI669681B (en) | Electronic calculatiing apparatus, system, and method for providing body posture health information | |
| US8162857B2 (en) | Limb movement monitoring system | |
| JP2021535465A (en) | Camera-guided interpretation of neuromuscular signals | |
| CN108885500A (en) | Tactile data conversion equipment, tactile data conversion method and tactile data conversion program | |
| CN109074157A (en) | Tactile data conversion equipment, tactile data conversion method and tactile data conversion program and component arrangement body | |
| US20180242654A1 (en) | Smart apparel | |
| WO2016084499A1 (en) | Behavior classification system, behavior classification device, and behavior classification method | |
| CN111539245B (en) | CPR (CPR) technology training evaluation method based on virtual environment | |
| CN108236468A (en) | Body action analysis system, mancarried device and body action analysis method | |
| CN106308811A (en) | Wearable intelligent health monitoring system | |
| CN116869490A (en) | Vestibule rehabilitation training dynamic evaluation system based on artificial intelligence | |
| Skach et al. | Sensing social behavior with smart trousers | |
| Basmaji et al. | Posture detection framework using the internet of wearable things | |
| CN109993037A (en) | Action identification method, device, wearable device and computer readable storage medium | |
| Minaoglou et al. | A review on wearable product design and applications | |
| Wang | Designing smart garments for rehabilitation | |
| Visser | Haptic feedback in a posture correcting wearable | |
| JP2022187952A (en) | Program, method, and information processing device | |
| Bleser et al. | Human motion capturing and activity recognition using wearable sensor networks | |
| JP7445933B2 (en) | Information processing device, information processing method, and information processing program | |
| Jensen | Design Considerations and Application Examples for Embedded Classification Systems | |
| Zhou | Designing a wearable interaction system to improve teenagers' sitting posture |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180213 |
|
| WD01 | Invention patent application deemed withdrawn after publication |