CN107239147A - A kind of human body context aware method based on wearable device, apparatus and system - Google Patents
A kind of human body context aware method based on wearable device, apparatus and system Download PDFInfo
- Publication number
- CN107239147A CN107239147A CN201710566180.3A CN201710566180A CN107239147A CN 107239147 A CN107239147 A CN 107239147A CN 201710566180 A CN201710566180 A CN 201710566180A CN 107239147 A CN107239147 A CN 107239147A
- Authority
- CN
- China
- Prior art keywords
- wearable device
- human body
- data
- signal
- contextual status
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
Abstract
The embodiments of the invention provide a kind of human body context aware method based on wearable device, apparatus and system, wherein, this method includes：Receive the first sensing data that wearable device is sent, wherein wearable device is worn on human calf rear portion, wearable device at least gathers human calf's motor message, electromyographic signal and ambient signal, the data that the first sensing data includes gathering wearable device mutually compensated for or Data Fusion after obtained data；The corresponding human body contextual status of first sensing data is determined according to default sorter model；Export the human body contextual status.The technical program is solved during context aware, because data acquisition accuracy and stability is not high, the problem of caused contextual status recognizes inaccurate, realizes the high accuracy perception to contextual status residing for human body.
Description
Technical field
The present embodiments relate to the technical field of context aware, more particularly to a kind of human body feelings based on wearable device
Border cognitive method, apparatus and system.
Background technology
In recent years, with the fast development of mobile device, the context aware technology by carrier of mobile device is got over
Carry out more extensive concern.Context aware technology refers to the ambient condition information sensed using sensor, judges that user works as
When residing contextual status, be that user or networking monitoring center provide reference information, to perform corresponding actions.
Existing human body context aware technical research, focuses mostly in the contextual status residing for known users, according to user's
The management rule of behavioural habits or system, makes corresponding behavior act or corresponding system operatio；In addition, also there is part research
Ambient condition information is gathered using mobile device, the contextual status residing for user is identified using the data collected, is had
There is certain context aware recognition capability.
But, often lead to collect with very big randomness using the condition and state of mobile device due to user
Data precision and stability it is low, the accuracy for causing contextual status to recognize is not high or the contextual status limited amount that can recognize
The problems such as.
The content of the invention
The present invention provides a kind of human body context aware method based on wearable device, apparatus and system, to realize to people
The high accuracy of contextual status residing for body is perceived.
In a first aspect, the embodiments of the invention provide a kind of human body context aware method based on wearable device, including：
The first sensing data that wearable device is sent is received, wherein, the wearable device is worn on after human calf
Portion, the wearable device at least gathers human calf's motor message, electromyographic signal and ambient signal, the first sensing number
According to including the data that the wearable device is gathered are mutually compensated for or Data Fusion after obtained data；
The corresponding human body contextual status of first sensing data is determined according to default sorter model；
Export the human body contextual status.
Second aspect, the embodiment of the present invention additionally provides a kind of human body context aware device based on wearable device, bag
Include：
Data reception module, the first sensing data for receiving wearable device transmission, wherein, the wearable device
It is worn on human calf rear portion, the wearable device at least gathers human calf's motor message, electromyographic signal and environment letter
Number, first sensing data includes mutually compensating for the data that the wearable device is gathered or Data Fusion
The data obtained afterwards；
Human body contextual status determining module, for determining that first sensing data is corresponding according to default sorter model
Human body contextual status；
Human body contextual status output module, for exporting the human body contextual status.
The third aspect, the embodiment of the present invention additionally provides a kind of human body context aware system based on wearable device, bag
Include：Wearable device and intelligent terminal；
The wearable device, for gathering sensing data；
The intelligent terminal is connected with the wearable device, and the intelligent terminal is included described in any embodiment of the present invention
The human body context aware device based on wearable device.
Human body context aware method provided in an embodiment of the present invention based on wearable device, apparatus and system, by wearing
Wear the wearable device in human calf rear portion and obtain sensing data, result in more stable and accurate human calf's motion letter
Number and electromyographic signal, while ambient signal can be obtained, there is provided the authentic data source of multiple types, there is provided more fully input data
Support；The data gathered are mutually compensated for or Data Fusion, can obtain more and more high stability
Data；It is accurate according to the human body contextual status that default sorter model and above-mentioned more stable, comprehensive and accurate data are determined
Degree is higher, realizes the high accuracy perception to contextual status residing for human body, solves during context aware, due to data
Gather accuracy and stability is not high, the problem of caused contextual status recognizes inaccurate.
Brief description of the drawings
Fig. 1 is a kind of flow signal of human body context aware method based on wearable device in the embodiment of the present invention one
Figure；
Fig. 2 is a kind of flow signal of human body context aware method based on wearable device in the embodiment of the present invention two
Figure；
Fig. 3 is a kind of flow signal of human body context aware method based on wearable device in the embodiment of the present invention three
Figure；
Fig. 4 is a kind of structure chart of human body context aware device based on wearable device in the embodiment of the present invention four；
Fig. 5 is a kind of structure chart of human body context aware system based on wearable device in the embodiment of the present invention five.
Embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention, rather than limitation of the invention.It also should be noted that, in order to just
Part related to the present invention rather than entire infrastructure are illustrate only in description, accompanying drawing.
Embodiment one
Fig. 1 is a kind of flow for human body context aware method based on wearable device that the embodiment of the present invention one is provided
Figure, this method is applied to the situation of contextual status residing for the data-speculative human body that intelligent terminal is collected using wearable device,
This method can be performed by the human body context aware device based on wearable device.Human body situation sense based on wearable device
Know that device can be realized by the mode of software or hardware, and be typically integrated in intelligent terminal, intelligent terminal can be intelligence
Mobile phone, tablet personal computer or personal computer equipment etc..As shown in figure 1, this method is specifically included：
Step 110, the first sensing data for receiving wearable device transmission, wherein, the wearable device is worn on people
Body posterior calf；The wearable device at least gathers human calf's motor message, electromyographic signal and ambient signal, described
One sensing data include mutually compensating for the data that the wearable device is gathered or Data Fusion after obtain
Data.
In the present embodiment, wearable device is installed and used on major muscles group, can detect people on rear side of human calf
Body movement signal and electromyographic signal.Because the muscle at human calf rear portion is more flourishing relative to the muscle at other positions of human body,
Therefore, motor message and electromyographic signal when the human body measured by the position is in motion or inactive state can it is sensitiveer with it is smart
Really, confidence level can be higher.
In the present embodiment, human calf's motor message, preferably can be the acceleration signal and angular speed of human calf
Signal；Ambient signal, preferably can be magnetic field variation signal, air pressure signal, temperature signal, the humidity letter in human body contextual status
Number and voice signal.
In the present embodiment, the first sensing data includes：Human calf's motor message, the flesh collected to wearable device
Electric signal and ambient signal, mutually compensated for or Data Fusion after obtained each data.First sensing data is excellent
Choosing can be electromyographic signal, motor message and ambient signal after treatment, wherein, the motor message after processing includes：Place
Acceleration signal, angular velocity signal, attitude signal and step height change signal after reason etc.；Ambient signal bag after processing
Include：Magnetic field variation signal, air pressure signal, temperature signal, moisture signal, voice signal after processing etc..It is wherein, exemplary,
Attitude data can be the data such as prostration, sitting posture or standing.
The default sorter model of step 120, basis determines the corresponding human body contextual status of first sensing data.
In the present embodiment, sorter model is preset, can preferably be obtained using machine learning algorithm training.
In the present embodiment, human body contextual status can include but is not limited to rest, walk, run, riding a bicycle, being in
In car, it is in bus, in subway and medium in elevator.
Step 130, the output human body contextual status.
Wherein it is possible to show human body contextual status on a display screen so that user checks, speech play people can also be utilized
Body contextual status, or export human body contextual status using other modes.
The human body context aware method based on wearable device that the present embodiment is provided, by being worn on human calf rear portion
Wearable device obtain sensing data, result in more stable and accurate human calf's motor message and electromyographic signal, together
When can obtain ambient signal, and there is provided the authentic data source of multiple types, there is provided the support of more fully input data；To what is gathered
Data are mutually compensated for or Data Fusion, can obtain more and more high stability data；According to default point
The human body contextual status degree of accuracy that class device model and above-mentioned more stable, comprehensive and accurate data are determined is higher, realizes pair
The high accuracy of contextual status residing for human body is perceived, and is solved during context aware, due to data acquisition accuracy and surely
It is qualitative not high, the problem of caused contextual status recognizes inaccurate.
On the basis of above-described embodiment,
Further, human body contextual status is exported, can be by the first sensing data together with corresponding human body contextual status
Cloud Server is uploaded to, Cloud Server can be stored and shared to the data of upload.Preferably, can be only by human body situation
State uploads to Cloud Server, and Cloud Server is stored and shared to human body contextual status, reduces and uploads a large amount of initial data
Spent transmission time and data traffic.
Specifically, human body contextual status can be uploaded to Cloud Server by intelligent terminal by wireless transmission method, wherein
Wireless transmission method preferably can be WiFi or SIM card communication network (such as 3G network, 4G networks).
Further, at least it is integrated with the wearable device：Accelerometer, gyroscope, myoelectricity Acquisition Circuit, air pressure
Meter, thermometer, hygrometer, microphone and magnetometer.
The data gathered to wearable device are mutually compensated for or Data Fusion, comprise the following steps (1) extremely
(6)：
(1) wearable device collection human calf motor message, electromyographic signal and ambient signal, wherein the human body is small
Leg motor message includes：The acceleration signal and angular velocity signal of human calf；The ambient signal includes：Human body contextual status
In magnetic field variation signal, air pressure signal, temperature signal, moisture signal and voice signal.
(2) wearable device is believed acceleration signal, angular velocity signal, changes of magnetic field according to temperature signal and moisture signal
Number and air pressure signal compensate processing.
Wherein it is possible to obtain detection demarcation record (the i.e. known temperature under known temperature and known humidity first with experiment
Corresponding acceleration signal, angular velocity signal, magnetic field variation signal and air pressure signal with known humidity), compensation deals are specific
Method that can be by tabling look-up, searches the temperature signal currently gathered and the letter corresponding to moisture signal in detection demarcation record
Number record, adjusts accelerometer, gyroscope, magnetometer and barometrical data using the signal record found and calculates in real time
Conversion coefficient, according to the conversion coefficient after regulation to the acceleration signal of collection, angular velocity signal, magnetic field variation signal and air pressure
Signal carries out temperature and humidity compensation, to obtain the acceleration signal after temperature and humidity compensation is handled, angular velocity signal, changes of magnetic field
Signal and air pressure signal, so as to improve noise and data wander situation that different operating mode lower sensors occur.
(3) wearable device is believed the acceleration signal after the compensation deals, angular velocity signal and changes of magnetic field
Number carry out attitude data fusion, obtain attitude signal.
In the present embodiment, the contextual status residing for human body can more accurately be judged using the attitude information of human body,
Therefore, it can using attitude data blending algorithm to the acceleration signal after temperature and humidity compensation is handled, angular velocity signal and
Magnetic field variation signal carries out attitude data fusion, to obtain attitude signal.Wherein, attitude data blending algorithm is preferably but not limited to
Dynamic offset-type extended Kalman filter.
(4) wearable device melts to the acceleration signal after the compensation deals and air pressure signal progress altitude information
Close, obtain foot's height variable signal.
In the present embodiment, human foot's height change signal also contributes to accurately judge the contextual status residing for human body,
It therefore, it can carry out the acceleration signal after temperature and humidity compensation is handled and air pressure signal using altitude information blending algorithm
Altitude information is merged, to obtain foot's height variable signal.Wherein, altitude information blending algorithm is preferably but not limited to dynamic compensation
Type extended Kalman filter.
(5) wearable device is using the exercise data after the compensation deals and the Data Fusion, to institute
State electromyographic signal and the voice signal carries out motion noise compensation.
During data acquisition, electromyographic signal and voice signal can be due to the motions of human body, extra motion of adulterating
Noise, accordingly, it would be desirable to carry out motion noise compensation deals to electromyographic signal and voice signal.Specifically compensation way is：Calculate and add
The standard deviation and covariance of rate signal, the standard deviation of angular velocity signal and covariance, the standard deviation of attitude signal and covariance,
And the standard deviation and covariance of foot's height variable signal, according to calculate obtained standard deviation and covariance to electromyographic signal and
Voice signal is filtered the regulation of the dynamic regulation and gain parameter of frequency, filters out in electromyographic signal and voice signal and adulterates
Motion artifacts.Wherein, when calculating standard deviation and covariance, the sampling number to each data is preferably but not limited to include currently adopting
Collect first 50 including data.
(6) wearable device output is by the compensation deals, the Data Fusion and the noise compensation deals
Each data-signal afterwards.
In the present embodiment, wireless communication apparatus can be provided with wearable device, its can by bluetooth, WiFi or
The wireless communication modes such as SIM card communication network, by through mutually compensating for or Data Fusion after obtained data be sent to intelligence
Can terminal.
In the present embodiment, wearable device results in more stable and accurate human calf's motor message and myoelectricity letter
Number, while the magnetic field variation signal in human body contextual status, air pressure signal, temperature signal, moisture signal and sound can be obtained
There is provided the support of more fully input data there is provided the authentic data source of multiple types for the ambient signals such as signal；To the data gathered
Mutually compensated for or Data Fusion, more and more high stability data can be obtained, so as to subsequently be based on this
The human body contextual status degree of accuracy of a little stable, comprehensive and accurate data predictions is higher.
Embodiment two
The present embodiment is there is provided the preferred embodiment of step 120 on the basis of embodiment one, and Fig. 2 is of the invention real
A kind of flow chart of human body context aware method based on wearable device of the offer of example two is provided, should be based on wearable with reference to Fig. 2
The human body context aware method of equipment includes：
Step 210, the first sensing data for receiving wearable device transmission, wherein, the wearable device is worn on people
Body posterior calf；The wearable device at least gathers human calf's motor message, electromyographic signal and ambient signal, described
One sensing data include mutually compensating for the data that the wearable device is gathered or Data Fusion after obtain
Data.
Step 220, first sensing data is pre-processed, obtain effective sensing data.
In the present embodiment, intelligent terminal receives the first sensing number that wearable device is sent by wireless communication mode
According to, wherein, wireless communication mode preferably can be bluetooth, WiFi or SIM card communication network etc..Receiving the first sensing data
Afterwards, intelligent terminal can be pre-processed to the first sensing data, reduce the noise in each signal, wherein, pretreatment preferably may be used
To be filtering process.
Step 230, feature extraction is carried out to the effective sensing data, obtain the First Eigenvalue.
In the present embodiment, after being pre-processed to the first sensing data, effective sensing data is obtained, intelligent terminal is to having
Imitate sensing data and carry out feature extraction, to obtain the main characteristic parameters of each data, realize Data Dimensionality Reduction.Wherein, feature extraction
It is preferred that can be 8 sections of fft analysis, peakvalue's checking, signal averaging, signal spectrum energy and variance yields scheduling algorithm.
Step 240, the First Eigenvalue inputted to the default sorter model calculated, obtain it is described effectively
The corresponding contextual status of sensing data.
In the present embodiment, intelligent terminal by the main characteristic parameters of each data obtained after feature extraction input to
In default sorter model, preset sorter model and classified calculating is carried out to each parameter of input, obtain user currently corresponding
Contextual status, such as rests, walks or runs.Wherein, default sorter model can be trained using machine learning algorithm
Arrive, machine learning algorithm preferably can be algorithm of support vector machine, it is of course also possible to use other algorithms, such as K arest neighbors
Algorithm etc..
Step 250, the output human body contextual status.
Exemplary, when human body is in resting state, the signal related to motion such as acceleration of human calf is relative
Weaker, the ambient signal of surrounding is also relatively steady, will not produce too big fluctuation., can so that user is in resting state as an example
Wearable device collects the data such as human calf's motor message, electromyographic signal and the ambient signal under user's current state, and
These data are mutually compensated for or Data Fusion, obtain the first sensing data, and be sent to intelligent terminal.Intelligence is eventually
Termination receive wearable device transmission the first sensing data after, the first sensing data is filtered wait pretreatment, removal
Unnecessary noise in signal, obtains effective sensing data；Intelligent terminal is put down using 8 sections of fft analysis, peakvalue's checking, signal
Average, signal spectrum energy and variance yields scheduling algorithm carry out feature extraction to effective sensing data, obtain the main spy of each data
Parameter is levied, and (intelligent terminal is utilized by machine learning such as algorithm of support vector machine by the default sorter model of characteristic parameter input
What Algorithm for Training was obtained) in, it is resting state to judge the contextual status residing for human body, and by resting state output so as to end
End control performs next step action, for example, duration of monitoring resting state etc..
The present embodiment is pre-processed and feature extraction to the first sensing data, and utilizes default sorter model realization point
Class is calculated, and the human body contextual status higher thus, it is possible to obtain the degree of accuracy realizes the high precision to contextual status residing for human body
Degree is perceived.
Embodiment three
The present embodiment sets up default sorter model on the basis of the various embodiments described above there is provided a kind of intelligent terminal
Preferred embodiment：Obtain the second sensing data under known contextual status；Second sensing data is pre-processed；To pre- place
The second sensing data after reason carries out feature extraction, obtains Second Eigenvalue；According to the known contextual status and the situation shape
The Second Eigenvalue under state, matching primitives are carried out to model parameter, obtain default sorter model.Carrying out human body situation
, it is necessary to pre-establish default point (before receiving the first sensing data that wearable device is sent) before status predication process
Class device model.
Fig. 3 is a kind of flow for human body context aware method based on wearable device that the embodiment of the present invention three is provided
Figure, as shown in figure 3, this method comprises the following steps：
Step 310, the second sensing data obtained under known contextual status.
Wearable device needs to gather corresponding human calf's motor message, electromyographic signal under various known contextual status
And ambient signal, and each signal collected is mutually compensated for or Data Fusion, the second sensing data is finally given,
Second sensing data is sent to intelligent terminal.
Step 320, second sensing data is pre-processed.
Step 330, to pretreated second sensing data carry out feature extraction, obtain Second Eigenvalue.
The pretreatment and feature extraction carried out to the second sensing data and the process and method to the processing of the first sensing data
All same.
Step 340, according to the Second Eigenvalue under the known contextual status and the contextual status, to model parameter
Matching primitives are carried out, the default sorter model is obtained.
According to machine learning algorithms such as algorithm of support vector machine, the Second Eigenvalue under known contextual status, training are utilized
Go out the sorter model with high accuracy.Wherein, sorter model can train one as needed, can also train one
Group, is not specifically limited herein.
Step 350, the first sensing data for receiving wearable device transmission, wherein, the wearable device is worn on people
Body posterior calf, the wearable device at least gathers human calf's motor message, electromyographic signal and ambient signal, described
One sensing data include mutually compensating for the data that the wearable device is gathered or Data Fusion after obtain
Data.
The default sorter model of step 360, basis determines the corresponding human body contextual status of first sensing data.
Step 370, the output human body contextual status.
Exemplary, before specifically human body contextual status prediction process is carried out, being respectively at human body various can think
Under the contextual status arrived, under each contextual status, the wearable device being worn on the major muscles group of human calf rear portion is adopted
Collection human calf's motor message, electromyographic signal and ambient signal under the contextual status, and the signal of collection is carried out mutually
Compensation or Data Fusion, obtain the second sensing data under each contextual status.Intelligent terminal for reception is under each contextual status
The second sensing data after, pretreatment and the feature extraction such as can be filtered to the second sensing data under each contextual status,
It is used to train the Second Eigenvalue under each contextual status of sorter model to obtain.Obtain the second feature under each contextual status
After value, intelligent terminal is instructed using machine learning algorithms such as algorithm of support vector machine to the Second Eigenvalue under each contextual status
Practice to obtain ideal sort device model.Herein it should be clear that, sorter model train during, preset classification
The threshold value of device model accuracy rate, when the accuracy rate of system detectio to sorter model meets or exceeds the threshold value, machine learning
Algorithm is by deconditioning, and using the training result as final sorter model, wherein, sorter model accuracy rate threshold value is excellent
Choosing can be 98%, 97% or 95% etc.., just can be to situation shape residing for unknown human body after default sorter model is obtained
State is predicted and exported.
The present embodiment is trained using machine learning algorithm to grader, obtains meeting the grader mould of accuracy requirement
Type, the prediction of human body contextual status is carried out based on the model, accurate human body contextual status can be obtained, realize to people
The high accuracy of contextual status residing for body is perceived.
It is related to model training in the present embodiment and situation predicts two processes, wherein, model training mainly includes：Data are pre-
Processing, feature extraction, machine learning algorithm training, sorter model are set up；Situation prediction is main：Data prediction, feature are carried
Take, the situation based on sorter model is predicted, contextual status output.
Example IV
Fig. 4 is a kind of structure for human body context aware device based on wearable device that the embodiment of the present invention four is provided
Figure, the device can apply to intelligent terminal, as shown in figure 4, the device can include：
Data reception module 410, the first sensing data for receiving wearable device transmission, wherein, it is described wearable
Equipment is worn on human calf rear portion, and the wearable device at least gathers human calf's motor message, electromyographic signal and ring
Border signal, first sensing data includes mutually compensating for the data that the wearable device is gathered or data fusion
The data obtained after processing；
Human body contextual status determining module 420, for determining first sensing data pair according to default sorter model
The human body contextual status answered；
Human body contextual status output module 430, for exporting the human body contextual status.
The human body context aware device based on wearable device that the present embodiment is provided, by being worn on human calf rear portion
Wearable device obtain sensing data, result in more stable and accurate human calf's motor message and electromyographic signal, together
When can obtain ambient signal, and there is provided the authentic data source of multiple types, there is provided the support of more fully input data；To what is gathered
Data are mutually compensated for or Data Fusion, can obtain more and more high stability data；According to default point
The human body contextual status degree of accuracy that class device model and above-mentioned more stable, comprehensive and accurate data are determined is higher, realizes pair
The high accuracy of contextual status residing for human body is perceived, and is solved during context aware, due to data acquisition accuracy and surely
It is qualitative not high, the problem of caused contextual status recognizes inaccurate.
Above-mentioned data reception module 410, human body contextual status determining module 420 and human body contextual status output module 430
It can be used in predicting human body contextual status according to the sensing data of wearable device.Further, as shown in figure 4, said apparatus
The correlation module for setting up sorter model can also be included, it is as follows：
Data acquisition module 440, for obtaining the second sensing data under known contextual status；
Data preprocessing module 450, for being pre-processed to second sensing data；
Characteristic extracting module 460, for carrying out feature extraction to pretreated second sensing data, obtains second feature
Value；
Default sorter model sets up module 470, for according to the institute under the known contextual status and the contextual status
Second Eigenvalue is stated, matching primitives are carried out to model parameter, the default sorter model is obtained.
Further, the human body contextual status determining module 420 includes：
Data pre-processing unit, for being pre-processed to first sensing data, obtains effective sensing data；
Human body contextual status determining unit, for determining effective sensing data pair according to the default sorter model
The contextual status answered, is used as the corresponding human body contextual status of first sensing data.
Further, the human body contextual status determining unit can include：
Feature extraction subelement, for carrying out feature extraction to effective sensing data, obtains the First Eigenvalue；
Human body contextual status determination subelement, for inputting to the default sorter model the First Eigenvalue
Row is calculated, and obtains the corresponding contextual status of effective sensing data.
The executable present invention of the human body context aware device based on wearable device that the embodiment of the present invention is provided is any
The human body context aware method based on wearable device that embodiment is provided, possesses the corresponding functional module of execution method and has
Beneficial effect.
It is worth noting that, in the device of the present embodiment, included unit and module are simply according to function logic
Divided, but be not limited to above-mentioned division, as long as corresponding function can be realized, for example, model training mistake
Preprocessing function involved by journey, and the preprocessing function involved by contextual status prediction process, can use same unit or
Module is realized；Feature extraction functions involved by model training process, with the feature extraction involved by contextual status prediction process
Function, can use same unit or module to realize.In addition, the specific name of each functional unit is also only to facilitate mutual area
Point, the protection domain being not intended to limit the invention.
Embodiment five
Fig. 5 is a kind of structure for human body context aware system based on wearable device that the embodiment of the present invention five is provided
Figure, as shown in figure 5, the system includes：Wearable device 510 and intelligent terminal 520.
The wearable device 510, for gathering sensing data；
The intelligent terminal 520 is connected with wearable device 510, and intelligent terminal 520 can be included described in example IV
Any human body context aware device based on wearable device.
The human body context aware system based on wearable device that the present embodiment is provided, by being worn on human calf rear portion
Wearable device obtain sensing data, result in more stable and accurate human calf's motor message and electromyographic signal, together
When can obtain ambient signal, and there is provided the authentic data source of multiple types, there is provided the support of more fully input data；To what is gathered
Data are mutually compensated for or Data Fusion, can obtain more and more high stability data；According to default point
The human body contextual status degree of accuracy that class device model and above-mentioned more stable, comprehensive and accurate data are determined is higher, realizes pair
The high accuracy of contextual status residing for human body is perceived, and is solved during context aware, due to data acquisition accuracy and surely
It is qualitative not high, the problem of caused contextual status recognizes inaccurate.
Further, said system can also include：
Cloud Server 530, for receiving the human body contextual status of the upload of intelligent terminal 520, and enters to human body contextual status
Row is stored and shared.
Further, wearable device 510 is at least integrated with：Accelerometer 511, gyroscope 512, myoelectricity Acquisition Circuit
513rd, barometer 514, thermometer 515, hygrometer 516, microphone 517, magnetometer 518 and microcontroller 519.Microcontroller
519 are used to compensate processing, Data Fusion and noise compensation deals to the data collected.
Wireless communication apparatus 501 can also be set in wearable device 510, for being communicated with intelligent terminal 520.Tool
Body can using the wireless communication mode such as bluetooth, WiFi or SIM card communication network, will through mutually compensating for or Data Fusion after
Obtained data are sent to intelligent terminal 520.
In addition, power management module 502 can also be integrated with wearable device 510, for microcontroller 519 and nothing
Line communication device 501 charges.For example, power management module 502 can be lithium battery and its charging circuit.
The executable present invention of the human body context aware system based on wearable device that the embodiment of the present invention is provided is any
The human body context aware method based on wearable device that embodiment is provided, possesses the corresponding functional module of execution method and has
Beneficial effect.
Note, above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The invention is not restricted to specific embodiment described here, can carry out for a person skilled in the art it is various it is obvious change,
Readjust and substitute without departing from protection scope of the present invention.Therefore, although the present invention is carried out by above example
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
Other more equivalent embodiments can be included, and the scope of the present invention is determined by scope of the appended claims.
Claims (13)
1. a kind of human body context aware method based on wearable device, it is characterised in that including：
The first sensing data that wearable device is sent is received, wherein, the wearable device is worn on human calf rear portion, institute
State wearable device and at least gather human calf's motor message, electromyographic signal and ambient signal, the first sensing data bag
Include the data that the wearable device is gathered are mutually compensated for or Data Fusion after obtained data；
The corresponding human body contextual status of first sensing data is determined according to default sorter model；
Export the human body contextual status.
2. a kind of human body context aware method based on wearable device according to claim 1, it is characterised in that according to
Default sorter model determines the corresponding human body contextual status of first sensing data, including：
First sensing data is pre-processed, effective sensing data is obtained；
The corresponding contextual status of effective sensing data is determined according to the default sorter model, the described first sensing is used as
The corresponding human body contextual status of data.
3. a kind of human body context aware method based on wearable device according to claim 2, it is characterised in that according to
The default sorter model determines the corresponding contextual status of effective sensing data, including：
Feature extraction is carried out to effective sensing data, the First Eigenvalue is obtained；
The First Eigenvalue is inputted to the default sorter model and calculated, effective sensing data correspondence is obtained
Contextual status.
4. a kind of human body context aware method based on wearable device according to claim 1, it is characterised in that connecing
Receive before the first sensing data that wearable device is sent, methods described also includes：
Obtain the second sensing data under known contextual status；
Second sensing data is pre-processed；
Feature extraction is carried out to pretreated second sensing data, Second Eigenvalue is obtained；
According to the Second Eigenvalue under the known contextual status and the contextual status, matching meter is carried out to model parameter
Calculate, obtain the default sorter model.
5. according to any a kind of described human body context aware method based on wearable device, its feature in claim 1-4
It is, exports the human body contextual status, including：
The human body contextual status is uploaded into Cloud Server, data storage is carried out and shared for the Cloud Server.
6. according to any a kind of described human body context aware method based on wearable device, its feature in claim 1-4
It is, is at least integrated with the wearable device：It is accelerometer, gyroscope, myoelectricity Acquisition Circuit, barometer, thermometer, wet
Degree meter, microphone and magnetometer；
The data that the wearable device is gathered are mutually compensated for or Data Fusion, including：
The wearable device collection human calf motor message, electromyographic signal and ambient signal, wherein the human calf
Motor message includes：The acceleration signal and angular velocity signal of human calf；The ambient signal includes：In human body contextual status
Magnetic field variation signal, air pressure signal, temperature signal, moisture signal and voice signal；
The wearable device is believed the acceleration signal, the angular speed according to the temperature signal and the moisture signal
Number, the magnetic field variation signal and the air pressure signal compensate processing；
The wearable device enters to the acceleration signal after the compensation deals, angular velocity signal and magnetic field variation signal
Row attitude data is merged, and obtains attitude signal；
The wearable device carries out altitude information fusion to the acceleration signal after the compensation deals and air pressure signal,
Obtain foot's height variable signal；
The wearable device is using the exercise data after the compensation deals and the Data Fusion, to the flesh
Electric signal and the voice signal carry out motion noise compensation；
The wearable device output is after the compensation deals, the Data Fusion and the noise compensation deals
Each data-signal.
7. a kind of human body context aware device based on wearable device, it is characterised in that including：
Data reception module, the first sensing data for receiving wearable device transmission, wherein, the wearable device wearing
In human calf rear portion, the wearable device at least gathers human calf's motor message, electromyographic signal and ambient signal, institute
State the first sensing data include the data that the wearable device is gathered are mutually compensated for or Data Fusion after must
The data arrived；
Human body contextual status determining module, for determining the corresponding human body of first sensing data according to default sorter model
Contextual status；
Human body contextual status output module, for exporting the human body contextual status.
8. a kind of human body context aware device based on wearable device according to claim 7, it is characterised in that described
Human body contextual status determining module includes：
Data pre-processing unit, for being pre-processed to first sensing data, obtains effective sensing data；
Human body contextual status determining unit, for determining that effective sensing data is corresponding according to the default sorter model
Contextual status, is used as the corresponding human body contextual status of first sensing data.
9. a kind of human body context aware device based on wearable device according to claim 8, it is characterised in that described
Human body contextual status determining unit includes：
Feature extraction subelement, for carrying out feature extraction to effective sensing data, obtains the First Eigenvalue；
Human body contextual status determination subelement, based on the First Eigenvalue is inputted to the default sorter model carry out
Calculate, obtain the corresponding contextual status of effective sensing data.
10. a kind of human body context aware device based on wearable device according to claim 7, it is characterised in that institute
Stating device also includes：
Data acquisition module, for obtaining the second sensing data under known contextual status；
Data preprocessing module, for being pre-processed to second sensing data；
Characteristic extracting module, for carrying out feature extraction to pretreated second sensing data, obtains Second Eigenvalue；
Default sorter model sets up module, for special according to described second under the known contextual status and the contextual status
Value indicative, carries out matching primitives to model parameter, obtains the default sorter model.
11. a kind of human body context aware system based on wearable device, it is characterised in that including：Wearable device and intelligence
Terminal；
The wearable device, for gathering sensing data；
The intelligent terminal is connected with the wearable device, and the intelligent terminal includes any described in claim 7-10
Human body context aware device based on wearable device.
12. a kind of human body context aware system based on wearable device according to claim 11, it is characterised in that institute
Stating system also includes：Cloud Server, for receiving the human body contextual status that the intelligent terminal is uploaded, and to the human body situation
State is stored and shared.
13. a kind of human body context aware system based on wearable device according to claim 11, it is characterised in that institute
Wearable device is stated at least to be integrated with：Accelerometer, gyroscope, myoelectricity Acquisition Circuit, barometer, thermometer, hygrometer, Mike
Wind, magnetometer and microcontroller, the microcontroller be used to compensating the data that collect processing, Data Fusion and
Noise compensation deals.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710566180.3A CN107239147A (en) | 2017-07-12 | 2017-07-12 | A kind of human body context aware method based on wearable device, apparatus and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710566180.3A CN107239147A (en) | 2017-07-12 | 2017-07-12 | A kind of human body context aware method based on wearable device, apparatus and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107239147A true CN107239147A (en) | 2017-10-10 |
Family
ID=59991011
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710566180.3A Pending CN107239147A (en) | 2017-07-12 | 2017-07-12 | A kind of human body context aware method based on wearable device, apparatus and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107239147A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108307050A (en) * | 2018-01-18 | 2018-07-20 | 大连理工大学 | A kind of construction worker's action recognition integrated approach based on intelligent mobile phone sensor |
| CN109126045A (en) * | 2018-09-30 | 2019-01-04 | 常州太玄信息技术有限公司 | intelligent motion analysis and training system |
| CN110516760A (en) * | 2019-09-02 | 2019-11-29 | Oppo（重庆）智能科技有限公司 | Situation identification method, device, terminal and computer readable storage medium |
| CN114004247A (en) * | 2020-07-14 | 2022-02-01 | 荣耀终端有限公司 | Riding detection method, electronic device and computer readable storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110288379A1 (en) * | 2007-08-02 | 2011-11-24 | Wuxi Microsens Co., Ltd. | Body sign dynamically monitoring system |
| CN104407702A (en) * | 2014-11-26 | 2015-03-11 | 三星电子（中国）研发中心 | Method, device and system for performing actions based on context awareness |
| CN105597298A (en) * | 2016-04-05 | 2016-05-25 | 哈尔滨工业大学 | Fitness effect evaluation system based on electromyographic signal and body movement detection |
| CN105877757A (en) * | 2016-03-30 | 2016-08-24 | 哈尔滨理工大学 | Multi-sensor integrated human motion posture capturing and recognizing device |
| US20160253594A1 (en) * | 2015-02-26 | 2016-09-01 | Stmicroelectronics, Inc. | Method and apparatus for determining probabilistic context awreness of a mobile device user using a single sensor and/or multi-sensor data fusion |
| CN205625934U (en) * | 2015-11-02 | 2016-10-12 | 广州阿路比电子科技有限公司 | A new type sensor for gathering human motion signal |
-
2017
- 2017-07-12 CN CN201710566180.3A patent/CN107239147A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110288379A1 (en) * | 2007-08-02 | 2011-11-24 | Wuxi Microsens Co., Ltd. | Body sign dynamically monitoring system |
| CN104407702A (en) * | 2014-11-26 | 2015-03-11 | 三星电子（中国）研发中心 | Method, device and system for performing actions based on context awareness |
| US20160253594A1 (en) * | 2015-02-26 | 2016-09-01 | Stmicroelectronics, Inc. | Method and apparatus for determining probabilistic context awreness of a mobile device user using a single sensor and/or multi-sensor data fusion |
| CN205625934U (en) * | 2015-11-02 | 2016-10-12 | 广州阿路比电子科技有限公司 | A new type sensor for gathering human motion signal |
| CN105877757A (en) * | 2016-03-30 | 2016-08-24 | 哈尔滨理工大学 | Multi-sensor integrated human motion posture capturing and recognizing device |
| CN105597298A (en) * | 2016-04-05 | 2016-05-25 | 哈尔滨工业大学 | Fitness effect evaluation system based on electromyographic signal and body movement detection |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108307050A (en) * | 2018-01-18 | 2018-07-20 | 大连理工大学 | A kind of construction worker's action recognition integrated approach based on intelligent mobile phone sensor |
| CN109126045A (en) * | 2018-09-30 | 2019-01-04 | 常州太玄信息技术有限公司 | intelligent motion analysis and training system |
| CN110516760A (en) * | 2019-09-02 | 2019-11-29 | Oppo（重庆）智能科技有限公司 | Situation identification method, device, terminal and computer readable storage medium |
| CN114004247A (en) * | 2020-07-14 | 2022-02-01 | 荣耀终端有限公司 | Riding detection method, electronic device and computer readable storage medium |
| CN114004247B (en) * | 2020-07-14 | 2022-11-01 | 荣耀终端有限公司 | Riding detection method, electronic device and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ferrari et al. | On the personalization of classification models for human activity recognition | |
| Reddy et al. | Using mobile phones to determine transportation modes | |
| CN110428808A (en) | A kind of audio recognition method and device | |
| Sztyler et al. | Online personalization of cross-subjects based activity recognition models on wearable devices | |
| KR101690649B1 (en) | Activity classification in a multi-axis activity monitor device | |
| CN107239147A (en) | A kind of human body context aware method based on wearable device, apparatus and system | |
| CN106887115B (en) | A kind of Falls Among Old People monitoring device and fall risk appraisal procedure | |
| CN106096662B (en) | Human motion state identification based on acceleration transducer | |
| CN103455170B (en) | A kind of sensor-based motion of mobile terminals identifying device and method | |
| US20210064141A1 (en) | System for detecting a signal body gesture and method for training the system | |
| Zang et al. | Gait-cycle-driven transmission power control scheme for a wireless body area network | |
| CN108245880A (en) | Body-sensing detection method for visualizing and system based on more wearing annulus sensor fusions | |
| Altini et al. | Self-calibration of walking speed estimations using smartphone sensors | |
| KR101418333B1 (en) | Apparatus and method for recognizing user activity | |
| EP2835769A1 (en) | Method, device and system for annotated capture of sensor data and crowd modelling of activities | |
| Ahmed et al. | An approach to classify human activities in real-time from smartphone sensor data | |
| CN108958482B (en) | Similarity action recognition device and method based on convolutional neural network | |
| Oshin et al. | ERSP: An energy-efficient real-time smartphone pedometer | |
| KR20190047648A (en) | Method and wearable device for providing feedback on action | |
| Samiei-Zonouz et al. | Smartphone-centric human posture monitoring system | |
| CN114053112A (en) | Massage method, device, terminal equipment and medium | |
| Pascoal et al. | Activity recognition in outdoor sports environments: smart data for end-users involving mobile pervasive augmented reality systems | |
| Santos et al. | Context inference for mobile applications in the UPCASE project | |
| CN210697629U (en) | Signal acquisition device, mobile terminal and signal analysis system | |
| CN107688828A (en) | A kind of bus degree of crowding estimating and measuring method based on mobile phone sensor |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171010 |
|
| RJ01 | Rejection of invention patent application after publication |