CN104089625A - Step counting method with self-learning capability - Google Patents
Step counting method with self-learning capability Download PDFInfo
- Publication number
- CN104089625A CN104089625A CN201410371964.7A CN201410371964A CN104089625A CN 104089625 A CN104089625 A CN 104089625A CN 201410371964 A CN201410371964 A CN 201410371964A CN 104089625 A CN104089625 A CN 104089625A
- Authority
- CN
- China
- Prior art keywords
- threshold value
- meter
- data
- recording method
- learning capability
- 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 relates to a step counting method with self-learning capability. The step counting method is characterized in that the walking speed of a current user is automatically calculated and is compared with that of ordinary people in a natural state, and a step counting threshold value is automatically changed according to the deviation condition, so that the effect of accurately counting the steps can be realized. The method comprises the following steps of calculating the step number of the current user within 5 minutes or longer time according to a preset step counting threshold value, comparing the calculated step number with the step number range of ordinary people in the natural state in the same period, automatically correcting the step counting threshold value according to the deviation condition, and repeating the process until the deviation is 0. The step counting method with self-learning capability is suitable for multiple sensors, is low in requirement on wearing, and is still capable of accurately counting the steps after the user is replaced.
Description
Technical field
The present invention relates to a kind of step-recording method, especially, a kind of step-recording method that possesses self-learning capability.
Background technology
Along with social development and progress, intelligent wearable product progressively enters into daily life, becomes the easy way of motion and sleep detection.For example, the motion bracelet of most, all possesses step function, the total step number of the metering daily walking of user and running.
And at present main motion bracelet and meter step APP are used two axles or 3-axis acceleration sensor image data conventionally, and use default a few batch total step threshold values to calculate the data that collect, draw meter step information.This step-recording method has following shortcoming:
1. pre-designed step threshold value is limited several groups, cannot adapt to all people group, and meter step error is larger;
2. couple user the having relatively high expectations of attitude of walking, the attitude of walking is nonstandard or when different from self common attitude, meter step error is larger;
3. step count set/software is changed after different users, need to reselect the meter step threshold value that is applicable to oneself, more loaded down with trivial details in operation.
Therefore, existing step-recording method needs to improve.
Summary of the invention
Deficiency in view of above-mentioned existing step-recording method, the object of the present invention is to provide a kind of step-recording method that possesses self-learning capability, make the step count set/software of various employing sensors, can count more exactly step, serve widely crowd, request for utilization is lower.
Technical scheme of the present invention is as follows:
A step-recording method that possesses self-learning capability, comprising step:
A. data rule of thumb in advance, arrange a meter step threshold value, make step count set/software to go out certain step number according to this threshold calculations;
B. according to this predetermined threshold value, calculate user 5 minutes or the step number in the more time, and with ordinary people's state of nature under same period in step number scope compare, according to deviation situation, predetermined threshold value is adjusted in the same way;
C. according to the threshold value after adjusting, calculate the step number in user's next cycle, and again with ordinary people's state of nature under same period in step number scope compare, according to deviation situation, predetermined threshold value is adjusted in the same way;
D. repeat the step of above-mentioned C, until meter step data meets the step number scope in the same period under ordinary people's state of nature.
The described step-recording method with self-learning capability, wherein, meter step threshold value default in described steps A is the empirical value by the statistical study of a large amount of meter step datas is obtained, and its objective is and makes initial calculation result approach target interval as far as possible, accelerates self study speed.
The described step-recording method with self-learning capability, wherein, the meter step threshold value in described steps A is an empirical data, according to different sensor types and computing method, can have different numerical value.
The described step-recording method with self-learning capability, wherein, the computing method in described steps A there is no concrete restriction, according to the difference of sensor type, can be peak value, amplitude, frequency, waveform, add/retarded velocity etc.
The described step-recording method with self-learning capability, wherein, the step number scope in the same period under the ordinary people's state of nature in described steps A is an empirical value, can have different numerical value according to different countries, area, ethnic group, age.
The described step-recording method with self-learning capability, wherein, described step B specifically comprises:
B1, to sensor data samples, get 5 sampled values p.s. (in different embodiments, sampling numerical value there is no concrete restriction, can be 3,4,5,6,7 etc.), and calculate according to the data of 5 sampled values and predetermined threshold value the step number that this second kind user passes by;
B2, a totalizer of use, the data accumulation of each second in 5 minute cycle is got up (in different embodiments, computation period there is no concrete restriction, can be 5 minutes, 6 minutes, 7 minutes, 8 minutes, 9 minutes, 10 minutes etc.), obtain the total step number calculating in this period of cycle;
B3, by the total step number calculating, compare with the step number scope in same period under ordinary people's state of nature, if total step number is greater than the upper limit of this scope, improve meter step threshold value; If total step number is less than the lower limit of this scope, reduce meter step threshold value.
The described step-recording method with self-learning capability, wherein, the step number scope in the same period under the ordinary people's state of nature in described step B is the empirical value by the statistical study of a large amount of ordinary people's walking speeds is drawn.
The described step-recording method with self-learning capability, wherein, the threshold value after the adjustment in described step C, is the meter step threshold value after one or many is adjusted, irrelevant with predetermined threshold value.
The described step-recording method with self-learning capability, wherein, described step D specifically comprises:
The step of D1, repetition C, constantly adjusts meter step threshold value, and the step data that makes to succeed in one's scheme reaches target interval;
D2, store current meter step threshold value, and replace pre-designed step threshold value, as follow-up meter step threshold value, use.
The described step-recording method with self-learning capability, wherein, the meter step threshold value that described storage is current, is the long-term current meter step threshold value of preserving, and when sensor and user do not change, can use this meter step threshold value, without extra computation and adjustment always.
The described step-recording method with self-learning capability, wherein, the repetition above-mentioned steps in described step D, is one or many adjustment meter step threshold value, there is no the restriction of concrete number of times.
Compared with prior art, the invention has the beneficial effects as follows:
1, the present invention adopts the meter step sum of statistics user within one period of cycle, compare and dynamically adjust meter step threshold value with ordinary people's walking speed scope, because ordinary people's the large probability of walking speed is located in same scope interval, the method adapts to different crowds more widely than existing meter step technology, has higher accuracy of detection;
2, the present invention is the self study step-recording method based on result of calculation, and the attitude that user is walked no requirement (NR) do not require that user remains same attitude, has advantage than prior art aspect hommization yet;
3, the algorithm that the present invention adopts has self-learning capability, when the user of step count set/software changes, can automatic learning be applicable to new user's meter step threshold value, uses convenient.
Accompanying drawing explanation
Fig. 1 is the process flow diagram that possesses the step-recording method embodiment of self-learning capability in the present invention.
Embodiment
The invention provides a kind of step-recording method that possesses self-learning capability, for making the object of the invention, technical scheme and effect more clear, clear and definite, referring to accompanying drawing examples, the present invention is described in more detail.Should be appreciated that, specific embodiment described herein only, in order to explain the present invention, is not intended to limit the present invention.
The present invention possesses the step-recording method of self-learning capability, by the walking speed scope of calculating gained step number and ordinary people is compared, according to deviation, automatically adjusts meter step threshold value, thereby realizes accurately meter step, hommization operation and require lower to user.
As shown in Figure 1, Fig. 1 is an embodiment of the present invention's step-recording method of possessing self-learning capability, and it mainly comprises the following steps:
Step S100, according to pre-designed step threshold value, calculates the meter step data of some cycles, and it specifically comprises:
A, a totalizer is set, and the initial value of totalizer is set to 0.
B, first, the data that step count set/software is passed back sensor are sampled, the sampling period be every 1 second once sampling, the 3 groups of data of at every turn sampling.
C, according to default meter step threshold value, calculate this 3 groups of step numbers that data are corresponding.
D, these 3 groups of step numbers corresponding to data are added in totalizer respectively.
E, computation period are 5 minutes, each second of 5 minutes, successively the data that sample are carried out to meter step and calculate and accumulation calculating.Obtain the total meter step data of user in 5 minute cycle.
Wherein, in step S100, default meter step threshold value can be any or combination in sensor peak value, amplitude, frequency, waveform, the add/indexs such as retarded velocity.A concrete example is that one group of peak-data that hypothesis 3 axle sensors are passed back is respectively X, Y, Z, and default meter step threshold value is V, works as so X
2+ Y
2+ Z
2during > V, be just designated as a step.In the present invention, pre-designed step threshold value can be set voluntarily by step count set/software vendor, to adapt to different sensors, countries and regions, target group etc.
Wherein, in the present invention, computation period is not limited to 5 minutes, it can be any number, this numerical value can be set voluntarily by step count set/software vendor, the setting of this computation period need to be considered contingent pause in ordinary people's motion process, give way, plays the situations such as phone for motor vehicle, make the final computation period of setting can cover common fortuitous event, reach better statistics effect.
Wherein, in the present invention, sampling period and sampled data can be any values, do not limit, and in the present embodiment, the sampling period was 1 second, and sampled data is 3 groups of data, and just example, can increase or reduce in actual implementation process.
Step S102, judges whether to fall into pre-set interval.
In this step, the meter step data obtaining in step S100 and default ordinary people's walking speed interval is compared, first meter step data and interval lower limit are compared, if meter step data is less than pre-set interval lower limit, explanation meter step threshold value needs to adjust downwards, and do not need to carry out again other relatively, directly perform step S104; If meter step data is greater than pre-set interval lower limit, continue meter step data and the pre-set interval upper limit to compare, if meter step data is less than or equal to the pre-set interval upper limit, illustrate that current meter step threshold value is applicable to user's actual conditions, execution step S106; If meter step data is greater than the pre-set interval upper limit, explanation meter step threshold value need to adjust upward, execution step S104.
Wherein, default ordinary people's walking speed interval is an empirical value, and object is that the scope of self study is realized to constraint, makes the final meter step threshold value after self study can calculate rational meter step data simultaneously, makes it meet ordinary people's walking speed.
Step S104, adjusts meter step threshold value, calculates the data of some cycles.
In this step, the adjustment direction of calculating according to step S104, adjusts meter step threshold value, accurate in order to guarantee to adjust effect, and the present invention interval that step threshold value adjusts of falling into a trap is less, by fine tune repeatedly, finally reaches the effect of accurate meter step.Meter step threshold value is used the method for mentioning in step S100 to recalculate the meter step data of 5 minutes after adjusting.
In the present embodiment, take at every turn meter step threshold value to adjust upward 10 or adjust 10 downwards, and it is calculated as new meter step threshold value.
The fall into a trap adjustment threshold value of step threshold value of the present invention can also according to circumstances be set voluntarily by user, with person easy to use, according to displacement state, accelerates self study speed.
Step S106, preserving current meter step threshold value is predetermined threshold value, normal meter step.
In this step, the meter step threshold value after adjusting is for the last time charged to step count set/software, replace default meter step threshold value, and persistence, and in this step threshold value meter step, until new deviation appears in meter step data, again enter new self-studying mode.
Wherein, the new deviation of meter step data can be triggered by user's change, also can be triggered by the change of user's environment of living in, can also be triggered by the change of user's mode of motion.
In sum, the present invention possesses the step-recording method of self-learning capability, by calculating the meter step data in some cycles, and compare with a step interval, when meter step data is beyond interval range, automatically adjust meter step threshold value, the step data that makes to succeed in one's scheme walks interval convergence to target meter, and is finally consistent.The present invention has avoided different users to count the inaccurate problem of step data, user only need use a period of time, can obtain and count accurately step data, and, can also be after user changes automatic learning adapt to new user, without any manual intervention, still can obtain and count accurately step data.
Should be understood that, application of the present invention is not limited to above-mentioned giving an example, and for those of ordinary skills, can be improved according to the above description or convert, and all these improvement and conversion all should belong to the protection domain of claims of the present invention.
Claims (10)
1. a step-recording method that possesses self-learning capability, is characterized in that, comprises step:
A, data rule of thumb in advance, arrange a meter step threshold value, makes step count set/software to go out certain step number according to this threshold calculations;
B, according to this predetermined threshold value, calculate the step number in the certain computation period of user, and with ordinary people's state of nature under same period in step number scope compare, according to deviation situation, predetermined threshold value is adjusted in the same way;
C, according to the threshold value after adjusting, calculate the step number in the next computation period of user, and again with ordinary people's state of nature under same period in step number scope compare, according to deviation situation, predetermined threshold value is adjusted in the same way;
D, repeat the step of above-mentioned C, until meter step data meets the step number scope in the same period under ordinary people's state of nature.
2. the step-recording method that possesses self-learning capability according to claim 1, is characterized in that, its step threshold value of falling into a trap is self-adjusting according to the deviation of user's meter step data and a predetermined interval, without manual intervention.
3. the step-recording method that possesses self-learning capability according to claim 1, is characterized in that, after user changes, meter step threshold value can relearn automatically, without manual intervention.
4. the step-recording method that possesses self-learning capability according to claim 1, is characterized in that, the adjustment of meter step threshold value is the step data of usining in some cycles as foundation, during for user, there is no specific (special) requirements.
5. the step-recording method that possesses self-learning capability according to claim 1, is characterized in that, the adjustment of meter step threshold value is the step data of usining in some cycles as foundation, and the kind of sensor and precision be there is no to specific (special) requirements.
6. the step-recording method that possesses self-learning capability according to claim 1, is characterized in that, meter in described steps A step threshold value is a kind of in peak value, amplitude, frequency, waveform, add/retarded velocity.
7. the step-recording method that possesses self-learning capability according to claim 1, is characterized in that, described step B specifically comprises:
B1, to sensor data samples, within certain sampling period, obtaining the data to determined number, and the data and the predetermined threshold value that according to sampling, obtain are calculated the step number that this sampling period user passes by;
B2, a totalizer of use, get up the data accumulation in each sampling period in computation period, obtains the total step number calculating in this section of computation period;
B3, by the total step number calculating, compare with the step number scope in same period under ordinary people's state of nature, if total step number is greater than the upper limit of this scope, improve meter step threshold value; If total step number is less than the lower limit of this scope, reduce meter step threshold value.
8. the step-recording method that possesses self-learning capability according to claim 1, is characterized in that, described step D specifically comprises:
The step of D1, repetition C, constantly adjusts meter step threshold value, and the step data that makes to succeed in one's scheme reaches target interval;
D2, store current meter step threshold value, and replace pre-designed step threshold value, as follow-up meter step threshold value, use.
9. the step-recording method that possesses self-learning capability according to claim 7, is characterized in that, wherein sensor can be acceleration transducer, gravity sensor, 3 axle gyro sensors.
10. the step-recording method that possesses self-learning capability according to claim 8, is characterized in that, wherein stores current meter step threshold value, it is the long-term current meter step threshold value of preserving, when sensor and user do not change, can use this meter step threshold value, without extra computation and adjustment always.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410371964.7A CN104089625A (en) | 2014-07-31 | 2014-07-31 | Step counting method with self-learning capability |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410371964.7A CN104089625A (en) | 2014-07-31 | 2014-07-31 | Step counting method with self-learning capability |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104089625A true CN104089625A (en) | 2014-10-08 |
Family
ID=51637363
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410371964.7A Pending CN104089625A (en) | 2014-07-31 | 2014-07-31 | Step counting method with self-learning capability |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104089625A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105004349A (en) * | 2015-06-30 | 2015-10-28 | 深圳市元征科技股份有限公司 | Step calculation method |
| CN105698815A (en) * | 2016-03-24 | 2016-06-22 | 广东欧珀移动通信有限公司 | Method and device for regulating step counting data |
| WO2016165333A1 (en) * | 2015-09-08 | 2016-10-20 | 中兴通讯股份有限公司 | Method and apparatus for realizing step counting |
| CN109173147A (en) * | 2018-04-11 | 2019-01-11 | 北京慧动众人科技有限公司 | A kind of electric treadmill step-recording method |
| CN111189469A (en) * | 2019-12-31 | 2020-05-22 | 歌尔科技有限公司 | Step counting method, terminal device and storage medium |
| CN111879333A (en) * | 2020-07-30 | 2020-11-03 | 歌尔科技有限公司 | Motion determination method, determination device, and computer-readable storage medium |
-
2014
- 2014-07-31 CN CN201410371964.7A patent/CN104089625A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105004349A (en) * | 2015-06-30 | 2015-10-28 | 深圳市元征科技股份有限公司 | Step calculation method |
| WO2016165333A1 (en) * | 2015-09-08 | 2016-10-20 | 中兴通讯股份有限公司 | Method and apparatus for realizing step counting |
| CN106500717A (en) * | 2015-09-08 | 2017-03-15 | 中兴通讯股份有限公司 | A kind of method and device for realizing counting step |
| CN106500717B (en) * | 2015-09-08 | 2020-05-12 | 中兴通讯股份有限公司 | Method and device for realizing step counting |
| CN105698815A (en) * | 2016-03-24 | 2016-06-22 | 广东欧珀移动通信有限公司 | Method and device for regulating step counting data |
| CN105698815B (en) * | 2016-03-24 | 2018-07-27 | 广东欧珀移动通信有限公司 | Count the adjusting method and device of step data |
| CN109173147A (en) * | 2018-04-11 | 2019-01-11 | 北京慧动众人科技有限公司 | A kind of electric treadmill step-recording method |
| CN111189469A (en) * | 2019-12-31 | 2020-05-22 | 歌尔科技有限公司 | Step counting method, terminal device and storage medium |
| CN111879333A (en) * | 2020-07-30 | 2020-11-03 | 歌尔科技有限公司 | Motion determination method, determination device, and computer-readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104089625A (en) | Step counting method with self-learning capability | |
| CN103712632B (en) | A kind of step-recording method based on 3 axle accelerometers and pedometer | |
| CN104197952A (en) | User walk step counting method and device and mobile terminal | |
| CN103954295B (en) | A kind of step-recording method based on acceleration transducer | |
| CN102445214B (en) | Step counting method based on 3D (three-dimensional) gravity sensor of intelligent mobile phone | |
| Gidófalvi et al. | When and where next: Individual mobility prediction | |
| CN108810824B (en) | Mobile terminal positioning method, system, mobile terminal and readable storage medium | |
| CN104964685B (en) | A kind of decision method of mobile phone athletic posture | |
| JP6134872B1 (en) | Device, method and system for counting the number of cycles of periodic motion of a subject | |
| CN102455427B (en) | Method for correcting global position system (GPS) offset | |
| CN103425895A (en) | Multi-sensor motion and posture monitoring and analyzing method and system | |
| KR20150086293A (en) | Systems and methods for frequency-based stride length correction in a pedometer device | |
| CN106840163A (en) | A kind of indoor orientation method and system | |
| CN106017502A (en) | Step counting method and electronic device | |
| CN111415752A (en) | Hand-foot-and-mouth disease prediction method integrating meteorological factors and search indexes | |
| CN109844460A (en) | A kind of stride calibration method, relevant device and system | |
| CN103892840A (en) | Intelligent wearing device and method for extracting human body motion features | |
| CN106096047A (en) | User partition preference computational methods based on Information Entropy and system | |
| CN104729524A (en) | Step length estimation method, pedometer and step counting system | |
| CN106681619B (en) | Method and device for controlling energy consumption of electronic equipment | |
| Hongman et al. | Acceleration and orientation multisensor pedometer application design and implementation on the android platform | |
| CN103810401A (en) | Two-dimensional runoff restoration method for separating influences of human activities | |
| JP2017192563A (en) | Activity meter and exercise amount calculation device | |
| CN105574183A (en) | App (application) recommendation method based on collaborative filtering recommendation algorithm-KNN (K-nearest neighbor) classification algorithm | |
| CN106500717B (en) | Method and device for realizing step counting |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20141008 |