CN107270931A - A kind of IOS and the general gait auto-correlation pedometer of Android platform - Google Patents
A kind of IOS and the general gait auto-correlation pedometer of Android platform Download PDFInfo
- Publication number
- CN107270931A CN107270931A CN201611207910.2A CN201611207910A CN107270931A CN 107270931 A CN107270931 A CN 107270931A CN 201611207910 A CN201611207910 A CN 201611207910A CN 107270931 A CN107270931 A CN 107270931A
- Authority
- CN
- China
- Prior art keywords
- correlation
- data
- gait
- acceleration
- auto
- 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
Abstract
The present invention relates to a kind of general pedometer implementation method on IOS and Android platform and pedometer for a kind of IOS of denomination of invention and the general gait auto-correlation pedometer summary of Android platform, this method, which is based on mobile phone acceleration sensor data, realizes high-precision step function, and mobile phone can be carried by user with any attitude(Including it is hand-held, be placed on jacket/trouser pocket or knapsack etc.).This method includes:Acceleration transducer sample frequency is determined, low-pass filtering treatment elimination noise is carried out to initial data, utilizes dynamic auto-correlation algorithm to calculate degree of fluctuation and gait auto-correlation coefficient, statistics step number.The present invention can be used for the mobile phone of the operating systems such as the IOS/android with acceleration transducer to count daily walking step number; differentiate that user's walking behavior provides method for all kinds of APP; instrument is provided for user record daily routines; with training without preset user behavior, statistical accuracy is high, the features such as cross-platform can use.
Description
Technical field
The present invention relates to a kind of general pedometer implementation method on IOS and Android platform, more particularly to pair plus
Velocity sensor data are acquired, stable frequency, filtering and noise reduction and calculating reach that meter walks the technology of effect.
Background technology
With the popularization and the progress of sensor technology of smart mobile phone, various sensors built in increasing mobile phone,
The big data that these sensors are contained can help user/third party APP more easily to monitor user's daily behavior, and user is come
Daily exercise and consumption can conveniently be counted by saying, it will be seen that user's current state, scene is done to user for third party APP
Change is recommended/avoided to bother user.
Current ios device built-in pedometer, but module meter step accuracy rate is not high, meanwhile, the module can not be straight
Connect and be transplanted in android equipment.Except mobile phone, the wearable device of step can much be counted by also existing, but these equipment need
Some position of body is worn on, the wish that user is used for a long time is not high.
The content of the invention
Present invention mainly solves technical problem there is provided a kind of meter step side general in IOS and Android platform
Method, using wave filter and dynamic auto-correlation algorithm, realizes meter step effect.
Wherein, central scope of the invention derives from individual gait continuation and similitude, by previous time inframe
Data and the data of latter time frame carry out similarity identification, judge time inframe data fluctuations degree and with it is next when
Between frame data similarity degree whether reach threshold value determine whether count step number;In addition, the time frame of the present invention is that dynamic is true
Fixed, occur being careful/running even if user also not interfering with present invention meter step precision;3rd, the present invention is without preset user's row
Training is used as data, it might even be possible to which realization counts step while learning the step precision in terms of improving;Finally, the invention provides one
Plant in IOS and Android platform general step-recording method, reduce the cost of third party APP exploitations.
The present invention provides a kind of IOS and the general gait auto-correlation pedometer of Android platform, and it is included with lower module:
Frequency setting module, the frequency of sensing data is gathered for setting mobile phone, and the sensor refers to acceleration transducer;Filter
Ripple device module, the filter module is filtered noise reduction process for the feature that people walks to original sensor data;Processing
Device module, the processor module designs dynamic gait auto-correlation algorithm according to the walk frequency scope of people, described by judging
Meter step effect is periodically realized in the vibration amplitude of acceleration transducer and vibrations.
It is preferred that, the filter module be Butterworth LPF, the Butterworth LPF according to
The cadence design that people normally walks, the Butterworth LPF is filtered at noise reduction to the acceleration information of synthesis
Reason, improves meter step precision.
It is preferred that, the dynamic gait auto-correlation algorithm utilizes the stability Design of everyone itself walking habits, without
It is pre-configured with user's walking behavioral data.
It is preferred that, the processor module is from initial timeStart, cadence of normally being walked according to people setting time frame
Shortest time and maximum duration, successively calculate the shortest time until maximum duration in acceleration information with it is next identical when
Between acceleration information correlation size, every time circulation in choice of dynamical correlation maximum time frame length。
It is preferred that, the processor module is calculatedThe standard deviation of acceleration information and synthesized in time with the next time
The coefficient correlation of acceleration, judge whether to reach threshold value, if reaching, be calculated as 1 step, otherwise disregard step.
It is preferred that, after above three module is realized by C language code, IOS systems are supported to directly invoke this method
C code realizes meter step, and Android system also can call the C code of this method to realize step function after being bridged using JNI.
The invention provides the step-recording method that a kind of IOS and Android platform are general, following steps are specifically included:
Step 1, mobile phone sample frequency is set, and processing is sampled to mobile phone gathered data, to stablize data output frequencies;
Step 2, following processing is done to acceleration transducer data:
2-1)Euclidean distance of the original acceleration sensing data away from origin is calculated as resultant acceleration signal to eliminate mobile phone
The influence that any attitude is brought;
2-2)Low pass filter of the design suitable for pedometer;
2-3)Processing is filtered to the acceleration signal of synthesis to eliminate high frequency background noise;
Step 3, filtered data are carried out plus frame Similarity measures step effect in terms of realizing, specifically included:
3-1)Shortest time frame and maximum duration frame that cadence of normally being walked according to people is dynamically set;
3-2)Since constantly, successively calculate arrive untilArriveAcceleration information adds with next same time in time
The similarity of speed data and the period acceleration magnitude standard deviation;
3-3)The time frame length of correlation maximum is selected, is determinedWith to resultant acceleration similarity and arrivingWhether resultant acceleration standard deviation reaches threshold value in time, if reached, counts a step, otherwise disregards step;
3-4)FromMoment starts to repeat 3-2)And 3-3)Step realizes complete meter step.
As the further prioritization scheme of the present invention, step 1 uses recurrence method to be sampled to stablize number to initial data
According to frequency;
As the further prioritization scheme of the present invention, Butterworth low pass is constructed according to people's normal gait/running frequency in step 2
Wave filter realizes noise reduction;
As the further prioritization scheme of the present invention, in step 3 shortest time of time frame and maximum duration by people's normal gait/
Most fast cadence of running and normal cadence are determined;
As the further prioritization scheme of the present invention, Similarity Measure uses Pearson correlation coefficients algorithm in step 3;
As the further prioritization scheme of the present invention, step 3 continuously can often walk the time to userLearnt, judge user
Walking custom step accuracy rate in terms of improving;
Compared with prior art, the present invention is by contrasting different length time frame accelerometer data first for the above-mentioned each side of the present invention
The maximum time frame length of similitude dynamic select similarity measured as long step by step, changing cadence even if user also can be high
Precision statisticses step number;Secondly, the present invention is trained without preset user behavior data, improves the practicality of the present invention;The
Three, the present invention persistently can also learn to user's gait, make meter step more accurate;Finally, the present invention is real by C language code
After now, IOS systems support the C code for directly invoking this method to realize that meter step, Android system also may be used after using JNI bridge joints
The C code of this method is called to realize meter step, it is not necessary to extra hardware device.
Brief description of the drawings
Fig. 1 is to illustrate flow chart of the method for the present invention.
Fig. 2 is the contrast for illustrating present invention design wave filter using front and rear resultant acceleration value.
Embodiment
1 embodiment for introducing the present invention below in conjunction with the accompanying drawings, in figure each square frame represent One function module or
One section of executable code, these modules are that occur according to mark order in figure, in addition, all modules can pass through code in figure
Realized on IOS and Android platform.
The frequency of data in mobile phone is determined first, and ios platform can directly set the sensing data frequency of needs,
Android platform sensor frequency is divided into Normal, UI, Game and Fast fourth gear, corresponding frequency respectively 5HZ, 15HZ,
50HZ and 100HZ or so, therefore for ios device, initialization system sample frequency is 30HZ, sets to come for android
Say, initialization system sample frequency is Game;
Secondly obtained data of being sampled to mobile phone are done recurrence and sampled to stablize data frequency as 15HZ, so-called recurrence sampling, will
Sampling carries out recurrence decomposition.Such as actual certain second frequency of system returned data is 31HZ, then is front and rear 15 by sub-sample resolution
Individual point respectively adopts 7 points, retains intermediate point;15 points, which adopt 7 points and can further be decomposed into front and rear 7 points, adopts 3 points, in reservation
Between point;7 points, which adopt 3 points and can be further broken into front and rear 3 points, respectively adopts 1 point, retains intermediate point;3 points, which adopt 1 point, to be turned
Reservation central point is changed to, such recurrence sampling is completed.If system returned data second frequency is less than 15HZ, this second factor data frequency
Rate is too low to disregard step;
The acceleration information of system acquisition includes the axle of x, y and z tri-, and user's mobile phone when walking is likely to be at any appearance
State, therefore 3-axis acceleration data are synthesized, composite formula for calculate 3-axis acceleration data apart from origin it is European away from
From that is,.
Design Butterworth LPF, Butterworth LPF can use following amplitude square to the public affairs of frequency
Formula is represented:
Wherein,It is filter cut-off frequencies, n is filter order.Calculation formula with n is respectively:
Pedometer filter cut-off frequencies wave filter key parameter is as follows:Sample frequency is set to 15HZ, and cut-off frequecy of passband is
2.5HZ, pass band dampingLess than 0.5db, stopband initial point frequencyFor 5HZ, stopband attenuation is more than 150db.Judge this section of mobile phone
Data time length is collected, if less than 1 second, data time is too short to disregard step because collecting, if higher than 1 second, accelerating to synthesis
Degrees of data is filtered denoising.Because the frequency of people's normal gait is between 1 to 2.5HZ, present invention design wave filter can effectively be gone
Except high frequency background noise, filter effect is as shown in Figure 2.
Next dynamic autocorrelation analysis is done to the resultant acceleration data after denoising, 10 calculating of often step circulation, circulation
The Pearson correlation coefficients of the 1st to 6 point and the 7th to 12 point are calculated for the first time(Corresponding cadence is 2.5HZ), circulation i-th
({i|0<i<11})The 1st to 6+i-1 point of secondary calculating and 6+i to 2*(6+i-1), circulate the 1st to 15 point of the 10th calculating
With the 16th to 30 Point correlation coefficient(Corresponding cadence is 1HZ), the circulation of coefficient correlation maximum jth time is taken as output knot
Really, record jth time circulation coefficient correlation and the standard deviation of resultant acceleration the 1st to 6+j-1 point of data;
Above-mentioned circulation is repeated since the 6+j point, the Pearson correlation coefficients and standard deviation of each period is calculated, takes phase
Relation number maximum time is used as output result;
The above results are judged, if auto-correlation coefficient is more than 0.7 while standard deviation is calculated as a step, otherwise more than 0.5
Disregard step;
It is because a people is in each walking process to need exist for calculating Pearson correlation coefficients, it may appear that one accelerates and one
It is individual to slow down, and for same person, cadence is stablized relatively, therefore can use from the data for accelerating to this process of slowing down
Sequence similarity completes periodically meter step, it is to avoid stands/sits down because of user, the erroneous judgement caused by the operation such as taking and placing mobile phone;
Why dynamically to determine time window in calculating process, be because the frequency of each individual walking and different,
The frequency that same individual is walked under different scenes is also and different, but people normally walks(Including running)Frequency nearly all
In 1 to 2.5HZ frequency range, therefore dynamic determination window can be in the situation without preset user's gait data
Lower progress meter step, also meets the requirement of the accurate meter step after the change of user's gait under different scenes.
In addition, the present invention through after a while to the study of user's gait data after, duration 6+j can be often walked to user
Scope make estimation, this estimation on the one hand can improve meter step precision, for example exclude it is too low/excessively it is high-frequency meter step, separately
Outer one side can also reduce the cycle-index in dynamic autocorrelation analysis, further the consumption of reduction computing resource.
It will also be understood by those skilled in the art that above-mentioned step count set includes but is not limited to this, it is any to understand the present invention's
Technical staff the invention discloses technical scope within, it is possible to understand that or the conversion and replacement expected, should all cover
Within protection scope of the present invention.
Claims (18)
1. a kind of IOS and the general gait auto-correlation pedometer of Android platform, it is characterised in that including with lower module:
Frequency setting module, the frequency of sensing data is gathered for setting mobile phone, and the sensor refers to acceleration sensing
Device;
Filter module, the filter module is filtered at noise reduction for the feature that people walks to original sensor data
Reason;
Processor module, the processor module designs dynamic gait auto-correlation algorithm according to the walk frequency scope of people, passes through
The vibration amplitude and vibrations for judging the acceleration transducer periodically realize meter step effect.
2. gait auto-correlation pedometer according to claim 1, it is characterised in that:The filter module is Butterworth
Low pass filter, the cadence that the Butterworth LPF is normally walked according to people is designed, the Butterworth low pass
Ripple device is filtered noise reduction process to the acceleration information of synthesis, improves meter step precision.
3. gait auto-correlation pedometer according to claim 1, it is characterised in that the dynamic gait auto-correlation algorithm profit
With the stability Design of everyone itself walking habits, without being pre-configured with user's walking behavioral data.
4. gait auto-correlation pedometer according to claim 1, it is characterised in that the processor module is from initial timeStart, the shortest time of cadence of normally being walked according to people setting time frame and maximum duration, calculate the shortest time successively until
The correlation of acceleration information and acceleration information in next same time period in maximum duration, by contrasting during different length
Between the maximum time frame length of similitude size selection similarity of frame accelerometer data be used as the time frame of long metering step by step
Length。
5. gait auto-correlation pedometer according to claim 4, it is characterised in that the processor module is calculatedTime
The standard deviation of interior acceleration informationAnd with the next timeThe coefficient correlation of resultant acceleration, judges whether to reach threshold
Value, if reaching, is calculated as 1 step, otherwise disregards step.
6. according to the gait recognition method using the gait auto-correlation pedometer described in claim 1, it is characterised in that including with
Lower step:
First, the frequency that mobile phone gathers sensing data is set using frequency setting module, the sensor refers to acceleration
Sensor;
Secondly, noise reduction process is filtered to original sensor data for the feature that people walks using the filter module;
Finally, dynamic auto-correlation algorithm is designed according to the walk frequency scope of people using processor module, by judging sensor
Meter step effect is periodically realized in vibration amplitude and vibrations.
7. gait recognition method according to claim 6, it is characterised in that also include:
The data that mobile phone sensor is gathered are carried out recurrence sampling processing to stablize data frequency.
8. gait recognition method according to claim 6, it is characterised in that also include:
Synthesis processing is carried out to the original 3-axis acceleration data for stablizing frequency to eliminate because mobile phone any attitude puts the shadow brought
Ring, composite formula is。
9. the gait recognition method according to one of claim 6-8, it is characterised in that further comprise:
The cadence design Butterworth LPF normally walked according to people is filtered noise reduction to the acceleration information of synthesis
Processing, improves meter step precision.
10. gait recognition method according to claim 6, it is characterised in that further comprise:
Since initial time, the shortest time of cadence of normally being walked according to people setting time frame and maximum duration are calculated successively
Shortest time is until the correlation of the acceleration information and the acceleration information of next same time in maximum duration, choice of dynamical
The time frame length of correlation maximum.
11. gait recognition method according to claim 10, it is characterised in that further comprise:
The standard deviation of acceleration information and the coefficient correlation with next time resultant acceleration in the calculating time, judge
Whether threshold value is reached, if reaching, be calculated as 1 step, otherwise disregard step.
12. gait recognition method according to claim 6, it is characterised in that stable frequency is carried out to initial data and is taken out
Sample processing, it is to avoid mobile phone sensor frequency is unstable to cause the problem of meter step precision is low.
13. gait recognition method according to claim 6, it is characterised in that the Frequency Design walked according to people is applied to
The Butterworth LPF of step is counted, ambient noise is reduced, meter step accuracy is improved.
14. gait recognition method according to claim 6, it is characterised in that utilize the steady of everyone itself walking habits
Qualitative design dynamic gait auto-correlation algorithm, without being pre-configured with user's walking behavioral data.
15. gait recognition method according to claim 6, it is characterised in that received by the circuit-switched data of walking several times of user
After collection, this method can learn user and normally walk cadence, further improve meter step accuracy rate.
16. gait recognition method according to claim 1, it is characterised in that this method is realized it by C language code
Afterwards, IOS systems can directly invoke the C code of this method, and Android system can also call the C of this method after being bridged using JNI
Code.
17. a kind of IOS and the general step-recording method of Android platform, specifically include following steps:
Step 1, mobile phone sample frequency is set, and processing is sampled to mobile phone gathered data, to stablize data output frequencies;
Step 2, following processing is done to acceleration transducer data:
2-1)Euclidean distance of the original acceleration sensing data away from origin is calculated as resultant acceleration signal to eliminate mobile phone
The influence that any attitude is brought;
2-2)Low pass filter of the design suitable for pedometer;
2-3)Processing is filtered to the acceleration signal of synthesis to eliminate high frequency background noise;
Step 3, filtered data are carried out plus frame Similarity measures step effect in terms of realizing.
18. step-recording method according to claim 17, it is characterised in that the step 3 is specifically included:
3-1)The shortest time frame that cadence of normally being walked according to people is dynamically setWith maximum duration frame;
3-2)Since constantly, calculate arrive successivelyUntil accelerating to acceleration information in the time and next same time
The similarity of degrees of data and the period acceleration magnitude standard deviation;
3-3)The time frame length of correlation maximum is selected, is judgedTo withArriveResultant acceleration similarity
And whether threshold value is reached to resultant acceleration standard deviation in the time, if reached, a step is counted, step is otherwise disregarded;
3-4)3-2 is repeated since constantly)And 3-3)Step realizes complete meter step.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611207910.2A CN107270931A (en) | 2016-12-23 | 2016-12-23 | A kind of IOS and the general gait auto-correlation pedometer of Android platform |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611207910.2A CN107270931A (en) | 2016-12-23 | 2016-12-23 | A kind of IOS and the general gait auto-correlation pedometer of Android platform |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107270931A true CN107270931A (en) | 2017-10-20 |
Family
ID=60052204
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611207910.2A Pending CN107270931A (en) | 2016-12-23 | 2016-12-23 | A kind of IOS and the general gait auto-correlation pedometer of Android platform |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107270931A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108195395A (en) * | 2017-11-17 | 2018-06-22 | 捷开通讯(深圳)有限公司 | Mobile terminal and its step-recording method, storage device |
CN109141465A (en) * | 2018-07-19 | 2019-01-04 | 歌尔科技有限公司 | A kind of step-recording method, wearable device and computer readable storage medium |
CN109189221A (en) * | 2018-08-23 | 2019-01-11 | 郑州航空工业管理学院 | A kind of user behavior recognition method across cell phone platform |
CN112484747A (en) * | 2020-12-08 | 2021-03-12 | 北京小米松果电子有限公司 | Step counting method, step counting device and storage medium |
CN113008242A (en) * | 2021-03-19 | 2021-06-22 | 深圳市慧鲤科技有限公司 | User behavior data processing method, device, equipment and storage medium |
CN113295182A (en) * | 2021-05-20 | 2021-08-24 | 北京智慧图科技有限责任公司 | Wi-Fi SLAM track restoration method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012048530A (en) * | 2010-08-27 | 2012-03-08 | Yamaha Corp | Step number measuring device |
CN103997572A (en) * | 2014-06-03 | 2014-08-20 | 深圳市爱康伟达智能医疗科技有限公司 | Step counting method and device based on data of acceleration sensor of mobile phone |
CN104406604A (en) * | 2014-11-21 | 2015-03-11 | 中国科学院计算技术研究所 | Step counting method |
CN104580532A (en) * | 2015-02-05 | 2015-04-29 | 中国农业银行股份有限公司 | Cross-platform application system |
CN105043410A (en) * | 2015-06-30 | 2015-11-11 | 深圳市元征科技股份有限公司 | Pedometer |
-
2016
- 2016-12-23 CN CN201611207910.2A patent/CN107270931A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012048530A (en) * | 2010-08-27 | 2012-03-08 | Yamaha Corp | Step number measuring device |
CN103997572A (en) * | 2014-06-03 | 2014-08-20 | 深圳市爱康伟达智能医疗科技有限公司 | Step counting method and device based on data of acceleration sensor of mobile phone |
CN104406604A (en) * | 2014-11-21 | 2015-03-11 | 中国科学院计算技术研究所 | Step counting method |
CN104580532A (en) * | 2015-02-05 | 2015-04-29 | 中国农业银行股份有限公司 | Cross-platform application system |
CN105043410A (en) * | 2015-06-30 | 2015-11-11 | 深圳市元征科技股份有限公司 | Pedometer |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108195395A (en) * | 2017-11-17 | 2018-06-22 | 捷开通讯(深圳)有限公司 | Mobile terminal and its step-recording method, storage device |
CN109141465A (en) * | 2018-07-19 | 2019-01-04 | 歌尔科技有限公司 | A kind of step-recording method, wearable device and computer readable storage medium |
CN109189221A (en) * | 2018-08-23 | 2019-01-11 | 郑州航空工业管理学院 | A kind of user behavior recognition method across cell phone platform |
CN109189221B (en) * | 2018-08-23 | 2021-07-16 | 郑州航空工业管理学院 | User behavior identification method across mobile phone platforms |
CN112484747A (en) * | 2020-12-08 | 2021-03-12 | 北京小米松果电子有限公司 | Step counting method, step counting device and storage medium |
CN113008242A (en) * | 2021-03-19 | 2021-06-22 | 深圳市慧鲤科技有限公司 | User behavior data processing method, device, equipment and storage medium |
CN113295182A (en) * | 2021-05-20 | 2021-08-24 | 北京智慧图科技有限责任公司 | Wi-Fi SLAM track restoration method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107270931A (en) | A kind of IOS and the general gait auto-correlation pedometer of Android platform | |
AU2020273327B2 (en) | Systems and methods of swimming analysis | |
US10687707B2 (en) | Detecting activity by a wheelchair user | |
CN103884337B (en) | Multi-attitude step correction positioning system and method | |
JP6183906B2 (en) | Gait estimation device and program, fall risk calculation device and program | |
JP2016039999A (en) | Devices, methods, and apparatuses for inferring position of mobile device | |
CA2615214A1 (en) | Method and device for measuring the progress of a moving person | |
WO2017177582A1 (en) | Method and device for implementing speed measurement of sports apparatus | |
CA2615211A1 (en) | Method and device for measuring the progress of a moving person | |
US20190076063A1 (en) | Systems and methods of ski activity detection | |
FR2912813A1 (en) | MAGNETIC MEASUREMENT OF CADENCE AND ORIENTATION OF THE MOVEMENT OF AN OBJECT | |
CN107449418A (en) | User's walking direction calculating method based on accelerometer and magnetometer | |
CN112504295B (en) | Riding detection method, electronic device and computer readable storage medium | |
CN107277222A (en) | User behavior state judging method based on mobile phone built-in sensors | |
CN113713358A (en) | Swimming monitoring method, device, storage medium and program product based on multi-sensor fusion | |
US11051720B2 (en) | Fitness tracking for constrained-arm usage | |
JP6416722B2 (en) | Step counting device, step counting method, and program | |
CN110916639B (en) | Method, system, wearable device and computer-readable storage medium for acquiring exercise heart rate recovery rate | |
CN108072386B (en) | Step counting method and device | |
CN106500717B (en) | Method and device for realizing step counting | |
CN114341947A (en) | System and method for exercise type recognition using wearable devices | |
WO2014191803A1 (en) | Acceleration-based step activity detection and classification on mobile devices | |
CN109061215A (en) | A kind of speed detection method and wearable device based on wearable device | |
Truong et al. | Nonparametric regression-based step-length estimation for arm-swing walking using a smartphone | |
CN108981744B (en) | Step frequency real-time calculation method based on machine learning and low-pass filtering |
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: 20171020 |
|
RJ01 | Rejection of invention patent application after publication |