WO2016165333A1 - 一种实现计步的方法及装置 - Google Patents
一种实现计步的方法及装置 Download PDFInfo
- Publication number
- WO2016165333A1 WO2016165333A1 PCT/CN2015/095416 CN2015095416W WO2016165333A1 WO 2016165333 A1 WO2016165333 A1 WO 2016165333A1 CN 2015095416 W CN2015095416 W CN 2015095416W WO 2016165333 A1 WO2016165333 A1 WO 2016165333A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- frequency
- value
- effective
- corrected
- actual walking
- Prior art date
Links
Images
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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C25/00—Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
- A61B5/112—Gait analysis
Definitions
- This document relates to, but is not limited to, the field of mobile applications, and in particular to a method and apparatus for implementing a step.
- Recording the number of daily walking steps by the pedometer can monitor the user's movement and maintain the health of the user.
- the built-in three-axis accelerometer of the mobile phone it is not necessary to carry an extra step counting device.
- the counting data can be counted and managed to realize the effective motion data of the user. analysis. Therefore, compared with other step counting devices, it is more convenient to use a mobile phone for step counting, and statistical analysis of the step data is more effective.
- the application No. CN1909699A discloses a method and a device for implementing a step function of a mobile terminal, wherein the method comprises: measuring a motion acceleration curve of a moving object; Converted to a digital signal and extracts level information; the step information is derived from the level information.
- the method uses a fixed calculation method for step detection. It does not consider the factors such as the stride and frequency of each user during walking, and other factors affecting the stride and frequency, for example, different users due to walking gait and habits.
- the application of the application number CN102954803 discloses an adaptive step detection processing system and method based on feature point detection method; the method is adaptive to the user's walking state change by dividing the step counting process into a search rule stage and a determining rule stage.
- the step counting function includes: the step counting process initially enters the search rule stage, and whenever the walking signal meets the determination rule of the search rule stage, the search rule counter is incremented by one; when the search rule counter is accumulated to the upper limit value, The upper limit value is counted in the total counter, and automatically enters the determination rule phase; after entering the determination rule phase, each time the pedestrian signal is detected to satisfy the determination rule of the determination rule phase, the determination rule counter and the total counter are incremented by one.
- the determination parameters determining the determination rules of each stage can be changed according to the user's walking state. Adapt to the adjustment. When it is detected that the walking signal meets the requirements of the re-search rule, it re-enters the search rule phase.
- the step counting process includes the calculation of synthetic triaxial acceleration, analog-to-digital conversion and filtering, feature point detection and search rule judgment, counter accumulation, and determination of rule judgment.
- the calculation amount is large and the implementation is complicated, which is inconvenient in computing power and endurance. It is implemented on a weaker terminal device, so it is inconvenient to run on a portable terminal.
- the step-by-step detection using a fixed calculation method is easy to cause different degrees of error in the step data due to different walking steps and frequencies of the user during the detection process; in addition, the step of the three-axis acceleration sensor of different mobile phones Different frequencies will also cause errors in the step data.
- the complex step counting algorithm can improve the accuracy of the step counting data, the step counting process is computationally intensive and complicated in calculation, and it is inconvenient to run on a portable diagnostic device.
- Embodiments of the present invention provide a method and apparatus for implementing step counting, which can improve step counting accuracy by a simple algorithm.
- An embodiment of the present invention provides a method for implementing a step, including:
- the step is performed based on the corrected step count frequency of the triaxial acceleration sensor obtained by the correction and/or the corrected step rms value of the step.
- the method further comprises: correcting the step effective frequency to obtain a corrected step effective frequency, and processing the invalid step in the step data according to the corrected step effective frequency.
- the step counting frequency of the corrected three-axis acceleration sensor includes:
- the step frequency of the three-axis acceleration sensor is decreased in the preset frequency range according to the preset step length from the first frequency, and the preset frequency is a step counting frequency when the step value and the actual walking step error are the smallest in the range is determined as the corrected step frequency;
- the step frequency of the three-axis acceleration sensor is increased in the preset frequency range according to the preset step from the first frequency, and the preset frequency is The step counting frequency when the step value is less than the actual walking step number is determined as the corrected step frequency.
- step effective value of the corrected step is:
- the stride value of the step is increased in the effective stride interval according to the preset amplitude adjustment unit from the first step, and the effective stride interval is The step value of the step when the step value and the actual walking step have the smallest error is determined as the corrected step effective value;
- the step size is adjusted according to the preset amplitude adjustment unit from the first step, and the step value is reduced in the effective step interval, and the effective step is The step value of the step when the step value and the actual walking step error are the smallest in the interval is determined as the corrected step effective value.
- correcting the step effective frequency includes:
- the effective frequency value is reduced in the step effective frequency range according to the preset frequency adjustment unit from the first effective frequency, and the step effective frequency range is The effective frequency value when the step value is less than the actual walking step error is determined as the corrected step effective frequency;
- the effective frequency value is increased in the step effective frequency range according to the preset frequency adjustment unit from the first effective frequency, and the step effective frequency range is The effective frequency value when the step value and the actual walking step error are the smallest is determined as the corrected step effective frequency.
- correcting the step frequency of the three-axis acceleration sensor and the step effective value of the step include:
- the step counting frequency of the corrected triaxial acceleration sensor, the stride effective value of the step step, and the step effective frequency include:
- the step is further based on the step data of the corrected step frequency.
- the amplitude effective value is subjected to parameter correction; and the step effective frequency is corrected based on the step data of the corrected step effective value.
- the method further includes: when the correction of the step counting frequency does not reduce the error between the step value and the actual walking step, taking the first frequency as the corrected step frequency;
- the first step is taken as the corrected step effective value
- the first effective frequency is taken as the corrected step effective frequency.
- the corrected step frequency includes:
- the acceleration data information, the first frequency and the step value of the three-axis acceleration sensor of the actual walking step in the step counting data are input, and are performed in a preset frequency range according to the preset step number. Calculating a step counting frequency when the step value of the step frequency and the actual walking step are the smallest in the preset frequency range, and determining the frequency as the corrected step frequency;
- the acceleration data information, the first step and the step value of the three-axis acceleration sensor of the actual walking steps in the step counting data are input, and the effective step is adjusted according to the preset amplitude.
- the effective step is adjusted according to the preset amplitude.
- the acceleration data information, the first step effective frequency and the step value of the three-axis acceleration sensor of the actual walking steps in the step counting data are input, and the unit is adjusted according to the preset frequency.
- the present application further provides an apparatus for implementing a step counting, comprising: a first correcting unit and a step counting unit; wherein
- a first correcting unit configured to correct a step frequency of the triaxial acceleration sensor and/or a stride effective value of the step according to the step data obtained by the preset step number
- the step counting unit is configured to perform the step counting according to the corrected step frequency of the corrected three-axis acceleration sensor and/or the corrected step effective value of the step.
- the apparatus further includes a second correction unit configured to perform correction on the step effective frequency to obtain a corrected step effective frequency, and the counter step data of the step counter unit is invalid according to the corrected step effective frequency Step by step.
- the first correcting unit is configured to implement the step counting frequency of the corrected three-axis acceleration sensor by: step counting data according to the obtained preset step number,
- the step frequency of the three-axis acceleration sensor is decreased in the preset frequency range according to the preset step length from the first frequency, and the preset frequency is a step counting frequency when the step value and the actual walking step error are the smallest in the range is determined as the corrected step frequency;
- the step frequency of the three-axis acceleration sensor is increased in the preset frequency range according to the preset step from the first frequency, and the preset frequency is a step counting frequency when the step value and the actual walking step error are the smallest in the range is determined as the corrected step frequency;
- the stride value of the step is increased in the effective stride interval according to the preset amplitude adjustment unit from the first step, and the effective stride interval is The step value of the step when the step value and the actual walking step have the smallest error is determined as the corrected step effective value;
- the step size is adjusted according to the preset amplitude adjustment unit from the first step, and the step value is reduced in the effective step interval, and the effective step is The step value of the step when the step value and the actual walking step error are the smallest in the interval is determined as the corrected step effective value.
- the second correcting unit is configured to perform correction on the step effective frequency by: step counting data according to the obtained preset step number,
- the effective frequency value is decreased according to the preset frequency adjustment unit according to the preset frequency adjustment range, and the step is stepped.
- the effective frequency value when the step value and the actual walking step error are the smallest in the effective frequency range is determined as the corrected step effective frequency;
- the effective frequency value is increased in the step effective frequency range according to the preset frequency adjustment unit from the first effective frequency, and the step effective frequency range is The effective frequency value when the step value and the actual walking step error are the smallest is determined as the corrected step effective frequency.
- the first correcting unit is configured to implement the stepping frequency of the corrected triaxial acceleration sensor and the stride effective value of the step by: correcting the triaxial accelerometer according to the stepped data of the obtained preset step number After the step counting frequency obtains the corrected step frequency, the step effective value is corrected according to the step data of the corrected step frequency.
- the second correcting unit is configured to realize the step counting frequency of the corrected three-axis acceleration sensor, the step effective value of the step, and the step effective frequency by: completing the three-axis acceleration in the first correcting unit in order After the correction of the step frequency of the sensor and the correction of the step effective value of the step, the step effective frequency is corrected based on the step data of the corrected step effective value.
- the apparatus further includes determining the evaluation unit, configured to take the first frequency as the corrected step frequency when the correction of the step frequency does not reduce the error between the step value and the actual walking step;
- the first step is taken as the corrected step effective value
- the first effective frequency is taken as the corrected step effective frequency.
- the first correcting unit is configured to implement the corrected step counting frequency by using the machine learning algorithm and the migration learning algorithm, the acceleration data information of the triaxial acceleration sensor of the actual walking step number in the step counting data, first The frequency and the step value are input, and the calculation is performed in the preset frequency range according to the preset step size to obtain the step counting frequency when the step value and the actual walking step have the smallest error in the preset frequency range, and the frequency is determined to be Correcting the step counting frequency;
- the second correcting unit is configured to perform correction of the step effective frequency by: based on the machine learning algorithm and the migration learning algorithm, the acceleration data information of the triaxial acceleration sensor of the actual walking step in the step counting data, A step effective frequency and a step value are input, and the effective frequency value when the step value and the actual walking step have the smallest error in the step effective frequency range are determined according to the preset frequency, and the effective frequency value is determined.
- the correction step counts the effective frequency.
- the embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the above method.
- the technical solution of the present application includes: correcting the step frequency of the triaxial acceleration sensor and/or the stride effective value of the step according to the step data obtained by the preset step; according to the corrected triaxial acceleration sensor
- the corrected step count frequency and/or the stepped corrected step rms value are counted.
- the method of the embodiment of the invention corrects the step frequency of the triaxial acceleration sensor and/or the step value of the step step, thereby avoiding the step error caused by the user walking stride and the step frequency of the triaxial acceleration sensor being different.
- the accuracy of the step data is improved.
- the correction of the effective frequency of the step is used to process the invalid step in the step data, which further improves the accuracy of the step data.
- FIG. 1 is a flowchart of a method for implementing step counting according to an embodiment of the present invention
- FIG. 2 is a structural diagram of an apparatus for implementing a step counting according to an embodiment of the present invention
- Figure 3 is a schematic diagram showing a complete display of partially unread messages
- FIG. 4 is a structural diagram of an apparatus for displaying an unread message according to an embodiment of the present invention.
- FIG. 1 is a flowchart of a method for implementing step counting according to an embodiment of the present invention. As shown in FIG. 1 , the method includes:
- Step 100 Correct the step frequency of the triaxial acceleration sensor and/or the step value of the step according to the step data obtained by the preset step number; the preset step number generally refers to about 500 to 1000 steps.
- the step counting frequency of the corrected three-axis acceleration sensor includes:
- the step frequency of the triaxial acceleration sensor is decreased in the preset frequency range according to the preset step length from the first frequency, and the step frequency is preset within the preset frequency range.
- the step frequency at which the error between the value and the actual walking step is the smallest is determined as the corrected step frequency;
- the step frequency of the three-axis acceleration sensor is increased in the preset frequency range according to the preset step length from the first frequency, and the step value in the preset frequency range is The step frequency at which the actual walking step error is the smallest is determined as the corrected step frequency.
- the preset frequency range is a general working range determined by a person skilled in the art according to different triaxial acceleration sensors, and the general preset frequency range may be 8.0 Hz to 12.0 Hz; the first frequency may be in the field.
- the empirical value of the technician for example, 10.0 Hz; the preset step size may be 0.5 Hz.
- the RMS values for the corrected step are:
- the step value of the step is increased in the effective step range according to the preset amplitude adjustment unit from the first step, and the effective step interval is calculated.
- the step value of the step value when the step value and the actual walking step error are the smallest is determined as the corrected step effective value;
- the step value of the step is reduced in the effective step range according to the preset amplitude adjustment unit from the first step, and the effective step range is The step value of the step when the step value and the actual walking step have the smallest error is determined as the corrected step effective value.
- the effective stride interval is a numerical range determined by a person skilled in the art according to the stride of walking for a general user, and the general effective stride interval may be 5.0 cm to 10.0 cm; the first step may be The empirical value of the skilled person, for example, 7.5 cm; the preset amplitude adjustment unit may be 0.5 cm.
- Step 101 Perform step counting according to the corrected step frequency of the triaxial acceleration sensor obtained by the correction and/or the corrected step effective value of the step.
- the method of the embodiment of the present invention further includes: correcting the step effective frequency to obtain a corrected step effective frequency, to process the invalid step in the step data.
- correcting the step effective frequency includes:
- the effective frequency value is decreased in the step effective frequency range according to the preset frequency adjustment unit from the first effective frequency, and the stepping effective frequency range is counted.
- the effective frequency value when the error between the value and the actual walking step is minimum is determined as the corrected step effective frequency
- the effective frequency value is increased in the step effective frequency range according to the preset frequency adjustment unit from the first effective frequency, and the step frequency is calculated within the step effective frequency range.
- the effective frequency value when the value and the actual walking step error are the smallest is determined as the corrected step effective frequency.
- the step effective frequency range is a numerical range determined by a person skilled in the art according to the frequency of walking of a general user, and the general user walking speed does not exceed 3 steps per second.
- the user takes one step at a time of 200. From milliseconds to 1000 milliseconds, the general walking time is 300 milliseconds. When performing calibration, it can be adjusted with 50 milliseconds as the duration, and the corresponding frequency value can be calculated according to the above duration.
- the calibration process includes correcting the step frequency of the triaxial acceleration sensor and the step effective value of the step, the calibration process includes:
- the step effective value is corrected according to the step data of the corrected step frequency
- the calibration process includes correcting the step frequency of the triaxial acceleration sensor, the stride effective value of the step, and the step effective frequency, the calibration process includes:
- the step effective value is corrected according to the step data of the corrected step frequency; and the step effective frequency is corrected based on the step data of the corrected step effective value.
- the method of the embodiment of the present invention further includes: when the correction of the step frequency does not reduce the error between the step value and the actual walking step, taking the first frequency as the corrected step frequency;
- the first step is taken as the corrected step effective value
- the first effective frequency is taken as the corrected step effective frequency.
- the acceleration data information, the first frequency and the step value of the three-axis acceleration sensor of the actual walking steps in the step counting data are input, and are performed in a preset frequency range according to a preset step size. Calculating the step counting frequency when the step value between the step value and the actual walking step is the smallest in the preset frequency range, and determining the frequency as the corrected step frequency;
- Corrected step RMS values include:
- the acceleration data information, the first step and the step value of the three-axis acceleration sensor of the actual walking steps in the step counting data are input, and the effective step is adjusted according to the preset amplitude. Calculating in the interval to obtain the stride value of the step when the step value in the effective stride interval is the smallest and the actual walking step is the smallest, and determining the stride value of the step is the corrected stride effective value;
- the acceleration data information, the first step effective frequency and the step value of the three-axis acceleration sensor of the actual walking steps in the step counting data are input, and the unit is adjusted according to the preset frequency.
- the machine learning algorithm and the migration learning algorithm may be a decision tree algorithm, a neural network algorithm, a speed learning machine algorithm or a deep learning algorithm.
- the method of the embodiment of the invention corrects the step frequency of the triaxial acceleration sensor and/or the step value of the step step, thereby avoiding the step error caused by the user walking stride and the step frequency of the triaxial acceleration sensor being different.
- the accuracy of the step data is improved.
- the correction of the effective frequency of the step is used to process the invalid step in the step data, which further improves the accuracy of the step data.
- the embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the above method.
- FIG. 2 is a structural diagram of a device for implementing a step counting according to an embodiment of the present invention. As shown in FIG. 2, the method includes: a first correcting unit and a step counting unit;
- a first correcting unit configured to correct a step frequency of the triaxial acceleration sensor and/or a stride effective value of the step according to the step data obtained by the preset step number
- the first correcting unit is configured to obtain step counting data of the preset number of steps,
- the step frequency of the triaxial acceleration sensor is decreased in the preset frequency range according to the preset step length from the first frequency, and the step frequency is preset within the preset frequency range.
- the step frequency at which the error between the value and the actual walking step is the smallest is determined as the corrected step frequency;
- the step frequency of the three-axis acceleration sensor is increased in the preset frequency range according to the preset step length from the first frequency, and the step frequency is preset within the preset frequency range.
- the step counting frequency when the value is the smallest with the actual walking step error is determined as the corrected step frequency; and/or,
- the step value of the step is increased in the effective step range according to the preset amplitude adjustment unit from the first step, and the effective step interval is calculated.
- the step value of the step value when the step value and the actual walking step error are the smallest is determined as the corrected step effective value;
- the step value of the step is reduced in the effective step range according to the preset amplitude adjustment unit from the first step, and the effective step range is The step value of the step when the step value and the actual walking step have the smallest error is determined as the corrected step effective value.
- the step counting unit is configured to perform the step counting according to the corrected step frequency of the corrected three-axis acceleration sensor and/or the corrected step effective value of the step.
- the apparatus of the embodiment of the present invention further includes a second correcting unit configured to perform correction on the step effective frequency to obtain a corrected step effective frequency to process the invalid step in the step data of the step counter unit.
- the second correcting unit is configured to acquire the step data of the preset step number
- the effective frequency value is decreased in the step effective frequency range according to the preset frequency adjustment unit from the first effective frequency, and the stepping effective frequency range is counted.
- the effective frequency value when the error between the value and the actual walking step is minimum is determined as the corrected step effective frequency
- the effective frequency value is increased in the step effective frequency range according to the preset frequency adjustment unit from the first effective frequency, and the step frequency is calculated within the step effective frequency range.
- the effective frequency value when the value and the actual walking step error are the smallest is determined as the corrected step effective frequency.
- the first correcting unit is configured to correct the stepping frequency of the triaxial acceleration sensor according to the step counting data of the obtained preset step number, and obtain the corrected step counting frequency, and the step value of the step counting data according to the corrected step counting frequency Make corrections.
- the second correcting unit is configured to perform, after the first correcting unit completes the correction of the step frequency of the three-axis acceleration sensor and the step effective value of the step, in accordance with the step data of the corrected step effective value. Correction of the step effective frequency.
- the apparatus of the embodiment of the present invention further includes determining the evaluation unit, and setting
- the first frequency is taken as the corrected step frequency
- the first step is taken as the corrected step effective value
- the first effective frequency is taken as the corrected step effective frequency.
- the first correcting unit is configured to, based on the machine learning algorithm and the migration learning algorithm, the acceleration data information, the first frequency and the step value of the three-axis acceleration sensor of the actual walking step number in the step counting data as input, according to the preset step Calculate to obtain the step frequency when the step value between the step value and the actual walking step is the smallest in the preset frequency range, and determine the frequency as the corrected step frequency; and/or,
- the acceleration data information, the first step and the step value of the three-axis acceleration sensor of the actual walking steps in the step counting data are input, and the effective step is adjusted according to the preset amplitude. Calculating in the interval to obtain the stride value of the step when the step value in the effective stride interval is the smallest and the actual walking step is the smallest, and determining the stride value of the step is the corrected stride effective value;
- the second correcting unit is configured to, based on the machine learning algorithm and the migration learning algorithm, the acceleration data information of the triaxial acceleration sensor of the actual walking step number in the step counting data, and the first step effective frequency
- the rate and the step value are input, and the effective frequency value when the step value and the actual walking step have the smallest error in the step effective frequency range are determined according to the preset frequency, and the effective frequency value is determined as the corrected step effective frequency.
- step counting device in the embodiment of the present invention may perform step counting separately or may be embedded in a mobile terminal, such as a mobile phone, and the specific embedding method does not require creative work by those skilled in the art.
- the first frequency of the three-axis acceleration sensor is 10.0 Hz
- the preset frequency range may be 8.0 Hz to 12.0 Hz
- the preset step size is 0.5 Hz
- the effective step interval may be from 5.0 cm to 10.0.
- the centimeter, the first step is 7.5 cm
- the preset amplitude adjustment unit can be 0.5 cm
- the effective frequency range is determined according to the time of the user's step by step from 200 milliseconds to 1000 milliseconds, and the first effective frequency is 300 milliseconds for one step of walking.
- the frequency adjustment unit takes 50 milliseconds as the adjustment calculation unit.
- FIG. 3 is a flowchart of a method for performing step counting according to an embodiment of the present invention. As shown in FIG. 3, the method includes:
- Step 300 Obtain step counting data of the preset step number and the actual walking step number; wherein, obtaining the actual walking step number includes: real-time collecting three-axis acceleration data of x, y, and z, step counting value, and the like;
- Step 301 Determine whether to perform correction processing according to the step value in the step data and the actual walking step; optionally, when the step value is greater than or less than the actual walking step, the correction may be determined. It should be noted that, in practical applications, it is possible to set whether or not to correct the error ratio. For example, when the step value and the actual walking step error exceed 2%, correction is needed, and the specific percentage can be adjusted according to actual conditions.
- the correction process includes a step frequency of the triaxial acceleration sensor, and/or a step effective value of the step, and/or a correction of the step effective frequency.
- step 3020 there is an error between the design step value and the actual walking step number, and step 3020 needs to be performed for correction; if there is no error, or the error ratio is smaller than the error ratio set by the technician, no correction is performed, and the relevant parameters are directly used to perform step 3030. Step by step processing.
- Step 3020 correcting the step frequency of the triaxial acceleration sensor; optionally,
- the step frequency of the triaxial acceleration sensor is reduced in the range of 10 Hz to 8 Hz in steps of 0.5 Hz from 10.0 Hz, and the step frequency is preset within the range of the preset frequency.
- the step frequency at which the error between the value and the actual walking step is the smallest is determined as the corrected step frequency; here, the preset frequency range is theoretically 8 Hz to 12 Hz, since the first frequency is 10 Hz and the processing for lowering the step frequency is performed, Therefore, the adjustment range is theoretically reduced to 10 Hz to 8 Hz, and the relevant corrections in the following statements are similar and will not be described.
- the step counting frequency of the triaxial accelerometer is increased from 10 Hz to 12 Hz in the range of 10 Hz to 12 Hz from 10.0 Hz, and the step value in the preset frequency range is The step frequency at which the actual walking step error is the smallest is determined as the corrected step frequency.
- the first frequency is taken as the corrected step frequency
- Step 3021 correct the step effective value of the step; optionally,
- the step is adjusted from 7.5 cm according to the preset amplitude to 0.5 cm.
- the step value is increased from 7.5 cm to 10.0 cm, and the effective step interval is The step value of the step is determined as the corrected step effective value when the internal step value and the actual walking step error are the smallest;
- the unit is adjusted from 7.5 cm according to the preset amplitude to 0.5 cm, and the step value is reduced within 5 cm to 7.5 cm, and the effective step is The step value of the step is determined as the corrected step effective value when the error between the step value and the actual walking step is the smallest.
- the first step is taken as the corrected step effective value.
- FIG. 4 is a three-axis acceleration graph sampled during the step counting process of the embodiment, as shown in FIG. 4, wherein the horizontal axis is the sampling point, the vertical axis represents the acceleration value, and the solid line is the x-axis acceleration data, which is composed of points.
- the dotted line is the z-axis acceleration data, and the dotted line composed of small short lines is the y-axis acceleration data; the fluctuation law of the x-axis acceleration data in the figure is obvious, and can be determined as the acceleration in the gravity direction;
- the point is the point from the upward direction to the downward direction, that is, the upper extreme point; the point B is the point from the downward direction to the upward direction, that is, the lower extreme point.
- the subsequent C points, E points, G points, I points, and K points are upper extreme points; D points, F points, H points, J points, and L points are lower extreme points.
- the amplitude value between the upper and lower extreme points is used as the stride, and when the size of the stride RMS is satisfied, the step is performed.
- the first step is 7.5 cm.
- the amplitude values of points A and B are 12 cm and greater than 7.5 cm; the amplitudes of points B and C are 4.5 cm and less than 7.5 cm.
- the amplitude values of points C and D are 3 cm and less than 7.5 cm.
- the calculation can obtain the step value of 3 when the step is 7.5 cm according to the first step; if the step value is less than the actual walking step, then 0.5 cm is used as the amplitude adjustment unit reduction.
- Step value of the step that is, step by step in the order of 7 cm, 6.5 cm, 6 cm, assuming that the step value is reduced to 6 cm, and the step value is 5, assuming that the step data is accurate at this time, 6 cm was determined as the corrected step effective value.
- Step 3022 correct the stepped effective frequency; optionally,
- the effective frequency value is decreased in the step effective frequency range according to the preset frequency adjustment unit from the first effective frequency, and the stepping effective frequency range is counted.
- the effective frequency value when the error between the value and the actual walking step is minimum is determined as the corrected step effective frequency
- the effective frequency value is increased in the step effective frequency range according to the preset frequency adjustment unit from the first effective frequency, and the step frequency is calculated within the step effective frequency range.
- the effective frequency value when the value and the actual walking step error are the smallest is determined as the corrected step effective frequency.
- the first effective frequency is taken as the corrected step effective frequency.
- Step 3023 Perform step counting according to the corrected step counting frequency obtained by the correction, the corrected step effective value, and the corrected step effective frequency.
- step 3030 the step is directly performed.
- the step counting frequency of the triaxial acceleration sensor prevents the step counting frequency from being too large when the step counting frequency is too large, the step counting frequency is too small, and the step counting value is small; the step value of the step step is corrected to avoid different values.
- the user has different guides due to stride, walking posture, walking habits and placement of the step counter device. The stride data is disturbed; the step counter effective frequency is used to invalidate the step that is not caused by walking.
- the user places the mobile phone for step counting in the bag or holds it in the hand. Placed in the bag or due to the sliding of the hand, the triaxial acceleration sensor firstly affects the data outside the effective frequency range of the step, and processing the part of the data can improve the step counting accuracy. After the test of the method of the embodiment, it is determined that the step error is reduced from about 8% to about 2%, so that the step accuracy is improved.
- the calibration process may be implemented by a machine learning algorithm and a migration learning algorithm, and specifically may be a decision tree algorithm, a neural network algorithm, a speed learning algorithm, or a deep learning algorithm.
- each module/unit in the above embodiment may be implemented in the form of hardware, for example, by implementing an integrated circuit to implement its corresponding function, or may be implemented in the form of a software function module, for example, executing a program stored in the memory by a processor. / instruction to achieve its corresponding function.
- the invention is not limited to any specific form of combination of hardware and software.
- the above technical solution avoids the step error caused by the user walking stride and the three-axis accelerometer step counting frequency, and improves the accuracy of the step counting data; in addition, the step counting data is realized by the step counting effective frequency correction. The invalid step is processed to further improve the accuracy of the step data.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Manufacturing & Machinery (AREA)
- Measurement Of Distances Traversed On The Ground (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
一种实现计步的方法及装置，包括：根据获取的预设步数的计步数据校正三轴加速度传感器的计步频率和/或计步的步幅有效值（100）；根据校正后三轴加速度传感器的校正计步频率和/或计步的校正步幅有效值进行计步（101）。该技术方案通过对三轴加速度传感器的计步频率和/或计步的步幅有效值进行校正，避免了由于用户行走步幅、三轴加速度传感器计步频率不同造成的计步误差，提高了计步数据的准确性；另外，通过计步有效频率的校正，实现了对计步数据中无效计步进行处理，进一步提高了计步数据的准确性。
Description
本文涉及但不限于移动应用领域，尤指一种实现计步的方法及装置。
通过计步装置记录每日行走步数可以监测用户运动情况、对用户保持身体健康等有重要参考意义。通过使用手机内置三轴加速度传感器进行计步不需要携带额外的计步装置，同时，利用手机将计步数据上传至服务器后，可以对计步数据进行统计和管理，实现对用户运动数据的有效分析。因而相对于其他计步装置，采用手机进行计步更为便捷、统计分析计步数据更为有效。
目前，计步方法主要有以下两种：申请号为CN1909699A的申请公开了一种实现移动终端计步功能的方法及装置，其中，所述方法包括：测量移动物体运动加速度变化曲线；将模拟信号转变为数字信号并提取电平信息；根据电平信息得出计步信息。该方法采用固定不变的计算方式进行计步检测，未考虑各用户在行走过程中存在步幅、频率，以及对步幅和频率造成影响的其他因素，例如、不同用户由于行走步态和习惯造成检测到的步幅存在差异；将计步装置放置在包袋或配置在手臂、腰部等不同位置，造成检测的频率存在差异；另外，不同手机的三轴加速度传感器的计步频率不同，过高或过低的频率，都将对计步数据造成影响，使计步数据出现不同程度的误差。申请号为CN102954803的申请公开一种基于特征点检测法的自适应计步处理系统及方法；该方法通过将计步过程划分为搜索规则阶段和确定规则阶段实现随使用者步行状态变化而自适应计步的功能，包括：计步处理初始默认进入搜索规则阶段，每当监测到步行信号满足搜索规则阶段的判定规则时，就将搜索规则计数器加一；当搜索规则计数器累加至上限值时，将该上限值计入总计数器中，且自动进入确定规则阶段；进入到确定规则阶段后，每当检测到步行信号满足确定规则阶段的判定规则时，就将确定规则计数器和总计数器加一；其中，确定各阶段判定规则的判定参数能随使用者步行状态变化做自
适应调整。当监测到步行信号满足重新搜索规则的要求时，便重新进入到搜索规则阶段。该计步过程包含合成三轴加速度的计算、模数转换和滤波、特征点检测与搜索规则判断、计数器累加、确定规则判断等过程，计算量大且实现复杂，不便于在计算能力和续航能力较弱的终端设备上实现，因此不便在便携式终端上运行使用。
综上，采用固定不变的计算方式进行计步检测，容易因为检测过程中用户行走步幅、频率等不同造成计步数据出现不同程度的误差；另外，不同手机的三轴加速度传感器的计步频率不同，也会造成计步数据出现误差。采用复杂的计步算法虽然可以提高计步数据的准确性，但计步过程计算量大且计算复杂，不便在便携的诊断设备上运行使用。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本发明实施例提供一种实现计步的方法及装置，能够通过简单算法提高计步准确性。
本发明实施例提供了一种实现计步的方法，包括：
根据获取的预设步数的计步数据校正三轴加速度传感器的计步频率和/或计步的步幅有效值；
根据校正获得的三轴加速度传感器的校正计步频率和/或计步的校正步幅有效值进行计步。
可选地，该方法还包括：对计步有效频率进行校正获得校正计步有效频率，根据所述校正计步有效频率以对所述计步数据中无效计步进行处理。
可选地，校正三轴加速度传感器的计步频率包括：
当所述计步数据中的计步数值大于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内降低所述三轴加速度传感器的计步频率，将预设频率范围内所述计步数值与所述实际行走步数误差最小时的计步频率确定为所述校正计步频率；
当所述计步数据中的计步数值小于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内提高所述三轴加速度传感器的计步频率，将预设频率范围内所述计步数值与所述实际行走步数误差最小时的计步频率确定为所述校正计步频率。
可选地，校正计步的步幅有效值包括：
当所述计步数据中的计步数值大于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内加大计步的步幅值，将有效步幅区间内所述计步数值与所述实际行走步数误差最小时计步的步幅值确定为所述校正步幅有效值；
当所述计步数据中的计步数值小于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内减小计步的步幅值，将有效步幅区间内所述计步数值与所述实际行走步数误差最小时计步的步幅值确定为所述校正步幅有效值。
可选地，对计步有效频率进行校正包括：
当所述计步数据中的计步数值大于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内降低有效频率值，将计步有效频率范围内所述计步数值与所述实际行走步数误差最小时的有效频率值确定为所述校正计步有效频率；
当所述计步数据中的计步数值小于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内提高有效频率值，将计步有效频率范围内所述计步数值与所述实际行走步数误差最小时的有效频率值确定为所述校正计步有效频率。
可选地，校正三轴加速度传感器的计步频率和计步的步幅有效值包括：
在获得所述校正计步频率后，根据校正计步频率的计步数据对所述步幅有效值进行校正；
所述校正三轴加速度传感器的计步频率、计步的步幅有效值和计步有效频率包括：
在获得所述校正计步频率后，再根据校正计步频率的计步数据对所述步
幅有效值进行参数校正；基于所述校正步幅有效值的计步数据进行所述计步有效频率的校正。
可选地，该方法还包括：当所述计步频率的校正未减少计步数值与实际行走步数的误差时，取所述第一频率作为所述校正计步频率；
当所述步幅有效值的校正未减少计步数值与实际行走步数的误差时，取所述第一步幅作为所述校正步幅有效值；
当所述计步有效频率的校正未减少计步数值与实际行走步数的误差时，取所述第一有效频率作为所述校正计步有效频率。
可选地，校正计步频率包括：
基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一频率及计步数值为输入，按照预设步数在预设频率范围内进行计算获得预设频率范围内所述计步数值与实际行走步数误差最小时的计步频率，确定该频率为所述校正计步频率；
所述校正步幅有效值包括：
基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一步幅及计步数值为输入，按照预设幅值调整单位在有效步幅区间内进行计算获得有效步幅区间内所述计步数值与实际行走步数误差最小时计步的步幅值，确定该计步的步幅值为所述校正步幅有效值；
所述对计步有效频率进行校正包括：
基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一计步有效频率及计步数值为输入，按照预设的频率调整单位在计步有效频率范围内所述计步数值与实际行走步数误差最小时的有效频率值，确定该有效频率值为所述校正计步有效频率。
另一方面，本申请还提供一种实现计步的装置，包括：第一校正单元和计步单元；其中，
第一校正单元，设置为根据获取的预设步数的计步数据校正三轴加速度传感器的计步频率和/或计步的步幅有效值；
计步单元，设置为根据校正后三轴加速度传感器的校正计步频率和/或计步的校正步幅有效值进行计步。
可选地，该装置还包括第二校正单元，设置为对计步有效频率进行校正获得校正计步有效频率，根据所述校正计步有效频率对所述计步单元的计步数据中无效计步进行处理。
可选地，第一校正单元是设置为通过如下方式实现校正三轴加速度传感器的计步频率：，根据获取的预设步数的计步数据，
当所述计步数据中的计步数值大于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内降低所述三轴加速度传感器的计步频率，将预设频率范围内所述计步数值与所述实际行走步数误差最小时的计步频率确定为所述校正计步频率；
当所述计步数据中的计步数值小于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内提高所述三轴加速度传感器的计步频率，将预设频率范围内所述计步数值与所述实际行走步数误差最小时的计步频率确定为所述校正计步频率；
所述第一校正单元是设置为通过如下方式实现校正计步的步幅有效值：
当所述计步数据中的计步数值大于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内加大计步的步幅值，将有效步幅区间内所述计步数值与所述实际行走步数误差最小时计步的步幅值确定为所述校正步幅有效值；
当所述计步数据中的计步数值小于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内减小计步的步幅值，将有效步幅区间内所述计步数值与所述实际行走步数误差最小时计步的步幅值确定为所述校正步幅有效值。
可选地，第二校正单元是设置为通过如下方式实现对计步有效频率进行校正：根据获取的预设步数的计步数据，
当所述计步数据中的计步数值大于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内降低有效频率值，将计步
有效频率范围内所述计步数值与所述实际行走步数误差最小时的有效频率值确定为所述校正计步有效频率；
当所述计步数据中的计步数值小于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内提高有效频率值，将计步有效频率范围内所述计步数值与所述实际行走步数误差最小时的有效频率值确定为所述校正计步有效频率。
可选地，第一校正单元是设置为通过如下方式实现校正三轴加速度传感器的计步频率和计步的步幅有效值：根据获取的预设步数的计步数据，校正三轴加速度传感器的计步频率获得所述校正计步频率后，根据校正计步频率的计步数据对所述步幅有效值进行校正。
所述第二校正单元是设置为通过如下方式实现校正三轴加速度传感器的计步频率、计步的步幅有效值和计步有效频率：在第一校正单元按照先后顺序完成所述三轴加速度传感器的计步频率的校正和所述计步的步幅有效值的校正后，基于所述校正步幅有效值的计步数据进行所述计步有效频率的校正。
可选地，该装置还包括判断赋值单元，设置为当所述计步频率的校正未减少计步数值与实际行走步数的误差时，取所述第一频率作为所述校正计步频率；
当所述步幅有效值的校正未减少计步数值与实际行走步数的误差时，取所述第一步幅作为所述校正步幅有效值；
当所述计步有效频率的校正未计步数值与实际行走步数的误差时，取所述第一有效频率作为所述校正计步有效频率。
可选地，第一校正单元是设置为通过如下方式实现校正计步频率：基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一频率及计步数值为输入，按照预设步长在预设频率范围内进行计算获得预设频率范围内所述计步数值与实际行走步数误差最小时的计步频率，确定该频率为所述校正计步频率；
所述第一校正单元是设置为通过如下方式实现校正步幅有效值：
基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴
加速度传感器的加速度数据信息、第一步幅及计步数值为输入，按照预设幅值调整单位在有效步幅区间内进行计算获得有效步幅区间内所述计步数值与实际行走步数误差最小时计步的步幅值，确定该计步的步幅值为所述校正步幅有效值；
所述第二校正单元是设置为通过如下方式实现对计步有效频率进行校正：基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一计步有效频率及计步数值为输入，按照预设的频率调整单位在计步有效频率范围内所述计步数值与实际行走步数误差最小时的有效频率值，确定该有效频率值为所述校正计步有效频率。
本发明实施例还提供了一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行上述的方法。
与相关技术相比，本申请技术方案包括：根据获取的预设步数的计步数据校正三轴加速度传感器的计步频率和/或计步的步幅有效值；根据校正后三轴加速度传感器的校正计步频率和/或计步的校正步幅有效值进行计步。本发明实施例方法通过对三轴加速度传感器的计步频率和/或计步的步幅有效值进行校正，避免了由于用户行走步幅、三轴加速度传感器计步频率不同造成的计步误差，提高了计步数据的准确性；另外，通过计步有效频率的校正，实现了对计步数据中无效计步进行处理，进一步提高了计步数据的准确性。
在阅读并理解了附图和详细描述后，可以明白其他方面。
附图概述
图1为本发明实施例实现计步的方法的流程图；
图2为本发明实施例实现计步的装置的结构程图；
图3是完整显示部分未读消息的示意图；
图4为本发明实施例未读消息显示的装置的结构程图。
下文中将结合附图对本发明的实施例进行详细说明。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互任意组合。
图1为本发明实施例实现计步的方法的流程图，如图1所示，包括：
步骤100、根据获取的预设步数的计步数据校正三轴加速度传感器的计步频率和/或计步的步幅有效值；这里预设步数一般指500～1000步左右。
本步骤中，校正三轴加速度传感器的计步频率包括：
当计步数据中的计步数值大于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内降低三轴加速度传感器的计步频率，将预设频率范围内计步数值与实际行走步数误差最小时的计步频率确定为校正计步频率；
当数据中的计步数值小于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内提高三轴加速度传感器的计步频率，将预设频率范围内计步数值与实际行走步数误差最小时的计步频率确定为校正计步频率。
需要说明的是，预设频率范围是本领域技术人员根据不同三轴加速度传感器确定的一般的工作范围，一般的预设频率范围可以取值为8.0Hz～12.0Hz；第一频率可以是本领域技术人员的经验值，例如、10.0Hz；预设步长可以是0.5Hz。
校正计步的步幅有效值包括：
当计步数据中的计步数值大于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内加大计步的步幅值，将有效步幅区间内计步数值与实际行走步数误差最小时计步的步幅值确定为校正步幅有效值；
当计步数据中的计步数值小于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内减小计步的步幅值，将有效步幅区间内计步数值与实际行走步数误差最小时计步的步幅值确定为校正步幅有效值。
需要说明的是，有效步幅区间是本领域技术人员根据对一般用户行走的步幅确定的数值范围，一般的有效步幅区间可以取值为5.0厘米～10.0厘米；第一步幅可以是本领域技术人员的经验值，例如、7.5厘米；预设幅值调整单位可以是0.5厘米。
步骤101、根据校正获得的三轴加速度传感器的校正计步频率和/或计步的校正步幅有效值进行计步。
可选的，本发明实施例方法还包括：对计步有效频率进行校正获得校正计步有效频率，以对计步数据中无效计步进行处理。
可选的，对计步有效频率进行校正包括：
当计步数据中的计步数值大于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内降低有效频率值，将计步有效频率范围内计步数值与实际行走步数误差最小时的有效频率值确定为校正计步有效频率；
当计步数据中的计步数值小于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内提高有效频率值，将计步有效频率范围内计步数值与实际行走步数误差最小时的有效频率值确定为校正计步有效频率。
需要说明的是，计步有效频率范围是本领域技术人员根据对一般用户行走的频率确定的数值范围，一般的用户行走速度每秒不超过3步，按照分析，用户每走一步的时间在200毫秒～1000毫秒，一般行走一步时间为300毫秒，进行校正时，可以以50毫秒作为时长进行调整，根据上述时长可以计算相应的频率值。
本发明实施例方法如果校正过程包括校正三轴加速度传感器的计步频率和计步的步幅有效值，校正过程包括：
在获得校正计步频率后，根据校正计步频率的计步数据对步幅有效值进行校正；
本发明实施例方法如果校正过程包括校正三轴加速度传感器的计步频率、计步的步幅有效值和计步有效频率，校正过程包括：
在获得校正计步频率后，再根据校正计步频率的计步数据对步幅有效值进行参数校正；基于校正步幅有效值的计步数据进行计步有效频率的校正。
本发明实施例方法还包括：当计步频率的校正未减少计步数值与实际行走步数的误差时，取第一频率作为校正计步频率；
当步幅有效值的校正未减少计步数值与实际行走步数的误差时，取第一步幅作为校正步幅有效值；
当计步有效频率的校正未减少计步数值与实际行走步数的误差时，取第一有效频率作为校正计步有效频率。
本发明实施例方法校正计步频率包括：
基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一频率及计步数值为输入，按照预设步长在预设频率范围内进行计算获得预设频率范围内计步数值与实际行走步数误差最小时的计步频率，确定该频率为校正计步频率；
校正步幅有效值包括：
基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一步幅及计步数值为输入，按照预设幅值调整单位在有效步幅区间内进行计算获得有效步幅区间内计步数值与实际行走步数误差最小时计步的步幅值，确定该计步的步幅值为校正步幅有效值；
对计步有效频率进行校正包括：
基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一计步有效频率及计步数值为输入，按照预设的频率调整单位在计步有效频率范围内计步数值与实际行走步数误差最小时的有效频率值，确定该有效频率值为校正计步有效频率。
需要说明的是，机器学习算法和迁移学习算法可以是决策树算法、神经网络算法、极速学习机算法或深度学习算法等。
本发明实施例方法通过对三轴加速度传感器的计步频率和/或计步的步幅有效值进行校正，避免了由于用户行走步幅、三轴加速度传感器计步频率不同造成的计步误差，提高了计步数据的准确性；另外，通过计步有效频率的校正，实现了对计步数据中无效计步进行处理，进一步提高了计步数据的准确性。
本发明实施例还提供了一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行上述的方法。
图2为本发明实施例实现计步的装置的结构程图，如图2所示，包括：第一校正单元和计步单元；其中，
第一校正单元，设置为根据获取的预设步数的计步数据校正三轴加速度传感器的计步频率和/或计步的步幅有效值；
第一校正单元是设置为，获取预设步数的计步数据，
当计步数据中的计步数值大于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内降低三轴加速度传感器的计步频率，将预设频率范围内计步数值与实际行走步数误差最小时的计步频率确定为校正计步频率；
当计步数据中的计步数值小于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内提高三轴加速度传感器的计步频率，将预设频率范围内计步数值与实际行走步数误差最小时的计步频率确定为校正计步频率；和/或，
当计步数据中的计步数值大于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内加大计步的步幅值，将有效步幅区间内计步数值与实际行走步数误差最小时计步的步幅值确定为校正步幅有效值；
当计步数据中的计步数值小于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内减小计步的步幅值，将有效步幅区间内计步数值与实际行走步数误差最小时计步的步幅值确定为校正步幅有效值。
计步单元，设置为根据校正后三轴加速度传感器的校正计步频率和/或计步的校正步幅有效值进行计步。
本发明实施例装置还包括第二校正单元，设置为对计步有效频率进行校正获得校正计步有效频率，以对计步单元的计步数据中无效计步进行处理。
第二校正单元是设置为，获取预设步数的计步数据，
当计步数据中的计步数值大于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内降低有效频率值，将计步有效频率范围内计步数值与实际行走步数误差最小时的有效频率值确定为校正计步有效频率；
当计步数据中的计步数值小于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内提高有效频率值，将计步有效频率范围内计步数值与实际行走步数误差最小时的有效频率值确定为校正计步有效频率。
第一校正单元是设置为，根据获取的预设步数的计步数据，校正三轴加速度传感器的计步频率获得校正计步频率后，根据校正计步频率的计步数据对步幅有效值进行校正。
第二校正单元是设置为，在第一校正单元按照先后顺序完成三轴加速度传感器的计步频率的校正和计步的步幅有效值的校正后，基于校正步幅有效值的计步数据进行计步有效频率的校正。
本发明实施例装置还包括判断赋值单元，设置为
当计步频率的校正未减少计步数值与实际行走步数的误差时，取第一频率作为校正计步频率；
当步幅有效值的校正未减少计步数值与实际行走步数的误差时，取第一步幅作为校正步幅有效值；
当计步有效频率的校正未计步数值与实际行走步数的误差时，取第一有效频率作为校正计步有效频率。
第一校正单元是设置为，基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一频率及计步数值为输入，按照预设步长在预设频率范围内进行计算获得预设频率范围内计步数值与实际行走步数误差最小时的计步频率，确定该频率为校正计步频率；和/或，
基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一步幅及计步数值为输入，按照预设幅值调整单位在有效步幅区间内进行计算获得有效步幅区间内计步数值与实际行走步数误差最小时计步的步幅值，确定该计步的步幅值为校正步幅有效值；
第二校正单元是设置为，基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一计步有效频
率及计步数值为输入，按照预设的频率调整单位在计步有效频率范围内计步数值与实际行走步数误差最小时的有效频率值，确定该有效频率值为校正计步有效频率。
需要说明的是，本发明实施例计步装置可以单独进行计步，也可以嵌入到移动终端，例如手机中进行计步，具体嵌入方式不需要本领域技术人员进行创造性劳动。
以下通过具体实施例对本发明方法进行清楚详细的说明，实施例仅用于陈述本发明，并不用于限定本发明方法的保护范围。
本实施例以三轴加速度传感器的第一频率为10.0Hz、预设频率范围可以取值为8.0Hz～12.0Hz和预设步长为0.5Hz；有效步幅区间可以取值为5.0厘米～10.0厘米、第一步幅为7.5厘米、预设幅值调整单位可以是0.5厘米；有效频率范围根据用户每走一步的时间在200毫秒～1000毫秒确定，第一有效频率以行走一步时间为300毫秒为例，频率调整单位以50毫秒作为调整计算单位。
图3为本发明实施例进行计步的方法的流程图，如图3所示，包括：
步骤300、获取预设步数的计步数据以及实际行走步数；其中，获取实际行走步数包括：三轴加速度传感器实时采集x、y、z三轴加速度数据，计步数值等；
步骤301、根据计步数据中的计步数值与实际行走步数做比较，确定是否进行校正处理；可选的，计步数值大于或小于实际行走步数时，就可以确定进行校正。需要说明的是，在实际应用中可以设置是否进行校正的误差比例，例如、计步数值与实际行走步数误差超过2％时，需要进行校正，具体百分比大小可以根据实际情况进行调整。这里，校正处理包括三轴加速度传感器的计步频率、和/或计步的步幅有效值、和/或计步有效频率的校正。本实施例假设计步数值与实际行走步数存在误差，需要执行步骤3020进行校正；如果没有误差，或误差比例小于技术人员设定的误差比例，则不做校正，直接采用相关的参数进行步骤3030的计步处理。
步骤3020、校正三轴加速度传感器的计步频率；可选的，
当计步数据中的计步数值大于实际行走步数时，从10.0Hz开始按照0.5HZ为步长在10Hz～8Hz范围内降低三轴加速度传感器的计步频率，将预设频率范围内计步数值与实际行走步数误差最小时的计步频率确定为校正计步频率；这里，预设频率范围理论上是8Hz～12Hz，由于第一频率是10Hz且做的是降低计步频率的处理，因此调整范围从理论上缩小至10Hz～8Hz，下述陈述中相关校正类似，不做赘述。
当数据中的计步数值小于实际行走步数时，从10.0Hz开始按照0.5HZ作为步长在10Hz～12Hz范围内提高三轴加速度传感器的计步频率，将预设频率范围内计步数值与实际行走步数误差最小时的计步频率确定为校正计步频率。
当计步频率的校正未减少计步数值与实际行走步数的误差时，取第一频率作为校正计步频率；
步骤3021、校正计步的步幅有效值；可选的，
当计步数据中的计步数值大于实际行走步数时，从7.5厘米开始按照预设幅值调整单位为0.5厘米在7.5厘米～10.0厘米内加大计步的步幅值，将有效步幅区间内计步数值与实际行走步数误差最小时计步的步幅值确定为校正步幅有效值；
当计步数据中的计步数值小于实际行走步数时，从7.5厘米开始按照预设幅值调整单位为0.5厘米在5厘米～7.5厘米内减小计步的步幅值，将有效步幅区间内计步数值与实际行走步数误差最小时计步的步幅值确定为校正步幅有效值。
当步幅有效值的校正未减少计步数值与实际行走步数的误差时，取第一步幅作为校正步幅有效值。
图4为本实施例计步过程中采样的三轴加速度曲线图，如图4所示，其中，横轴为采样点、纵轴代表加速度数值，实线为x轴加速度数据，由点构成的虚线为z轴加速度数据，由小短线构成的虚线为y轴加速度数据；图中x轴加速度数据波动性规律较为明显，可以判定为重力方向的加速度；图中A
点处为方向从向上变化到向下的点，即上极值点；B点为方向从向下变化到向上的点，即下极值点。同理，后续的C点、E点、G点、I点、K点为上极值点；D点、F点、H点、J点、L点为下极值点。上下两极值点间的幅度值作为步幅，满足步幅有效值的大小要求时，进行计步。本实施例未进行校正前，第一步幅为7.5厘米，根据图中A点与B点的幅度值为12厘米，大于7.5厘米；B点与C点的幅度值为4.5厘米，小于7.5厘米；C点与D点的幅度值为3厘米，小于7.5厘米。依次类推，计算可得当按照第一步幅7.5厘米计步时，将获得计步数值为3；假设，此时计步数值小于实际行走步数，则以0.5厘米作为幅值调整单位减小计步的步幅值，即按7厘米、6.5厘米、6厘米的顺序逐步减少步幅值，假设步幅值降低到6厘米，此时计步数值为5，假设此时计步数据准确，则将6厘米确定为校正步幅有效值。
步骤3022、校正计步有效频率；可选的，
当计步数据中的计步数值大于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内降低有效频率值，将计步有效频率范围内计步数值与实际行走步数误差最小时的有效频率值确定为校正计步有效频率；
当计步数据中的计步数值小于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内提高有效频率值，将计步有效频率范围内计步数值与实际行走步数误差最小时的有效频率值确定为校正计步有效频率。
当计步有效频率的校正未减少计步数值与实际行走步数的误差时，取第一有效频率作为校正计步有效频率。
步骤3023、根据校正获得的校正计步频率、校正步幅有效值和校正计步有效频率进行计步。
步骤3030、直接进行计步。
本实施例通过三轴加速度传感器的计步频率避免了计步频率过大时是计步数值偏大，计步频率过小时，计步数值偏小；通过校正计步的步幅有效值避免不同用户由于步幅、行走姿态、行走习惯以及放置计步装置位置不同导
致步幅数据受到干扰；通过校正计步有效频率对不是由于行走造成的计步进行无效处理，例如，用户将计步用的手机放置在了包袋或握持在手中，行走时，手机因为放置在包袋或由于手的滑动，三轴加速度传感器首次影响产生了计步有效频率范围外的数据，对该部分数据进行处理，可以提高计步准确性。本实施例方法经过试验后确定，计步误差由原来的超过8％，降低到2％左右，因此，计步准确性得到提高。
本实施例，校正过程可以通过机器学习算法和迁移学习算法实现，具体的可以是决策树算法、神经网络算法、极速学习机算法或深度学习算法等。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序来指令相关硬件(例如处理器)完成，所述程序可以存储于计算机可读存储介质中，如只读存储器、磁盘或光盘等。可选地，上述实施例的全部或部分步骤也可以使用一个或多个集成电路来实现。相应地，上述实施例中的各模块/单元可以采用硬件的形式实现，例如通过集成电路来实现其相应功能，也可以采用软件功能模块的形式实现，例如通过处理器执行存储于存储器中的程序/指令来实现其相应功能。本发明不限制于任何特定形式的硬件和软件的结合。
任何本发明所属领域内的技术人员，在不脱离本发明所揭露的精神和范围的前提下，可以在实施的形式及细节上进行任何的修改与变化，但本发明的专利保护范围，仍须以所附的权利要求书所界定的范围为准。
上述技术方案避免了由于用户行走步幅、三轴加速度传感器计步频率不同造成的计步误差，提高了计步数据的准确性；另外，通过计步有效频率的校正，实现了对计步数据中无效计步进行处理，进一步提高了计步数据的准确性。
Claims (16)
- 一种实现计步的方法，包括：根据获取的预设步数的计步数据校正三轴加速度传感器的计步频率和/或计步的步幅有效值；根据校正获得的三轴加速度传感器的校正计步频率和/或计步的校正步幅有效值进行计步。
- 根据权利要求1所述的方法，该方法还包括：对计步有效频率进行校正获得校正计步有效频率；根据所述校正计步有效频率对所述计步数据中无效计步进行处理。
- 根据权利要求2所述的方法，其中，所述校正三轴加速度传感器的计步频率包括：当所述计步数据中的计步数值大于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内降低所述三轴加速度传感器的计步频率，将预设频率范围内所述计步数值与所述实际行走步数误差最小时的计步频率确定为所述校正计步频率；当所述计步数据中的计步数值小于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内提高所述三轴加速度传感器的计步频率，将预设频率范围内所述计步数值与所述实际行走步数误差最小时的计步频率确定为所述校正计步频率。
- 根据权利要求2所述的方法，其中，所述校正计步的步幅有效值包括：当所述计步数据中的计步数值大于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内加大计步的步幅值，将有效步幅区间内所述计步数值与所述实际行走步数误差最小时计步的步幅值确定为所述校正步幅有效值；当所述计步数据中的计步数值小于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内减小计步的步幅值，将有效步幅区间内所述计步数值与所述实际行走步数误差最小时计步的步幅值确定为所述校 正步幅有效值。
- 根据权利要求3或4所述的方法，其中，所述对计步有效频率进行校正包括：当所述计步数据中的计步数值大于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内降低有效频率值，将计步有效频率范围内所述计步数值与所述实际行走步数误差最小时的有效频率值确定为所述校正计步有效频率；当所述计步数据中的计步数值小于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内提高有效频率值，将计步有效频率范围内所述计步数值与所述实际行走步数误差最小时的有效频率值确定为所述校正计步有效频率。
- 根据权利要求2所述的方法，其中，所述校正三轴加速度传感器的计步频率和计步的步幅有效值包括：在获得所述校正计步频率后，根据校正计步频率的计步数据对所述步幅有效值进行校正；所述校正三轴加速度传感器的计步频率、计步的步幅有效值和计步有效频率包括：在获得所述校正计步频率后，再根据校正计步频率的计步数据对所述步幅有效值进行参数校正；基于所述校正步幅有效值的计步数据进行所述计步有效频率的校正。
- 根据权利要求5所述的方法，该方法还包括：当所述计步频率的校正未减少计步数值与实际行走步数的误差时，取所述第一频率作为所述校正计步频率；当所述步幅有效值的校正未减少计步数值与实际行走步数的误差时，取所述第一步幅作为所述校正步幅有效值；当所述计步有效频率的校正未减少计步数值与实际行走步数的误差时，取所述第一有效频率作为所述校正计步有效频率。
- 根据权利要求2所述的方法，其中，所述校正计步频率包括：基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一频率及计步数值为输入，按照预设步数在预设频率范围内进行计算获得预设频率范围内所述计步数值与实际行走步数误差最小时的计步频率，确定该频率为所述校正计步频率；所述校正步幅有效值包括：基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一步幅及计步数值为输入，按照预设幅值调整单位在有效步幅区间内进行计算获得有效步幅区间内所述计步数值与实际行走步数误差最小时计步的步幅值，确定该计步的步幅值为所述校正步幅有效值；所述对计步有效频率进行校正包括：基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一计步有效频率及计步数值为输入，按照预设的频率调整单位在计步有效频率范围内所述计步数值与实际行走步数误差最小时的有效频率值，确定该有效频率值为所述校正计步有效频率。
- 一种实现计步的装置，包括：第一校正单元，设置为根据获取的预设步数的计步数据校正三轴加速度传感器的计步频率和/或计步的步幅有效值；计步单元，设置为根据校正后三轴加速度传感器的校正计步频率和/或计步的校正步幅有效值进行计步。
- 根据权利要求9所述的装置，该装置还包括：第二校正单元，设置为对计步有效频率进行校正获得校正计步有效频率，根据所述校正计步有效频率对所述计步单元的计步数据中无效计步进行处理。
- 根据权利要求10所述的方法，其中，所述第一校正单元是设置为通过如下方式实现校正三轴加速度传感器的计步频率：根据获取的预设步数的计步数据，当所述计步数据中的计步数值大于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内降低所述三轴加速度传感器的计步频率，将预设频率范围内所述计步数值与所述实际行走步数误差最小时的计步频率确定为所述校正计步频率；当所述计步数据中的计步数值小于实际行走步数时，从第一频率开始按照预设步长在预设频率范围内提高所述三轴加速度传感器的计步频率，将预设频率范围内所述计步数值与所述实际行走步数误差最小时的计步频率确定为所述校正计步频率；所述第一校正单元是设置为通过如下方式实现校正计步的步幅有效值：当所述计步数据中的计步数值大于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内加大计步的步幅值，将有效步幅区间内所述计步数值与所述实际行走步数误差最小时计步的步幅值确定为所述校正步幅有效值；当所述计步数据中的计步数值小于实际行走步数时，从第一步幅开始按照预设幅值调整单位在有效步幅区间内减小计步的步幅值，将有效步幅区间内所述计步数值与所述实际行走步数误差最小时计步的步幅值确定为所述校正步幅有效值。
- 根据权利要求11所述的装置，其中，所述第二校正单元是设置为通过如下方式实现对计步有效频率进行校正：根据获取的预设步数的计步数据，当所述计步数据中的计步数值大于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内降低有效频率值，将计步有效频率范围内所述计步数值与所述实际行走步数误差最小时的有效频率值确定为所述校正计步有效频率；当所述计步数据中的计步数值小于实际行走步数时，从第一有效频率开始按照预设的频率调整单位在计步有效频率范围内提高有效频率值，将计步有效频率范围内所述计步数值与所述实际行走步数误差最小时的有效频率值 确定为所述校正计步有效频率。
- 根据权利要求12所述的装置，其中，所述第一校正单元是设置为通过如下方式实现校正三轴加速度传感器的计步频率和计步的步幅有效值：根据获取的预设步数的计步数据，校正三轴加速度传感器的计步频率获得所述校正计步频率后，根据校正计步频率的计步数据对所述步幅有效值进行校正；所述第二校正单元是设置为通过如下方式实现校正三轴加速度传感器的计步频率、计步的步幅有效值和计步有效频率：在第一校正单元按照先后顺序完成所述三轴加速度传感器的计步频率的校正和所述计步的步幅有效值的校正后，基于所述校正步幅有效值的计步数据进行所述计步有效频率的校正。
- 根据权利要求12或13所述的装置，该装置还包括：判断赋值单元，设置为当所述计步频率的校正未减少计步数值与实际行走步数的误差时，取所述第一频率作为所述校正计步频率；当所述步幅有效值的校正未减少计步数值与实际行走步数的误差时，取所述第一步幅作为所述校正步幅有效值；当所述计步有效频率的校正未计步数值与实际行走步数的误差时，取所述第一有效频率作为所述校正计步有效频率。
- 根据权利要求10所述的装置，其中，所述第一校正单元是设置为通过如下方式实现校正计步频率：基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一频率及计步数值为输入，按照预设步长在预设频率范围内进行计算获得预设频率范围内所述计步数值与实际行走步数误差最小时的计步频率，确定该频率为所述校正计步频率；所述第一校正单元是设置为通过如下方式实现校正步幅有效值：基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一步幅及计步数值为输入，按照预设幅值调整单位在有效步幅区间内进行计算获得有效步幅区间内所述计步数值与实际行走步数误差最小时计步的步幅值，确定该计步的步幅值为所述校正步 幅有效值；所述第二校正单元是设置为通过如下方式实现对计步有效频率进行校正：基于机器学习算法和迁移学习算法，以计步数据中实际行走步数的三轴加速度传感器的加速度数据信息、第一计步有效频率及计步数值为输入，按照预设的频率调整单位在计步有效频率范围内所述计步数值与实际行走步数误差最小时的有效频率值，确定该有效频率值为所述校正计步有效频率。
- 一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行权利要求1～8中任一项所述的方法。
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US15/757,537 US20180252549A1 (en) | 2015-09-08 | 2015-11-24 | Method and apparatus for realizing step counting |
| EP15889029.3A EP3330674A4 (en) | 2015-09-08 | 2015-11-24 | Method and apparatus for realizing step counting |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510567832.6A CN106500717B (zh) | 2015-09-08 | 2015-09-08 | 一种实现计步的方法及装置 |
| CN201510567832.6 | 2015-09-08 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2016165333A1 true WO2016165333A1 (zh) | 2016-10-20 |
Family
ID=57126963
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2015/095416 WO2016165333A1 (zh) | 2015-09-08 | 2015-11-24 | 一种实现计步的方法及装置 |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US20180252549A1 (zh) |
| EP (1) | EP3330674A4 (zh) |
| CN (1) | CN106500717B (zh) |
| WO (1) | WO2016165333A1 (zh) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110749336A (zh) * | 2019-09-24 | 2020-02-04 | 集美大学 | 一种计步器计步修正方法、装置及非暂存性计算机可读存储介质 |
| CN114563012B (zh) * | 2020-11-27 | 2024-06-04 | 北京小米移动软件有限公司 | 计步方法、装置、设备及存储介质 |
| CN113295182B (zh) * | 2021-05-20 | 2022-11-15 | 北京智慧图科技有限责任公司 | 一种Wi-Fi SLAM轨迹还原方法 |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101206124A (zh) * | 2006-12-21 | 2008-06-25 | 陈侑郁 | 具有自动步伐校正装置的计步器 |
| WO2008081553A1 (ja) * | 2006-12-28 | 2008-07-10 | Fujitsu Limited | 携帯機器、運動強度算出方法、歩幅補正方法、情報機器、及びプログラム |
| CN101881625A (zh) * | 2008-08-19 | 2010-11-10 | 幻音科技(深圳)有限公司 | 步幅校正方法、测距方法及计步装置 |
| US20130273939A1 (en) * | 2012-04-13 | 2013-10-17 | Electronics And Telecommunications Research Institute | Method and apparatus for estimating location of pedestrian using step length estimation model parameters |
| CN104089625A (zh) * | 2014-07-31 | 2014-10-08 | 北京蟠桃互联科技有限公司 | 一种具备自学习能力的计步方法 |
| US20150141873A1 (en) * | 2015-01-29 | 2015-05-21 | Physical Enterprises, Inc. | Systems and Methods for Stride Length Calibration |
| CN104729524A (zh) * | 2013-12-18 | 2015-06-24 | 中国移动通信集团公司 | 一种步长估计方法、计步器及计步系统 |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6473483B2 (en) * | 1998-10-28 | 2002-10-29 | Nathan Pyles | Pedometer |
| KR20050013904A (ko) * | 2003-07-29 | 2005-02-05 | 박승훈 | 휴대기기의 지피에스 등을 이용한 조깅관리장치 및 그 방법 |
| CN101354265B (zh) * | 2008-08-19 | 2010-11-03 | 幻音科技(深圳)有限公司 | 计步方法及计步装置 |
| CN102654405B (zh) * | 2011-03-04 | 2014-09-10 | 美新半导体（无锡）有限公司 | 基于加速度传感器的计步方法及其装置 |
| CN102353383B (zh) * | 2011-06-16 | 2013-04-17 | 浙江大学 | 基于单轴陀螺仪的计步及里程推算方法 |
| US20130085700A1 (en) * | 2011-09-30 | 2013-04-04 | Apple Inc. | Techniques for improved pedometer readings |
| US9734304B2 (en) * | 2011-12-02 | 2017-08-15 | Lumiradx Uk Ltd | Versatile sensors with data fusion functionality |
| US10422814B2 (en) * | 2013-07-18 | 2019-09-24 | Vital Connect, Inc. | Fall detection using machine learning |
| CN104406603B (zh) * | 2014-11-12 | 2018-05-11 | 上海卓易科技股份有限公司 | 一种基于加速度传感器的计步方法和装置 |
-
2015
- 2015-09-08 CN CN201510567832.6A patent/CN106500717B/zh active Active
- 2015-11-24 WO PCT/CN2015/095416 patent/WO2016165333A1/zh active Application Filing
- 2015-11-24 US US15/757,537 patent/US20180252549A1/en not_active Abandoned
- 2015-11-24 EP EP15889029.3A patent/EP3330674A4/en not_active Withdrawn
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101206124A (zh) * | 2006-12-21 | 2008-06-25 | 陈侑郁 | 具有自动步伐校正装置的计步器 |
| WO2008081553A1 (ja) * | 2006-12-28 | 2008-07-10 | Fujitsu Limited | 携帯機器、運動強度算出方法、歩幅補正方法、情報機器、及びプログラム |
| CN101881625A (zh) * | 2008-08-19 | 2010-11-10 | 幻音科技(深圳)有限公司 | 步幅校正方法、测距方法及计步装置 |
| US20130273939A1 (en) * | 2012-04-13 | 2013-10-17 | Electronics And Telecommunications Research Institute | Method and apparatus for estimating location of pedestrian using step length estimation model parameters |
| CN104729524A (zh) * | 2013-12-18 | 2015-06-24 | 中国移动通信集团公司 | 一种步长估计方法、计步器及计步系统 |
| CN104089625A (zh) * | 2014-07-31 | 2014-10-08 | 北京蟠桃互联科技有限公司 | 一种具备自学习能力的计步方法 |
| US20150141873A1 (en) * | 2015-01-29 | 2015-05-21 | Physical Enterprises, Inc. | Systems and Methods for Stride Length Calibration |
Non-Patent Citations (1)
| Title |
|---|
| See also references of EP3330674A4 * |
Also Published As
| Publication number | Publication date |
|---|---|
| EP3330674A4 (en) | 2018-10-17 |
| EP3330674A1 (en) | 2018-06-06 |
| US20180252549A1 (en) | 2018-09-06 |
| CN106500717B (zh) | 2020-05-12 |
| CN106500717A (zh) | 2017-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI518304B (zh) | 一種基於三軸加速計的計步方法和計步器 | |
| Mladenov et al. | A step counter service for Java-enabled devices using a built-in accelerometer | |
| CN103954295B (zh) | 一种基于加速度传感器的计步方法 | |
| US20140074431A1 (en) | Wrist Pedometer Step Detection | |
| US20130085711A1 (en) | Techniques for improved pedometer readings | |
| US20130085700A1 (en) | Techniques for improved pedometer readings | |
| US9702899B2 (en) | Pedometer with lag correction | |
| KR20110139143A (ko) | 보수계 | |
| RU2601152C2 (ru) | Устройство, способ и компьютерная программа для обеспечения предоставления информации пользователю | |
| KR20130138321A (ko) | 모바일 디바이스에서의 우도 함수 값들의 결합을 사용하여 사용자 활동을 분류하기 위한 시스템들, 방법들 및 장치들 | |
| WO2018076205A1 (zh) | 一种步幅校准方法、相关设备及系统 | |
| WO2016165333A1 (zh) | 一种实现计步的方法及装置 | |
| CN109893111B (zh) | 一种动态血压测量模式选择方法及装置 | |
| CN107270931A (zh) | 一种IOS和android平台通用的步态自相关计步器 | |
| US20150241243A1 (en) | Method for counting steps and electronic apparatus using the same | |
| CN105138127A (zh) | 用于穿戴设备的左右手分辨方法、装置及穿戴设备 | |
| CN117782148A (zh) | 步数计量方法、装置、电子设备及可读存储介质 | |
| JP2017192563A (ja) | 活動量計および運動量算出装置 | |
| CN103984857A (zh) | 一种帕金森病情监控系统和方法 | |
| WO2018086321A1 (zh) | 一种计步方法及装置 | |
| JP2018159569A (ja) | 姿勢算出装置、姿勢計測システム、及び姿勢算出方法 | |
| WO2022267290A1 (zh) | 一种计步方法、装置、电子设备及可读存储介质 | |
| JP6033418B2 (ja) | オフセット推定装置、オフセット推定方法、およびプログラム | |
| Uddin et al. | SmartSpaghetti: Accurate and robust tracking of Human's location | |
| TWI699544B (zh) | 定位方法及裝置 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 15889029 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 15757537 Country of ref document: US |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |