CN108981745A - A kind of step-recording method, device, equipment and storage medium - Google Patents
A kind of step-recording method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN108981745A CN108981745A CN201811158417.5A CN201811158417A CN108981745A CN 108981745 A CN108981745 A CN 108981745A CN 201811158417 A CN201811158417 A CN 201811158417A CN 108981745 A CN108981745 A CN 108981745A
- Authority
- CN
- China
- Prior art keywords
- value
- step counting
- exercise data
- axis acceleration
- axis
- 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
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C22/00—Measuring distance traversed on the ground by vehicles, persons, animals or other moving solid bodies, e.g. using odometers, using pedometers
- G01C22/006—Pedometers
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Measurement Of Distances Traversed On The Ground (AREA)
Abstract
The invention discloses a kind of step counting algorithms, acquire the exercise data of sensor in motion process, the sensor is 3-axis acceleration sensor, obtain the characteristic value of collected exercise data, characteristic value is standardized to obtain standardized feature value, according to the standardized feature value, step counting state is judged using step counting model, if be in step counting state, step number step counting is then carried out according to the size of the peak value of step counting exercise data and reference threshold, the characteristic value for the exercise data that the present invention acquires is the characteristic of 20 dimensions, the true step counting state of user can accurately be reacted, effectively improve the accuracy of step counting.
Description
Technical field
The present invention relates to wearable device field, especially a kind of step-recording method, device, equipment and storage medium.
Background technique
With the continuous improvement of living standards, people increasingly focus on the health of oneself.For body-building by scientific methods, step counting
Device receives everybody as a kind of easily movement monitoring equipment and widely approves.Traditional step counting based on acceleration transducer
Method mainly judges paces by the extreme point of acceleration signal.
Wherein the Chinese patent application of Publication No. CN101354265B proposes a kind of basic skills and device.The program
The value of three axis resultant accelerations is calculated first, is then running or walking according to the judgement of the amplitude of three axis resultant accelerations, most
Step number is calculated according to motion state afterwards.The Chinese patent application of Publication No. CN104406604B proposes a kind of step-recording method.
In the program, the resultant acceleration value of collected 3-axis acceleration is calculated, and using sef-adapting filter in current data window
Resultant acceleration data be filtered, if the wave crest in filtered three axis resultant acceleration data meets certain threshold value simultaneously
And the time interval between adjacent peaks is more than certain threshold value, then the wave crest is Valid peak, each Valid peak table
Show a step.It but in the prior art, is using the threshold value set to determine whether starting step counting, and most of pedometer is only
It is the amplitude for using acceleration information, variance, mean value, maximum value, several as characteristic in minimum value, leads to step counting not
Accurately, it is therefore desirable to propose that one kind can be improved the accurate step counting algorithm of step counting.
Summary of the invention
The present invention is directed to solve at least some of the technical problems in related technologies.For this purpose, of the invention
One purpose is to provide a kind of step-recording method, device, equipment and storage medium
The technical scheme adopted by the invention is that:
A kind of step-recording method, comprising steps of
S1: the exercise data of sensor in acquisition motion process, the sensor are 3-axis acceleration sensor；S2: it obtains
The characteristic value of collected exercise data is taken, and calculates step counting exercise data；S3: the characteristic value is standardized
To standardized feature value；S4: according to the standardized feature value, step counting state is judged using step counting model；S5: if in meter
When step state, then step number step counting is carried out according to the size of the peak value of step counting exercise data and reference threshold；The exercise data
It include: x-axis acceleration value, y-axis acceleration value, z-axis acceleration value and three axis resultant acceleration square values；The step counting campaign number
According to for three axis resultant acceleration square values.
Further, in step S2, the characteristic value includes following characteristics value:
The mean value, the mean value of x-axis acceleration value of three axis resultant acceleration square values in time window, y-axis acceleration value it is equal
Value, z-axis acceleration value mean value,
The side of the variance, the variance of x-axis acceleration value of three axis resultant acceleration square values in time window, y-axis acceleration value
Poor, z-axis acceleration value variance,
The maximum value, the maximum value of x-axis acceleration value, y-axis acceleration value of three axis resultant acceleration square values in time window
Maximum value, the maximum value of z-axis acceleration value,
The minimum value, the minimum value of x-axis acceleration value, y-axis acceleration value of three axis resultant acceleration square values in time window
Minimum value, the minimum value of z-axis acceleration value,
The maximum value of the maximum value and minimum difference, x-axis acceleration value of three axis resultant acceleration square values in time window
With the maximum value and minimum value difference of minimum difference, the maximum value of y-axis acceleration value and minimum difference and z-axis acceleration value
Value.
Further, the formula of standardization described in step S3 specifically:
Wherein,It indicates into the characteristic value data after standardization excessively, XiIndicate i-th dimension characteristic value, μiAnd σiFor mark
Standardization parameter, n indicate the dimension of characteristic value.
Further, the μiAnd σiCalculation formula specifically:
Wherein m indicates the number of sample in training set, xi (j)Indicate the i-th dimension characteristic value of j-th of sample.
Further, judge that step counting state specifically includes in step S4:
S41: to the characteristic value data in S3 after standardization, step counting state, step counting are judged using step counting model
Model specifically:
Wherein,For the vector of the standardized feature value composition obtained in step S3, W indicates the vector being made of coefficient, T
Indicate the transposition of vector；
The calculation formula of the W are as follows:
Wherein, m indicates the number of sample in training set,Indicate the standardized feature value of i-th of sample, y(i)Indicate the
The mark value of i sample, α are regularization parameter, and are solved by gradient descent method and obtain W value；
S42: obtaining step counting state value according to step counting model and carry out the judgement of step counting state,
If the F (X) > 0.5 in S41, is determined as step counting state and executes S5；
If the F (X)≤0.5 in S41, it is judged to non-step counting state and returns to S1 to restart step counting.
It further, further include data filtering before step S5, specifically:
The step counting exercise data is filtered by Fast Fourier Transform (FFT) principle.
Further, the reference threshold includes dynamic peak value threshold value and the first preset threshold, and step S5 is specifically included:
S51: judging the peak value of step counting exercise data, specifically:
ai>ai-1And ai>ai+1
Wherein aiFor the step counting exercise data at current time, ai-1For the step counting exercise data of previous moment, ai+1It is latter
The step counting exercise data at moment；
S52: according to peak computational dynamic peak value threshold value, specifically:
Wherein B is dynamic peak value threshold value, and N is the number of peaks in time window, AiFor i-th peak value in time window
Numerical value, C are coefficient；
S53: carrying out step number step counting, specifically:
By each peak value in time window compared with dynamic peak value threshold value B and the first preset threshold A, if should
Peak value had both been greater than A also greater than B, then pedometer adds one.
On the other hand, the present invention also provides a kind of step count sets, comprising:
Acquisition device: for acquiring the exercise data of sensor in motion process；
Characteristic value acquisition device: for obtaining the characteristic value of collected exercise data；
Characteristic value modular station: the characteristic value is standardized to obtain standardized feature value；
Step counting state judging device: for judging step counting state using step counting model according to the standardized feature value；
Step number step count set, if when in step counting state, according to the peak value of step counting exercise data and referring to threshold
The size of value carries out step number step counting.
On the other hand, the present invention also provides a kind of control equipment of step-recording method, comprising:
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
A processor executes, so that at least one described processor is able to carry out step-recording method as described in any one of the above embodiments.
On the other hand, the present invention also provides a kind of computer readable storage medium, the computer readable storage medium is deposited
Computer executable instructions are contained, the computer executable instructions are based on keeping computer execution as described in any one of the above embodiments
One step process.
The beneficial effects of the present invention are:
A kind of step-recording method of the invention is obtained by the exercise data of 3-axis acceleration sensor in acquisition motion process
The characteristic value of collected exercise data is standardized the characteristic value to obtain standardized feature value, then utilize
Machine learning method (logistical linear classifier), step counting state is judged according to standardized feature value, when in step counting
When state, step number step counting is carried out according to the size of the peak value of step counting exercise data and preset threshold, wherein due to the work of user
There are many kinds of dynamic states, such as walks, runs, rides, sits, swims, therefore a small amount of feature is only used only and is difficult accurately to sentence
It is disconnected whether to start step counting, cause step counting inaccurate, the characteristic value for the exercise data that the present invention acquires is the characteristic of 20 dimensions, energy
Enough accurately reaction true step counting states of user, effectively improve the accuracy of step counting.
Detailed description of the invention
Fig. 1 is the step-recording method basic flow chart of one embodiment of the present invention；
Fig. 2 is the step-recording method specific flow chart of one embodiment of the present invention；
Fig. 3 is the step count set structural block diagram of one embodiment of the present invention.
Specific embodiment
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.
Embodiment one:
As shown in Figure 1, the basic flow chart of the step-recording method for the present embodiment, including S11: in acquisition motion process
The exercise data of sensor, wherein sensor is acceleration transducer；S12: obtaining the characteristic value of collected exercise data, and
Step counting exercise data is calculated, wherein step counting exercise data is three axis resultant acceleration square values；S13: standard is carried out to characteristic value
Change handles to obtain standardized feature value；S14: according to the standardized feature value, step counting state is judged using step counting model；S15:
By Fast Fourier Transform (FFT) principle to step counting exercise data in S12, i.e. three axis resultant acceleration square Value Datas are filtered；
S16: step number step counting is carried out.
Exercise data includes: x-axis acceleration value in step S11Y-axis acceleration valueZ-axis acceleration valueWith three
Axis resultant acceleration square value ai, wherein
The characteristic value of exercise data includes but is not limited to following 20 dimensional feature values in step S12, and calculation formula is as follows:
The mean value mean of three axis resultant acceleration square values:
The mean value mean of x-axis acceleration valuex:
The mean value mean of y-axis acceleration valuey:
The mean value meanz of z-axis acceleration value:
The variances sigma of three axis resultant acceleration square values:
The variances sigma of x-axis acceleration valuex:
The variances sigma of y-axis acceleration valuey:
The variances sigma z of z-axis acceleration value:
Maximum value MAX:MAX=max (a of three axis resultant acceleration square valuesi),i∈{1,2…,n} (10)
The maximum value MAX of x-axis acceleration valuex: MAXx=max (ai x),i∈{1,2…,n} (11)
The maximum value MAX of y-axis acceleration valuey: MAXy=max (ai y),i∈{1,2…,n} (12)
The maximum value MAX of z-axis acceleration valuez: MAXz=max (ai z),i∈{1,2…,n} (13)
The minimum value MIN of three axis resultant acceleration square values: MIN=min (ai),i∈{1,2…,n} (14)
The minimum value MIN of x-axis acceleration valuex: MINx=min (ai x),i∈{1,2…,n} (15)
The minimum value MIN of y-axis acceleration valuey: MINy=min (aiy),i∈{1,2…,n} (16)
The minimum value MIN of z-axis acceleration valuez: MINz=min (ai z),i∈{1,2…,n} (17)
The maximum value and minimum difference R:R=of three axis resultant acceleration square values | MAX-MIN | (18)
The maximum value and minimum difference R of x-axis acceleration valuex: Rx=| MAXx-MINx| (19)
The maximum value and minimum difference R of y-axis acceleration valuey: Ry=| MAXy-MINy| (20)
The maximum value and minimum difference R of z-axis acceleration valuez: Rz=| MAXz-MINz| (21)
It can also include that dominant frequency, energy, x and z-axis related coefficient, x, tri- axis of y, z add other than 20 above-mentioned dimensional feature values
The characteristic values such as the section of speed and resultant acceleration square distribution.
Illustrate a kind of step counting algorithm detailed process of the invention below with reference to Fig. 2.
1) initiation parameter first, including (unit is by step Numerical Num, time window T (unit is the second), sample frequency F
Hz) and the first preset threshold A, in some embodiments, the value of sample frequency F can be 20Hz, 25Hz or 30Hz, step Numerical
Num's is initially divided into two kinds, the first is that booting carries out step number step counting, and the value of Num is 0 at this time, and another kind is to start step counting
Continue step counting afterwards, the value of Num is the step Numerical of a preceding step counting at this time.
Wherein the first preset threshold A is obtained by grid search, grid search refer to traversal the first preset threshold A is all can
The value of energy, therefrom selects end value of the most accurate value of step counting data as parameter A.For example, acquisition it is many walk number
According to, and know in these data that user truly walks step number, at this point it is possible to assume A be (1.5,1,5,10,15,20 ...,
100), true step number is 20000, by the possible value of each A, brings into the step counting algorithm of the present embodiment, is calculated
Pedometer, for example, the pedometer that is calculated is when the pedometer being calculated as A=1 is 10000, A=5
The pedometer being calculated when 19000 ..., A=15 is 15000, it is known that calculating step number as A=1 is 10000, very
Solid step number is 20000, and it is 19000 that step number is calculated when error 10000, A=5, and whens error 1000, A=15 calculates step number and is
15000, error 5000, therefore error is minimum when A=5, for calculated value closest to the step number of true value, taking A at this time is 5, real
Border is in application, carry out grid search to a large amount of walking data to determine a suitable A value, particularly, A value once calculates
Just no longer change, use as an empirical value constant, A value is 2 in the present embodiment.
2) exercise data of sensor in motion process is acquired, exercise data includes: x-axis acceleration value, y-axis acceleration
Value, z-axis acceleration value and three axis resultant acceleration square values.
Sample size in each time window are as follows: M=F*T, the sampled data cached are the whole of M until data volume
Several times, for example, obtaining one section of 10 seconds exercise data from acceleration transducer, the time window T set is 2s, sampling frequency
Rate F is 50HZ, the then timestamp generated when being acquired according to data, the initial time t1 and knot of the exercise data of available acquisition
Beam time t2, then by this section of 10s data according to time window T be divided into 5 be not present data overlap time windows, first
The time range of a time window be [t1, t1+2), i.e. t1≤t < t1+2, the time range of the last one window be (t2-2,
T2], i.e. t2-2 < t≤t2, the number of sampled samples M in each time window is 100=2s*50Hz.
3) characteristic value of collected exercise data is obtained, i.e., is accelerated by x-axis acceleration value, y-axis acceleration value, z-axis
Angle value and three axis resultant acceleration square values extract 20 dimensional feature values according to formula (1)~formula (21).
4) characteristic value is standardized to obtain standardized feature value, the formula of standardization specifically:
Wherein,It indicates into the characteristic value data after standardization excessively, XiIndicate i-th dimension characteristic value, μiAnd σiFor mark
Standardization parameter, n indicate the dimension of characteristic value.
Wherein μiAnd σiCalculation formula specifically:
Wherein m indicates the number of sample in training set, xi (j)Indicate the i-th dimension characteristic value of j-th of sample.
In actual algorithm, μ can be obtained by great amount of samples data in advanceiAnd σi, can be as a constant
For in formula (22).
5) judge whether it is step counting state, specifically include step:
First to the characteristic value data after standardization, step counting state, step counting model are judged using step counting model
Specifically:
Wherein,For the vector of the standardized feature value composition in step 4) Jing Guo standardization, T indicates turning for vector
It sets, W indicates the vector being made of coefficient.
Such as in the present embodiment, the 20 dimensional feature values point obtained in a time window are calculated in step 3)
Not Wei (- 0.91427, -0.22424 .., -0.29868, -0.74205), then It is deployable to be Indicate the standardized feature value of the n-th dimension, WnIndicate n coefficient, n indicates the number of characteristic value, i.e. n dimension.
The calculation formula of the vector W of coefficient composition are as follows:
Wherein m indicates the number of sample in training set,Indicate the standardized feature value of i-th of sample, y(i)Indicate i-th
The mark value of a sample, general value are that 1 or -1, α is regularization parameter, and value 0.1 is solved by gradient descent method and obtained
W value, W value will receive the influence of the several factors such as the sample number of machine learning model and training set, the W's obtained in varied situations
Optimal value is different, for example, in some embodiment W by the value after machine learning be (- 3.691413e-2 ,-
8.136777e-2 ..., 1.968664e+00), and as the first preset threshold A, W value of front is once calculated also no longer
Change, is applied in formula (25) as a constant.
Then step counting state value is obtained according to step counting model and carries out the judgement of step counting state,
As F (X) > 0.5, then it is determined as step counting state and continues step counting；
As F (X)≤0.5, then it is determined as non-step counting state and return step 1) initiation parameter, restarts step counting.
6) by Fast Fourier Transform (FFT) principle to three axis resultant acceleration square value aiIt is filtered, due to human motion
In the process, acceleration transducer has certain unordered shaking, can introduce biggish noise in the data of acquisition, but human body
The frequency of motion process is in certain range, therefore can be filtered out using Fast Fourier Transform (FFT) not in human normal
Frequency signal within the scope of motion frequency, for example, frequency range of people when on foot be 0.5~5HZ, then be higher than 5HZ or
Useless noise is defined as lower than the signal of 0.5HZ, by Fast Fourier Transform (FFT) by collected three axis resultant acceleration
Square value aiSignal is converted into frequency domain by time domain, and the signal amplitude except 0.5~5HZ of frequency range is then set as 0, is finally led to
It crosses inverse fast Fourier transform and converts time domain data for frequency domain data, three axis resultant acceleration square value a can be completediNumber
According to filtering.
7) size of the peak value and reference threshold that compare step counting exercise data carries out step counting, specifically includes:
First determine whether three axis resultant acceleration square value aiData peak value, peak value judgment formula are as follows:
ai>ai-1And ai>ai+1 (26)
Wherein aiFor the step counting exercise data at current time, ai-1For the step counting exercise data of previous moment, ai+1It is latter
The step counting exercise data at moment；
Then according to peak computational dynamic peak value threshold value B, specific formula are as follows:
Wherein, N is the number of peaks in time window, AiFor the numerical value of i-th of peak value in time window, C is coefficient；
Step number step counting is finally carried out, in advance with dynamic peak value threshold value B and first by each peak value in time window
If threshold value A compares, if the peak value had both been greater than A also greater than B, pedometer adds one.
The machine learning algorithm that the present embodiment uses is logistical linear classifier, can also use random forest,
Decision tree, the other machines such as support vector base learning algorithms substitute.
Embodiment two:
As shown in figure 3, being a kind of step count set structural block diagram of the present embodiment, including acquisition device: for acquiring movement
The exercise data of sensor in the process；Characteristic value acquisition device: for obtaining the characteristic value of collected exercise data；Characteristic value
Modular station: characteristic value is standardized to obtain standardized feature value；Step counting state judging device: for according to institute
Standardized feature value is stated, judges step counting state using step counting model；Step number step count set, if when in step counting state,
Step number step counting is then carried out according to the size of the peak value of step counting exercise data and reference threshold.
The invention also discloses a kind of control equipment of step-recording method, comprising: at least one processor；And at least
The memory of one processor communication connection；Wherein, memory is stored with the instruction that can be executed by least one processor, instruction
It is executed by least one processor, so that at least one processor is able to carry out a kind of step-recording method of any of the above-described.
The invention also discloses a kind of computer readable storage medium, computer-readable recording medium storage has computer can
It executes instruction, computer executable instructions are for making a kind of step-recording method of computer any of the above-described.
A kind of step-recording method of the invention is obtained by the exercise data of 3-axis acceleration sensor in acquisition motion process
The characteristic value of collected exercise data is standardized the characteristic value to obtain standardized feature value, then utilize
Machine learning method (logistical linear classifier), step counting state is judged according to standardized feature value, when in step counting
When state, step number step counting is carried out according to the size of the peak value of step counting exercise data and preset threshold, wherein due to the work of user
There are many kinds of dynamic states, such as walks, runs, rides, sits, swims, therefore a small amount of feature is only used only and is difficult accurately to sentence
It is disconnected whether to start step counting, cause step counting inaccurate, the characteristic value for the exercise data that the present invention acquires is the characteristic of 20 dimensions, energy
Enough accurately reaction true step counting states of user, effectively improve the accuracy of step counting.
It is to be illustrated to preferable implementation of the invention, but the invention is not limited to the implementation above
Example, those skilled in the art can also make various equivalent variations on the premise of without prejudice to spirit of the invention or replace
It changes, these equivalent deformations or replacement are all included in the scope defined by the claims of the present application.
Claims (10)
1. a kind of step-recording method, which is characterized in that comprising steps of
S1: the exercise data of sensor in acquisition motion process, the sensor are 3-axis acceleration sensor；
S2: obtaining the characteristic value of collected exercise data, and calculates step counting exercise data；
S3: the characteristic value is standardized to obtain standardized feature value；
S4: according to the standardized feature value, step counting state is judged using step counting model；
S5: if be in step counting state, step number is carried out according to the size of the peak value of step counting exercise data and reference threshold
Step counting；
The exercise data includes: x-axis acceleration value, y-axis acceleration value, z-axis acceleration value and three axis resultant accelerations square
Value；
The step counting exercise data is three axis resultant acceleration square values.
2. a kind of step-recording method according to claim 1, which is characterized in that in step S2, the characteristic value include with
Lower eigenvalue:
Mean value, the mean value of x-axis acceleration value, the mean value of y-axis acceleration value, z of three axis resultant acceleration square values in time window
The mean value of axle acceleration value,
Variance, the variance of x-axis acceleration value, the variance of y-axis acceleration value, z of three axis resultant acceleration square values in time window
The variance of axle acceleration value,
The maximum value, the maximum value of x-axis acceleration value of three axis resultant acceleration square values, y-axis acceleration value be most in time window
Big value, the maximum value of z-axis acceleration value,
The minimum value, the minimum value of x-axis acceleration value of three axis resultant acceleration square values, y-axis acceleration value be most in time window
It is small value, z-axis acceleration value minimum value,
The maximum value and minimum difference, the maximum value of x-axis acceleration value of three axis resultant acceleration square values and most in time window
The maximum value and minimum difference of small value difference value, the maximum value of y-axis acceleration value and minimum difference and z-axis acceleration value.
3. a kind of step-recording method according to claim 1, which is characterized in that the formula of standardization described in step S3
Specifically:
Wherein i ∈ { 1,2 ..., n }
Wherein,It indicates into the characteristic value data after standardization excessively, XiIndicate i-th dimension characteristic value, μiAnd σiFor standardization ginseng
Number, n indicate the dimension of characteristic value.
4. a kind of step-recording method according to claim 3, which is characterized in that the μiAnd σiCalculation formula specifically:
Wherein m indicates the number of sample in training set, xi (j)Indicate the i-th dimension characteristic value of j-th of sample.
5. a kind of step-recording method according to claim 1, which is characterized in that judge that step counting state is specifically wrapped in step S4
It includes:
S41: to the characteristic value data in S3 after standardization, step counting state, step counting model are judged using step counting model
Specifically:
Wherein,For the vector of the standardized feature value composition obtained in step S3, W indicates that the vector being made of coefficient, T indicate
The transposition of vector；
The calculation formula of the W are as follows:
Wherein, m indicates the number of sample in training set,Indicate the standardized feature value of i-th of sample, y(i)It indicates i-th
The mark value of sample, α are regularization parameter, and are solved by gradient descent method and obtain W value；
S42: obtaining step counting state value according to step counting model and carry out the judgement of step counting state,
If the F (X) > 0.5 in S41, is determined as step counting state and executes S5；
If the F (X)≤0.5 in S41, it is judged to non-step counting state and returns to S1 to restart step counting.
6. a kind of step-recording method according to claim 2, which is characterized in that further include data filtering before step S5, specifically
Are as follows:
It is filtered by step counting exercise data described in Fast Fourier Transform (FFT) principle.
7. a kind of step-recording method according to claim 1, which is characterized in that the reference threshold includes dynamic peak value threshold value
With the first preset threshold, step S5 is specifically included:
S51: judging the peak value of step counting exercise data, specifically:
ai>ai-1And ai>ai+1
Wherein aiFor the step counting exercise data at current time, ai-1For the step counting exercise data of previous moment, ai+1For later moment in time
Step counting exercise data；
S52: according to peak computational dynamic peak value threshold value, specifically:
Wherein B is dynamic peak value threshold value, and N is the number of peaks in time window, AiFor the numerical value of i-th of peak value in time window,
C is coefficient；
S53: carrying out step number step counting, specifically:
By each peak value in time window compared with dynamic peak value threshold value B and the first preset threshold A, if the peak value
Numerical value had both been greater than A also greater than B, then pedometer adds one.
8. a kind of step count set characterized by comprising
Acquisition device: for acquiring the exercise data of sensor in motion process；
Characteristic value acquisition device: for obtaining the characteristic value of collected exercise data；
Characteristic value modular station: the characteristic value is standardized to obtain standardized feature value；
Step counting state judging device: for judging step counting state using step counting model according to the standardized feature value；
Step number step count set, if when in step counting state, according to the peak value of step counting exercise data and reference threshold
Size carries out step number step counting.
9. a kind of control equipment of step-recording method characterized by comprising
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
It manages device to execute, so that at least one described processor is able to carry out step-recording method as described in any one of claim 1 to 7.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer can
It executes instruction, the computer executable instructions are for making computer execute step counting side as described in any one of claim 1 to 7
Method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811158417.5A CN108981745A (en) | 2018-09-30 | 2018-09-30 | A kind of step-recording method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811158417.5A CN108981745A (en) | 2018-09-30 | 2018-09-30 | A kind of step-recording method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108981745A true CN108981745A (en) | 2018-12-11 |
Family
ID=64544280
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811158417.5A Pending CN108981745A (en) | 2018-09-30 | 2018-09-30 | A kind of step-recording method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108981745A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110595501A (en) * | 2019-10-09 | 2019-12-20 | 成都乐动信息技术有限公司 | Running distance correction method based on three-axis sensor |
| CN111765900A (en) * | 2020-07-30 | 2020-10-13 | 歌尔科技有限公司 | Step counting method, step counting device and computer readable storage medium |
| CN111765899A (en) * | 2020-06-30 | 2020-10-13 | 歌尔科技有限公司 | Step counting judgment method and device and computer readable storage medium |
| CN111780780A (en) * | 2020-06-16 | 2020-10-16 | 贵州省人民医院 | Step counting method and device based on filter bank |
| CN111780779A (en) * | 2020-06-15 | 2020-10-16 | Oppo（重庆）智能科技有限公司 | Data processing method, device, equipment and storage medium |
| CN111879334A (en) * | 2020-07-31 | 2020-11-03 | 歌尔科技有限公司 | Step counting method, step counting device and computer readable storage medium |
| CN111982149A (en) * | 2020-08-20 | 2020-11-24 | 歌尔科技有限公司 | Step counting identification method, step counting identification device, step counting identification equipment and readable storage medium |
| CN114034313A (en) * | 2021-11-10 | 2022-02-11 | 广东乐心医疗电子股份有限公司 | Step counting method and device |
| CN114563012A (en) * | 2020-11-27 | 2022-05-31 | 北京小米移动软件有限公司 | Step counting method, step counting device, step counting equipment and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060064276A1 (en) * | 2004-09-23 | 2006-03-23 | Inventec Appliances Corp. | Mobile phone with pedometer |
| CN101334280A (en) * | 2007-06-25 | 2008-12-31 | 雅马哈株式会社 | Apparatus and method for updating an offset value of magnetic sensor data as well as corresponding machine readable medium |
| EP2910901A1 (en) * | 2014-02-21 | 2015-08-26 | CSEM Centre Suisse d'Electronique et de Microtechnique SA | Method for determining an instant velocity of a user and for improving estimation of heart rate |
| CN105588577A (en) * | 2014-10-23 | 2016-05-18 | 中国移动通信集团公司 | Detection method and detection apparatus for abnormal step counting in exercise monitoring device |
| CN105844379A (en) * | 2015-12-09 | 2016-08-10 | 东软集团股份有限公司 | Vehicle risk assessment method and vehicle risk assessment device based on principal component analysis |
| CN105953814A (en) * | 2016-06-21 | 2016-09-21 | 广东欧珀移动通信有限公司 | Step counting method, step counting device and mobile terminal |
| CN106021961A (en) * | 2016-06-20 | 2016-10-12 | 吉林大学 | Urban standard cyclic working condition constructing method based on genetic algorithm optimization |
| CN107588784A (en) * | 2016-07-08 | 2018-01-16 | 深圳达阵科技有限公司 | A kind of state recognition and the method, apparatus and terminal distinguished |
| CN107613125A (en) * | 2017-09-22 | 2018-01-19 | 常州大学 | A kind of method of blind person's identification and guiding guide based on smart mobile phone |
| CN108446733A (en) * | 2018-03-19 | 2018-08-24 | 上海玺翎智能科技有限公司 | A kind of human body behavior monitoring and intelligent identification Method based on multi-sensor data |
-
2018
- 2018-09-30 CN CN201811158417.5A patent/CN108981745A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060064276A1 (en) * | 2004-09-23 | 2006-03-23 | Inventec Appliances Corp. | Mobile phone with pedometer |
| CN101334280A (en) * | 2007-06-25 | 2008-12-31 | 雅马哈株式会社 | Apparatus and method for updating an offset value of magnetic sensor data as well as corresponding machine readable medium |
| EP2910901A1 (en) * | 2014-02-21 | 2015-08-26 | CSEM Centre Suisse d'Electronique et de Microtechnique SA | Method for determining an instant velocity of a user and for improving estimation of heart rate |
| CN105588577A (en) * | 2014-10-23 | 2016-05-18 | 中国移动通信集团公司 | Detection method and detection apparatus for abnormal step counting in exercise monitoring device |
| CN105844379A (en) * | 2015-12-09 | 2016-08-10 | 东软集团股份有限公司 | Vehicle risk assessment method and vehicle risk assessment device based on principal component analysis |
| CN106021961A (en) * | 2016-06-20 | 2016-10-12 | 吉林大学 | Urban standard cyclic working condition constructing method based on genetic algorithm optimization |
| CN105953814A (en) * | 2016-06-21 | 2016-09-21 | 广东欧珀移动通信有限公司 | Step counting method, step counting device and mobile terminal |
| CN107588784A (en) * | 2016-07-08 | 2018-01-16 | 深圳达阵科技有限公司 | A kind of state recognition and the method, apparatus and terminal distinguished |
| CN107613125A (en) * | 2017-09-22 | 2018-01-19 | 常州大学 | A kind of method of blind person's identification and guiding guide based on smart mobile phone |
| CN108446733A (en) * | 2018-03-19 | 2018-08-24 | 上海玺翎智能科技有限公司 | A kind of human body behavior monitoring and intelligent identification Method based on multi-sensor data |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110595501A (en) * | 2019-10-09 | 2019-12-20 | 成都乐动信息技术有限公司 | Running distance correction method based on three-axis sensor |
| CN110595501B (en) * | 2019-10-09 | 2022-10-04 | 成都乐动信息技术有限公司 | Running distance correction method based on three-axis sensor |
| CN111780779A (en) * | 2020-06-15 | 2020-10-16 | Oppo（重庆）智能科技有限公司 | Data processing method, device, equipment and storage medium |
| CN111780780A (en) * | 2020-06-16 | 2020-10-16 | 贵州省人民医院 | Step counting method and device based on filter bank |
| CN111780780B (en) * | 2020-06-16 | 2022-06-03 | 贵州省人民医院 | Step counting method and device based on filter bank |
| CN111765899B (en) * | 2020-06-30 | 2022-04-22 | 歌尔科技有限公司 | Step counting judgment method and device and computer readable storage medium |
| CN111765899A (en) * | 2020-06-30 | 2020-10-13 | 歌尔科技有限公司 | Step counting judgment method and device and computer readable storage medium |
| CN111765900A (en) * | 2020-07-30 | 2020-10-13 | 歌尔科技有限公司 | Step counting method, step counting device and computer readable storage medium |
| CN111765900B (en) * | 2020-07-30 | 2022-02-22 | 歌尔科技有限公司 | Step counting method, step counting device and computer readable storage medium |
| CN111879334A (en) * | 2020-07-31 | 2020-11-03 | 歌尔科技有限公司 | Step counting method, step counting device and computer readable storage medium |
| CN111982149A (en) * | 2020-08-20 | 2020-11-24 | 歌尔科技有限公司 | Step counting identification method, step counting identification device, step counting identification equipment and readable storage medium |
| CN111982149B (en) * | 2020-08-20 | 2023-02-17 | 歌尔科技有限公司 | Step counting identification method, step counting identification device, step counting identification equipment and readable storage medium |
| CN114563012A (en) * | 2020-11-27 | 2022-05-31 | 北京小米移动软件有限公司 | Step counting method, step counting device, step counting equipment and storage medium |
| CN114563012B (en) * | 2020-11-27 | 2024-06-04 | 北京小米移动软件有限公司 | Step counting method, device, equipment and storage medium |
| CN114034313A (en) * | 2021-11-10 | 2022-02-11 | 广东乐心医疗电子股份有限公司 | Step counting method and device |
| CN114034313B (en) * | 2021-11-10 | 2024-04-02 | 广东乐心医疗电子股份有限公司 | Step counting method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108981745A (en) | A kind of step-recording method, device, equipment and storage medium | |
| CN105388495B (en) | Estimating local motion in physical exercise | |
| EP3110313A1 (en) | Method and apparatus for generating assessments using physical activity and biometric parameters | |
| US20100292618A1 (en) | Motion analysis device, program and method thereof and motion analysis system | |
| CN102727185B (en) | Motion energy consumption measuring apparatus based on heart rate and acceleration as well as measuring method | |
| US20160058373A1 (en) | Running Energy Efficiency | |
| JP6951516B2 (en) | Methods and systems for detecting human pace | |
| Panagiota et al. | Assessment of human gait speed and energy expenditure using a single triaxial accelerometer | |
| Soltani et al. | Real-world gait bout detection using a wrist sensor: An unsupervised real-life validation | |
| CN107966161A (en) | Walking detection method based on FFT | |
| Novatchkov et al. | Machine learning methods for the automatic evaluation of exercises on sensor-equipped weight training machines | |
| CN106422274A (en) | Multi-sensor-based assessment system for yoga | |
| JP2004358229A (en) | Physical motion analyzing apparatus and physical motion analyzing system | |
| CN106446530A (en) | Human body motion thermal physiology-based simulation risk assessment method and system | |
| CN105771224B (en) | A kind of locomotion evaluation system based on multisensor | |
| CN107270931A (en) | A kind of IOS and the general gait auto-correlation pedometer of Android platform | |
| CN103785157A (en) | Human body motion type identification accuracy improving method | |
| US10238320B2 (en) | Method and apparatus for calculating amount of exercise performed | |
| CA3046375A1 (en) | Systems, devices, and methods for biometric assessment | |
| Lueken et al. | Peak detection algorithm for gait segmentation in long-term monitoring for stride time estimation using inertial measurement sensors | |
| CN106730763A (en) | A kind of evaluation method of the students in middle and primary schools' physical training exercise load based on heart rate data and run duration | |
| JP5360093B2 (en) | Energy consumption measuring device, energy consumption measuring method, and motion analysis system | |
| Viana et al. | GymApp: A real time physical activity trainner on wearable devices | |
| Lu et al. | Test method for health-related physical fitness of college students in mobile internet environment | |
| CN112107290A (en) | Systems, methods, and software applications for KAM predicting multiple gait cycles of a subject |
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 | ||
| AD01 | Patent right deemed abandoned | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20220909 |