CN108854032A - A kind of method, apparatus and intelligent wearable device of detection swimming switch-back point - Google Patents
A kind of method, apparatus and intelligent wearable device of detection swimming switch-back point Download PDFInfo
- Publication number
- CN108854032A CN108854032A CN201810590022.6A CN201810590022A CN108854032A CN 108854032 A CN108854032 A CN 108854032A CN 201810590022 A CN201810590022 A CN 201810590022A CN 108854032 A CN108854032 A CN 108854032A
- Authority
- CN
- China
- Prior art keywords
- peak
- time window
- value
- data
- absolute difference
- 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
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
- A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
- A63B71/0619—Displays, user interfaces and indicating devices, specially adapted for sport equipment, e.g. display mounted on treadmills
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B2220/00—Measuring of physical parameters relating to sporting activity
- A63B2220/80—Special sensors, transducers or devices therefor
- A63B2220/803—Motion sensors
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B2244/00—Sports without balls
- A63B2244/20—Swimming
Abstract
The invention discloses the method, apparatus and intelligent wearable device of a kind of detection swimming switch-back point, this method includes:It acquires exercise data when user swims and calculates the arm cycle of user;According to the length of arm cycle setting time window, the data of acquisition are successively handled according to time window, obtains the number of effectively striking in each time window;When the number of effectively striking in some time window is less than default value, judge the end time point of time window for switch-back point of swimming.The present invention can accurate judgement switch-back point so that determine user whether swum a swimming pool length and reached swimming pool edge, it checks when having swum one time motion state without carrying out button operation when reaching swimming pool edge, user is after swimming it can be seen that the motion state of a swimming, using easier, it can also prevent underwater button operation from destroying the waterproof performance of intelligent wearable device.
Description
Technical field
The present invention relates to intelligent wearable device field, in particular to a kind of the method, apparatus and intelligence of detection swimming switch-back point
It can wearable device.
Background technique
With the development of the society, the run duration of people is fewer and fewer, the working method of sitting makes many human body quality
It gradually goes down hill, fat and various chronic diseases etc. have seriously affected people's lives quality.For the health of itself, people are to fortune
Dynamic also increasingly to pay attention to, domestic various marathon races at present also like the mushrooms after rain, have further confirmed movement in people
The promotion of status in mind.Along with the upsurge of movement, the motion state that user is detected using intelligent wearable device becomes heat
Point.Moving state identification is intelligent wearable device algorithm as the technical foundation for solving motion monitoring and motion state prompting
One of core and difficult point.
It is well known that swimming is the movement for needing whole body to participate in, can mobilize than other movements has more muscle
Group participates in metabolic function.It can be improved the muscle of many muscle powers and harmony, especially trunk, shoulder belt and upper limb.Cause
Need to overcome biggish resistance to swim in the water, swimming is again periodically to move, and long-term take exercise can make muscle strength, speed
The flexibility of degree, endurance and joint is all improved, and is improved cardiovascular system, is improved lung capacity, improve muscle systems ability, change
Kind Thermoregulation ability etc..Increasingly pay attention to today of health and movement in people, swimming also becomes the aerobic fortune that public heat is held in both hands
Dynamic form.Therefore also become one of the emphasis of intelligent wearable device to the monitoring of swimming exercise.
When swimmer wears the intelligent wearable device progress swimming exercise for having motion tracking function, for switch-back point of swimming
Judgement be particularly important.When swimming exercise carries out, user can not once chase after reaching to operate at switch-back point every time
Track program, and for swimming pool swimming, the measurement of movement total length can not be carried out using GPS data, can only pass through swimming pool
Length and swim ring number measure, therefore calculating for the predicting relation for switch-back point of swimming to final swimming exercise total amount.
And traditional intelligence wearable device can only record the information such as simple speed, locomitivity, the heat consumption etc. to estimate user are believed
Breath is unable to judge accurately switch-back point and then can not determine whether user has swum a swimming pool length and reached swimming pool edge, and
It checks and needs to carry out button operation when reaching swimming pool edge when having swum one time motion state, destroy the anti-of intelligent wearable device
Aqueous energy, use is not convenient enough, is unfavorable for motion state and ability that user preferably grasps oneself.
Summary of the invention
The method, apparatus and intelligent wearable device of a kind of detection swimming switch-back point provided by the invention, with solution or part
It solves the problem above-mentioned.
According to an aspect of the invention, there is provided a kind of method of detection swimming switch-back point, the method includes:
It acquires exercise data when user swims and calculates the arm cycle of user;
According to the length of the arm cycle setting time window, the data of acquisition are successively located according to the time window
Reason obtains the number of effectively striking in each time window;
When the number of effectively striking in some described time window is less than default value, the end time of the time window is judged
Point is swimming switch-back point.
Optionally, the exercise data of described pair of acquisition is successively handled according to the time window, when obtaining each described
Between number of effectively striking in window include:
The exercise data in the time window is successively obtained, the exercise data in each time window is filtered,
Any one uniaxial data is chosen from the filtered exercise data;
The uniaxial data are handled, the number of effectively striking in each time window is obtained.
Optionally, described that the uniaxial data are handled, obtain the number packet of effectively striking in each time window
It includes:
Wave crest and trough of the uniaxial data in each time window are recorded, the wave crest and trough are labeled as
Peak value obtains the corresponding peak lists of each time window;
One peak difference threshold value is set according to the peak change range of the uniaxial data, according to the peak difference threshold value to each
The corresponding peak lists of the time window are handled, and the number of the corresponding effective peak of each time window is obtained;
The number of the corresponding effective peak of each time window is halved, effectively striking in each time window is obtained
Number.
Optionally, described that the corresponding peak lists of each time window are handled according to the peak difference threshold value, it obtains every
The number of the corresponding effective peak of a time window includes:
Calculate first between each peak value in the corresponding peak lists of each time window and the adjacent peak on the left of it
Absolute difference and the second absolute difference between the adjacent peak on the right side of it,
When first absolute difference and second absolute difference are all larger than the peak difference threshold value, by the peak
Value is labeled as effective peak;
It, will when there is one to be less than the peak difference threshold value in first absolute difference and second absolute difference
The peak markers are invalid peak value, and are removed from the peak lists;
When first absolute difference and second absolute difference are respectively less than the peak difference threshold value, further obtain
The peak value and the absolute difference being separated by between a peak value on the left of it are taken, if the absolute difference is poor greater than the peak
The peak markers are then effective peak, and the adjacent peak at left and right sides of the peak value are labeled as invalid peak by threshold value
Value, and removed from the peak lists.
Optionally, the time window is at least 1.5 arm cycles, and the arm cycle is according to the exercise data acquired
Wave period determines.
According to another aspect of the present invention, it provides a kind of detection swimming to strike several devices, described device includes depositing
By internal bus communication connection, the memory is stored with can for reservoir and processor, the memory and the processor
The computer program executed by the processor, detection that can be above-mentioned when the computer program is executed by the processor are swum
The method for switch-back point of swimming.
According to a further aspect of the invention, a kind of intelligent wearable device is provided, the intelligent wearable device is built-in with
Inertial sensor, memory and processor, the inertial sensor and memory are connected to the processor respectively, the storage
Device is stored with the computer program that can be executed by the processor, and the computer program can when being executed by the processor
Realize following method and step:
It acquires exercise data when user swims and calculates the arm cycle of user;
According to the length of the arm cycle setting time window, the data of acquisition are successively located according to the time window
Reason obtains the number of effectively striking in each time window;
When the number of effectively striking in some described time window is less than default value, the end time of the time window is judged
Point is swimming switch-back point.
Optionally, the exercise data of described pair of acquisition is successively handled according to the time window, when obtaining each described
Between several step of effectively striking in window be specially:
The exercise data in the time window is successively obtained, the exercise data in each time window is filtered,
Any one uniaxial data is chosen from the filtered exercise data;
The uniaxial data are handled, the number of effectively striking in each time window is obtained.
Optionally, described that the uniaxial data are handled, obtain the number packet of effectively striking in each time window
It includes:
Wave crest and trough of the uniaxial data in each time window are recorded, the wave crest and trough are labeled as peak value
Obtain the corresponding peak lists of each time window;
One peak difference threshold value is set according to the peak change range of the uniaxial data, according to the peak difference threshold value to each
The corresponding peak lists of the time window are handled, and the number of the corresponding effective peak of each time window is obtained;
The number of the corresponding effective peak of each time window is halved, effectively drawing in each time window is obtained
Water number.
Optionally, described that the corresponding peak lists of each time window are handled according to the peak difference threshold value, it obtains
The number of the corresponding effective peak of each time window the step of be specially:
Calculate first between each peak value in the corresponding peak lists of each time window and the adjacent peak on the left of it
Absolute difference and the second absolute difference between the adjacent peak on the right side of it,
When first absolute difference and second absolute difference are all larger than the peak difference threshold value, by the peak
Value is labeled as effective peak;
It, will when there is one to be less than the peak difference threshold value in first absolute difference and second absolute difference
The peak markers are invalid peak value, and are removed from the peak lists;
When first absolute difference and second absolute difference are respectively less than the peak difference threshold value, further obtain
The peak value and the absolute difference being separated by between a peak value on the left of it are taken, if the absolute difference is poor greater than the peak
The peak markers are then effective peak, and the adjacent peak at left and right sides of the peak value are labeled as invalid peak by threshold value
Value, and removed from the peak lists.
Optionally, it is filtered using exercise data of the low pass Chebyshev filter to acquisition;
And/or the time window is at least 1.5 arm cycles, the arm cycle is according to the exercise data acquired
Wave period determines.
The beneficial effect of the embodiment of the present invention is:By acquisition user swimming when exercise data and calculate striking for user
Period;According to the length of arm cycle setting time window, the data of acquisition are successively handled according to time window, is obtained each
Number of effectively striking in time window;When the number of effectively striking in some time window is less than default value, the knot of time window is judged
Beam time point is swimming switch-back point.It can only be recorded simply compared to current intelligent wearable device, such as conventional motion wrist-watch
Velocity information, the present invention can accurately record the swimming data of user, and accurate judgement switch-back point determines whether user swims in turn
A complete swimming pool length and reach swimming pool edge, check when having swum one time motion state without reaching the progress of swimming pool edge when
Button operation, use is easier, can also prevent underwater button operation from destroying the waterproof performance of intelligent wearable device;It can be complete quasi-
Really the exercise data all to user is recorded and is analyzed, and is accurately switched to the real time kinematics state of user automatically,
Facilitate motion state and ability that user better grasps oneself.
Detailed description of the invention
Fig. 1 is a kind of method flow diagram of detection swimming switch-back point provided in an embodiment of the present invention;
Fig. 2 is the method flow diagram of another detection swimming switch-back point provided in an embodiment of the present invention;
Fig. 3 is untreated three-axis moving data waveform figure;
Fig. 4 is the three-axis moving data waveform figure by filtering noise reduction;
Fig. 5 is the x-axis waveforms of the motion data figure chosen;
Fig. 6 is the schematic diagram that wave crest and trough are labeled as peak value;
Fig. 7 be when one fluctuate rising edge or failing edge occur small vibration when the case where schematic diagram;
Fig. 8 is the case where wave crest fluctuated when one or trough are there are when small vibration schematic diagram;
Fig. 9 is a kind of device figure of detection swimming switch-back point provided in an embodiment of the present invention;
Figure 10 is a kind of intelligent wearable device schematic diagram provided in an embodiment of the present invention.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention
Formula is described in further detail.Embodiment described in following exemplary embodiment does not represent consistent with the application
All embodiments.On the contrary, they are only and some aspects phase one as detailed in the attached claim, the application
The example of the device and method of cause.
It is only to be not intended to be limiting the application merely for for the purpose of describing particular embodiments in term used in this application.
It is also intended in the application and the "an" of singular used in the attached claims, " described " and "the" including majority
Form, unless the context clearly indicates other meaning.
Fig. 1 is a kind of method flow diagram of detection swimming switch-back point provided in an embodiment of the present invention, and this method includes following
Step:
Step S11:It acquires exercise data when user swims and calculates the arm cycle of user;
Step S12:According to the length of arm cycle setting time window, the data of acquisition are successively located according to time window
Reason obtains the number of effectively striking in each time window;Specially:The successively exercise data in acquisition time window, to each time
Exercise data in window is filtered, any one uniaxial data is chosen from filtered exercise data;Record number of axle evidence exists
Wave crest and trough are obtained the corresponding peak lists of each time window labeled as peak value by wave crest and trough in each time window;
One peak difference threshold value is set according to the peak change range of uniaxial data, according to peak difference threshold value to the corresponding peak value of each time window
List is handled, and the number of the corresponding effective peak of each time window is obtained;
Step S13:When the number of effectively striking in some time window is less than default value, the end time of time window is judged
Point is swimming switch-back point.
Wherein, described in step S12 " the corresponding peak lists of each time window are handled according to peak difference threshold value,
Obtain the number of the corresponding effective peak of each time window " include:Calculate each of corresponding peak lists of each time window
The first absolute difference between peak value and adjacent peak on the left of it and second between the adjacent peak on the right side of it poor
It is worth absolute value, is effective peak by peak markers when the first absolute difference and the second absolute difference are all larger than peak difference threshold value
Value;It is invalid peak by peak markers when there is one to be less than peak difference threshold value in the first absolute difference and the second absolute difference
Value, and removed from peak lists;When the first absolute difference and the second absolute difference are respectively less than peak difference threshold value, further
Peak value and the absolute difference being separated by between a peak value on the left of it are obtained, it, will if absolute difference is greater than peak difference threshold value
Peak markers are effective peak, and the adjacent peak at left and right sides of peak value are labeled as invalid peak value, and from peak lists
Removal.
Wherein time window is at least 1.5 arm cycles, and arm cycle is true according to the wave period of the exercise data of acquisition
It is fixed;Denoiser is filtered using low pass Chebyshev, noise reduction is filtered to the exercise data in each time window.
The present invention passes through a Chebyshev using the data of the MEMS sensor acquisition built in intelligent wearable device first
Low-pass filtering denoiser (Chebyshev Filter), is filtered noise reduction process to initial data, then according to the dynamic of user
Make feature and judges whether user returns to corresponding result after the completion of switch-back point, judgement.Intelligent wearable device include Intelligent bracelet,
Smart motion wrist-watch etc. elaborates several detection method of striking of swimming by taking smart motion wrist-watch as an example, and Fig. 2 is that the present invention is real
The method flow diagram that another detection swimming switch-back point of example offer is provided, as shown in Fig. 2, this approach includes the following steps:
Step S21:Acquire exercise data;User first turns on the swimming tracking function of smart motion wrist-watch, built in
MEMS sensor starts to acquire three-axis moving data;
Step S22:Judge whether to start to strike;Whether automatic monitoring user starts arm stroke, if so, operation
Step S23, if it is not, then operating procedure S21;
Step S23:It acquires and obtains user and start the exercise data after striking;
Step S24:The exercise data of acquisition is successively handled according to time window, is obtained effective in each time window
It strikes number;
The time window of a default appropriate length, the length of time window is too long to will increase calculation amount, too short, is unable to get
Effective information.It is found by investigation, the length of time window at least will be more than that 1.5 arm cycles could effecting reaction movement shape
State, but do not exceed the time for having swum a swimming pool length yet.
Original exercise data includes noise, is unfavorable for handling, as shown in figure 3, Fig. 3 is untreated three-axis moving
The waveform burr of data waveform figure, exercise data is excessive, and contains high fdrequency component, these invalid data are unfavorable for movement
Analysis, therefore be filtered noise reduction process firstly the need of to initial data.We use a low pass Chebyshev filter
Noise reduction process is filtered to data, cut-off frequecy of passband is set as 1Hz, after Chebyshev filter is handled, data
High frequency noise components are just filtered.Fig. 4 is by the three-axis moving data waveform figure of filtering noise reduction, as shown in figure 4, from process
The three-axis moving data that Chebyshev filter was treated once swim, it can be seen that when swimming is struck, three axis fortune
Dynamic data trend having the same (waveform between the 1770-1820 second of Fig. 3).Therefore we roll over only with x-axis data
Return the benefit the calculating of judgement, not only ensure that accuracy of judgement degree in this way, but also can reduce operand to the maximum extent.
Fig. 5 is the x-axis waveforms of the motion data figure chosen, from figure 5 it can be seen that exercise data when striking is in periodically
Fluctuation, each period correspondence once strike, and when strike stop when, then be that user reaches swimming pool edge, to carry out once
It turns back movement.We just use this feature to judge switch-back point.
All wave crest and trough are found in data in time window, Fig. 6 is wave crest and trough showing labeled as peak value
Be intended to, as shown in fig. 6, black circle is labeled as trough, ash circle is labeled as wave crest, wave crest and trough by find in region maximum and
The method of minimum determines;Then wave crest and trough are uniformly labeled as peak value and obtain the corresponding peak lists of each time window,
Remove the lower vibration of amplitude in waveform, retains the main component in waveform and be labeled as once effectively striking.Specific way is first
A peak difference threshold value is first arranged according to the peak change range of the uniaxial data, then calculate some peak value and adjacent peak it
Between absolute difference whether be greater than the threshold value as judgment criteria, obtain the number of the corresponding effective peak of each time window;
It is specifically divided into following several situations:
Each peak value f (x in the corresponding peak lists of each time window is calculated firstn) with its on the left of adjacent peak f
(xn-1) between the first absolute difference | f (xn-1)-f(xn) | and with the adjacent peak f (x on the right side of itn+1) between second
Absolute difference | f (xn)-f(xn+1) |,
1, when primary fluctuation itself is the main component of waveform, it is desirable that the peak value specifically shaken and its left and right sides phase
The absolute difference of adjacent peak value is all larger than threshold value.I.e.:
|f(xn)-f(xn+1) | > T& | f (xn-1)-f(xn) | > T
Meet this requirement, then judges that the secondary undulation itself is the main component of waveform, retaining the peak value is effective peak
Value.
2, it when small fluctuation occur in the rising edge or failing edge that once fluctuate, needs to disappear the influence of this minor swing
It removes.It is when the peak value and the absolute difference of its left and right peak value of this minor swing are all larger than preset threshold, i.e., in algorithm:
|f(xn)-f(xn+1) | > T& | f (xn-1)-f(xn) | > T
Then judge the peak value of the secondary minor swing for effective peak;Peak value at left and right sides of the peak value of this minor swing and its
This peak value is judged as invalid peak value when absolute difference cannot meet simultaneously greater than threshold value, while it being arranged from effective peak
It is removed in table.
As it can be seen that the decision condition is identical as the Rule of judgment in the 1st kind of situation.Fig. 7 be when one fluctuation rising edge or
The case where when small vibration occurs in failing edge schematic diagram, as shown in fig. 7, in Fig. 7 A, peak value 2 and 3 is what rising edge side occurred
Small sample perturbations judge that peak value 2 and 3 is invalid peak value by algorithm, and peak value 1,4 and 5 is effective peak.Display passes through in Fig. 7 B
This Rule of judgment has successfully filtered the peak value of invalid fluctuation within a narrow range, remains effective peak 1,4 and 5.
3, when the peak value once fluctuated nearby minor swing occurs, it is also desirable to eliminate the influence of this minor swing.At this time
The characteristics of be to occur 3 peak values near this effective peak, but one can only be retained.Method proposed by the present invention is to retain
That among 3 peak values.The characteristics of interpeak is that the absolute difference of it and two neighboring peak value is less than threshold value simultaneously,
But the peak difference values absolute value being separated by with left side meets the requirement for being greater than threshold value, i.e.,:
|f(xn)-f(xn+1) | < T& | f (xn-1)-f(xn) | < T& | f (xn)-f(xn-2) | > T
Fig. 8 is the case where wave crest fluctuated when one or trough are there are when small vibration schematic diagram, as shown in figure 8, Fig. 8 A
In, occur two small sample perturbations at left and right sides of peak value 3, the peak value of the two small sample perturbations is peak value 2 and 4, peak value 3 and peak value
2 and 4 absolute difference is respectively less than threshold value, but is greater than threshold value with the absolute difference of peak value 1, meets above-mentioned Rule of judgment, because
This peak value 3 is effective peak, and peak value 2 and 4 is invalid peak value.As shown in Figure 8 B, two at effective peak 3 have been filtered small to disturb
The raw peak value 2 and 4 of movable property, retains an effective peak 3.
The peak value of small sample perturbations generation can be filtered out by above-mentioned algorithm, retain effective peak.Count each time window
The number of effective peak is halved the number of effectively striking obtained in each time window by the number of corresponding effective peak.
Step S25:Judgement effectively strikes number whether less than default value, if so, operating procedure S26;If it is not, then running
Step S23;
Step S26:The end time point for counting the time window for whether being less than default value that will effectively strike is determined as swimming folding
It returns the benefit;
When effectively strike stop when, i.e., prompt current point in time be swimming switch-back point.Specific practice is when in time window
When number of effectively striking is less than preset value, that is, judge the end time point of the time window for switch-back point of swimming.The setting of preset value with
Time window length is related, and preset value outline is lower than the arm cycle number that time window includes.If the time window being arranged
Length is 1.5 arm cycles, then at the end of judging the time window when the number of striking in some time window is less than 1 time
Between point for swimming switch-back point.
Fig. 9 is a kind of device figure that detection swimming is struck several provided in an embodiment of the present invention, as shown in figure 9, the device 90
Including:Including memory 901 and processor 902, connected between memory 901 and processor 902 by the communication of internal bus 903
It connects, memory 901 is stored with the computer program that can be executed by processor 902, energy when computer program is executed by processor
Enough realize the method and step of above-mentioned detection swimming switch-back point.
In various embodiments, memory 901 can be memory or nonvolatile memory.It is wherein non-volatile to deposit
Reservoir can be:Memory driver (such as hard disk drive), solid state hard disk, any kind of storage dish (such as CD, DVD),
Perhaps similar storage medium or their combination.Memory can be:RAM (Radom Access Memory, arbitrary access
Memory), volatile memory, nonvolatile memory, flash memory.Further, nonvolatile memory and memory can as machine
Storage medium is read, can store the computer program executed by processor 902 thereon, realizes that detection above-mentioned is swum several side of striking
Method, this method have been elaborated in the embodiment that Fig. 1 and Fig. 2 are provided, and details are not described herein.
Figure 10 is a kind of intelligent wearable device schematic diagram provided in an embodiment of the present invention, and as shown in Figure 10, this is intelligently dressed
Equipment 100 is built-in with inertial sensor 1001, memory 901 and processor 902, and inertial sensor 1001 and memory 901 divide
It is not connect with processor 902, memory 901 is stored with the computer program that can be executed by processor 902, computer program quilt
Processor 902 can be realized following method and step when executing:
Exercise data when user swims is acquired using inertial sensor 1001;
The exercise data of acquisition is successively handled according to time window, obtains the number of effectively striking in each time window;
When the number of effectively striking in some time window is less than default value, judge the end time point of time window for swimming
Switch-back point.
In some embodiments, the exercise data of acquisition is successively handled according to time window, obtains each time window
Interior several step of effectively striking is specially:
The successively exercise data in acquisition time window is filtered noise reduction to the exercise data in each time window, from filter
Any one uniaxial data is chosen in exercise data after wave noise reduction;
The record number of axle obtains each according to the wave crest and trough in each time window, by wave crest and trough labeled as peak value
The corresponding peak lists of time window;
One peak difference threshold value is set according to the peak change range of uniaxial data, according to peak difference threshold value to each time window pair
The peak lists answered are handled, and the number of the corresponding effective peak of each time window is obtained;
The number of the corresponding effective peak of each time window is halved, the number of effectively striking in each time window is obtained.
In some embodiments, the corresponding peak lists of each time window are handled according to peak difference threshold value, is obtained every
The step of number of the corresponding effective peak of a time window is specially:
Calculate first between each peak value in the corresponding peak lists of each time window and the adjacent peak on the left of it
Absolute difference and the second absolute difference between the adjacent peak on the right side of it,
It is effective by peak markers when the first absolute difference and second absolute difference are all larger than peak difference threshold value
Peak value;
It is nothing by peak markers when there is one to be less than peak difference threshold value in the first absolute difference and the second absolute difference
Peak value is imitated, and is removed from peak lists;
When the first absolute difference and the second absolute difference are respectively less than peak difference threshold value, further obtain peak value and it is left
Peak markers are effective if absolute difference is greater than peak difference threshold value by the absolute difference of side being separated by between a peak value
Peak value, and the adjacent peak at left and right sides of peak value is labeled as invalid peak value, and remove from peak lists.
In some embodiments, denoiser is filtered using low pass Chebyshev and drop is filtered to the exercise data of acquisition
It makes an uproar;And/or the length of time window is greater than 1.5 arm cycles.
For device embodiment, since it corresponds essentially to embodiment of the method, so related place is referring to method reality
Apply the part explanation of example.The apparatus embodiments described above are merely exemplary, can select according to the actual needs
Some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying wound
In the case that the property made is worked, it can understand and implement.
In conclusion passing through exercise data when acquisition user swims and calculating the arm cycle of user;According to week of striking
The length of phase setting time window successively handles the data of acquisition according to time window, obtains effective in each time window
It strikes number;When the number of effectively striking in some time window is less than default value, judge the end time point of time window for swimming
Switch-back point.The information such as simple speed can only be recorded compared to current intelligent wearable device, such as conventional motion wrist-watch, to estimate
The information such as locomitivity, the heat consumption of user are calculated, the present invention can accurately record the swimming data of user, accurate judgement folding
Return the benefit and then determine whether user has swum a swimming pool length and reached swimming pool edge, check when having swum one time motion state without
Button operation need to be carried out when reaching swimming pool edge, user is after swimming it can be seen that the motion state of a swimming, makes
With easier, it can also prevent underwater button operation from destroying the waterproof performance of intelligent wearable device;Can complete and accurate to user
All exercise datas are recorded and are analyzed, and are accurately switched to the real time kinematics state of user automatically, facilitate user
Better grasp oneself motion state and ability;And it is drawn compared to swimming is calculated using Fourier transformation or PCA scheduling algorithm
Water number, the present invention are calculated several method of striking when swimming using the Wave crest and wave trough number for extracting movement meter waveform, are greatly lowered
Operand is more applicable in the computing capability of intelligent wearable device.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent replacement, improvement and so within the spirit and principles in the present invention, are all contained in protection scope of the present invention
It is interior.
Claims (11)
1. a kind of method of detection swimming switch-back point, which is characterized in that the method includes:
It acquires exercise data when user swims and calculates the arm cycle of user;
According to the length of the arm cycle setting time window, the data of acquisition are successively handled according to the time window,
Obtain the number of effectively striking in each time window;
When the number of effectively striking in some described time window is less than default value, judge that the end time point of the time window is
Swimming switch-back point.
2. the method as described in claim 1, which is characterized in that described pair acquisition exercise data according to the time window successively
It is handled, the number of effectively striking obtained in each time window includes:
The exercise data in the time window is successively obtained, the exercise data in each time window is filtered, from filter
Any one uniaxial data is chosen in the exercise data after wave;
The uniaxial data are handled, the number of effectively striking in each time window is obtained.
3. method according to claim 2, which is characterized in that it is described that the uniaxial data are handled, obtain each institute
The effectively number of striking stated in time window includes:
Wave crest and trough of the uniaxial data in each time window are recorded, the wave crest and trough are labeled as peak value
Obtain the corresponding peak lists of each time window;
One peak difference threshold value is set according to the peak change range of the uniaxial data, according to the peak difference threshold value to each described
The corresponding peak lists of time window are handled, and the number of the corresponding effective peak of each time window is obtained;
The number of the corresponding effective peak of each time window is halved, the number of effectively striking in each time window is obtained.
4. method as claimed in claim 3, which is characterized in that described corresponding to each time window according to the peak difference threshold value
Peak lists are handled, and the number for obtaining the corresponding effective peak of each time window includes:
Calculate the first difference between each peak value in the corresponding peak lists of each time window and the adjacent peak on the left of it
Absolute value and the second absolute difference between the adjacent peak on the right side of it,
When first absolute difference and second absolute difference are all larger than the peak difference threshold value, by the peak value mark
It is denoted as effective peak;
It, will be described when there is one to be less than the peak difference threshold value in first absolute difference and second absolute difference
Peak markers are invalid peak value, and are removed from the peak lists;
When first absolute difference and second absolute difference are respectively less than the peak difference threshold value, institute is further obtained
Peak value and the absolute difference being separated by between a peak value on the left of it are stated, if the absolute difference is greater than the peak difference threshold
It is worth, then is effective peak by the peak markers, and the adjacent peak at left and right sides of the peak value is labeled as invalid peak value,
And it is removed from the peak lists.
5. the method as described in claim 1, which is characterized in that the time window is at least 1.5 arm cycles, described to strike
Period determines according to the wave period of the exercise data of acquisition.
Several device 6. a kind of detection swimming is struck, which is characterized in that described device includes memory and processor, the storage
By internal bus communication connection, the memory is stored with the calculating that can be executed by the processor for device and the processor
Machine program, the computer program can be realized detection described in claim 1-5 any one when being executed by the processor
The method for switch-back point of swimming.
7. a kind of intelligent wearable device, which is characterized in that the intelligent wearable device is built-in with inertial sensor, memory and place
Device is managed, the inertial sensor and memory are connected to the processor respectively, and the memory is stored with can be by the place
The computer program that device executes is managed, the computer program can be realized following method and step when being executed by the processor:
It acquires exercise data when user swims and calculates the arm cycle of user;
According to the length of the arm cycle setting time window, the data of acquisition are successively handled according to the time window,
Obtain the number of effectively striking in each time window;
When the number of effectively striking in some described time window is less than default value, judge that the end time point of the time window is
Swimming switch-back point.
8. intelligent wearable device as claimed in claim 7, which is characterized in that described pair acquisition exercise data according to it is described when
Between window successively handled, the several step of effectively striking obtained in each time window is specially:
The exercise data in the time window is successively obtained, the exercise data in each time window is filtered, from filter
Any one uniaxial data is chosen in the exercise data after wave;
The uniaxial data are handled, the number of effectively striking in each time window is obtained.
9. intelligent wearable device as claimed in claim 8, which is characterized in that it is described that the uniaxial data are handled, it obtains
Number of effectively striking in each time window includes:
Wave crest and trough of the uniaxial data in each time window are recorded, the wave crest and trough are obtained labeled as peak value
The corresponding peak lists of each time window;
One peak difference threshold value is set according to the peak change range of the uniaxial data, according to the peak difference threshold value to each described
The corresponding peak lists of time window are handled, and the number of the corresponding effective peak of each time window is obtained;
The number of the corresponding effective peak of each time window is halved, effectively striking in each time window is obtained
Number.
10. intelligent wearable device as claimed in claim 9, which is characterized in that it is described according to the peak difference threshold value to each institute
The step of stating the corresponding peak lists of time window to be handled, obtaining the number of the corresponding effective peak of each time window has
Body is:
Calculate the first difference between each peak value in the corresponding peak lists of each time window and the adjacent peak on the left of it
Absolute value and the second absolute difference between the adjacent peak on the right side of it,
When first absolute difference and second absolute difference are all larger than the peak difference threshold value, by the peak value mark
It is denoted as effective peak;
It, will be described when there is one to be less than the peak difference threshold value in first absolute difference and second absolute difference
Peak markers are invalid peak value, and are removed from the peak lists;
When first absolute difference and second absolute difference are respectively less than the peak difference threshold value, institute is further obtained
Peak value and the absolute difference being separated by between a peak value on the left of it are stated, if the absolute difference is greater than the peak difference threshold
It is worth, then is effective peak by the peak markers, and the adjacent peak at left and right sides of the peak value is labeled as invalid peak value,
And it is removed from the peak lists.
11. intelligent wearable device as claimed in claim 8, which is characterized in that
It is filtered using exercise data of the low pass Chebyshev filter to acquisition;
And/or the time window is at least 1.5 arm cycles, waveform of the arm cycle according to the exercise data of acquisition
Period determines.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810590022.6A CN108854032A (en) | 2018-06-08 | 2018-06-08 | A kind of method, apparatus and intelligent wearable device of detection swimming switch-back point |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810590022.6A CN108854032A (en) | 2018-06-08 | 2018-06-08 | A kind of method, apparatus and intelligent wearable device of detection swimming switch-back point |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108854032A true CN108854032A (en) | 2018-11-23 |
Family
ID=64338781
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810590022.6A Pending CN108854032A (en) | 2018-06-08 | 2018-06-08 | A kind of method, apparatus and intelligent wearable device of detection swimming switch-back point |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108854032A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115155044A (en) * | 2022-07-13 | 2022-10-11 | 杭州光粒科技有限公司 | Method, device, equipment and medium for determining swimming turn-around time |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2465824A (en) * | 2008-12-03 | 2010-06-09 | James Christopher Irlam | Motion analysis device for sports |
CN102481478A (en) * | 2009-03-27 | 2012-05-30 | 英福摩迅运动科技公司 | Monitoring of physical training events |
US20120245714A1 (en) * | 2009-07-17 | 2012-09-27 | Neal Mueller | System and method for counting swimming laps |
CN103837157A (en) * | 2014-03-26 | 2014-06-04 | 向杰 | Motion measuring method and device |
GB2511833A (en) * | 2013-03-15 | 2014-09-17 | Suunto Oy | Device and method for monitoring swimming performance |
CN105536228A (en) * | 2016-02-01 | 2016-05-04 | 安徽北斗易通信息技术有限公司 | Acceleration sensor based digital fighting evaluation system |
CN106334307A (en) * | 2015-07-07 | 2017-01-18 | 天彩电子(深圳)有限公司 | Method for monitoring swimming |
CN107376247A (en) * | 2017-08-16 | 2017-11-24 | 广东远峰电子科技股份有限公司 | A kind of swimming exercise analysis method based on intelligent watch and the intelligent watch |
CN107469326A (en) * | 2017-07-04 | 2017-12-15 | 广东乐心医疗电子股份有限公司 | Swimming monitoring method and device for wearable equipment and wearable equipment |
-
2018
- 2018-06-08 CN CN201810590022.6A patent/CN108854032A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2465824A (en) * | 2008-12-03 | 2010-06-09 | James Christopher Irlam | Motion analysis device for sports |
CN102481478A (en) * | 2009-03-27 | 2012-05-30 | 英福摩迅运动科技公司 | Monitoring of physical training events |
US20120245714A1 (en) * | 2009-07-17 | 2012-09-27 | Neal Mueller | System and method for counting swimming laps |
GB2511833A (en) * | 2013-03-15 | 2014-09-17 | Suunto Oy | Device and method for monitoring swimming performance |
CN103837157A (en) * | 2014-03-26 | 2014-06-04 | 向杰 | Motion measuring method and device |
CN106334307A (en) * | 2015-07-07 | 2017-01-18 | 天彩电子(深圳)有限公司 | Method for monitoring swimming |
CN105536228A (en) * | 2016-02-01 | 2016-05-04 | 安徽北斗易通信息技术有限公司 | Acceleration sensor based digital fighting evaluation system |
CN107469326A (en) * | 2017-07-04 | 2017-12-15 | 广东乐心医疗电子股份有限公司 | Swimming monitoring method and device for wearable equipment and wearable equipment |
CN107376247A (en) * | 2017-08-16 | 2017-11-24 | 广东远峰电子科技股份有限公司 | A kind of swimming exercise analysis method based on intelligent watch and the intelligent watch |
Non-Patent Citations (1)
Title |
---|
郑治真: "《波谱分析基础》", 28 February 1983, 地震出版社 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115155044A (en) * | 2022-07-13 | 2022-10-11 | 杭州光粒科技有限公司 | Method, device, equipment and medium for determining swimming turn-around time |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104406604B (en) | A kind of step-recording method | |
CN107376247B (en) | A kind of swimming exercise analysis method based on smartwatch and the smartwatch | |
JP6403696B2 (en) | Physical activity monitoring device and method | |
JP6564328B2 (en) | Extending gameplay with physical activity monitoring devices | |
CN106175781A (en) | Utilize method and the wearable device of wearable device monitoring swimming state | |
CN105509763B (en) | Minimizing technology and device are interfered in step counting | |
CN105902257B (en) | Sleep state analysis method and device, intelligent wearable device | |
CN101091832B (en) | Swimming lap counter | |
CN104656896B (en) | The method and apparatus for determining input information | |
CN105496416A (en) | Human motion state recognition method and device | |
EP2414055A1 (en) | Swim monitor | |
CN105664454B (en) | The method and apparatus for realizing sports equipment tachometric survey | |
CN104567912B (en) | Method for realizing pedometer on Android mobile phone | |
CN107469326A (en) | Swimming monitoring method and device for wearable equipment and wearable equipment | |
CN106289309A (en) | Step-recording method based on 3-axis acceleration sensor and device | |
CN106334307A (en) | Method for monitoring swimming | |
CN107421560A (en) | A kind of step-recording method, device and wrist type pedometer | |
CN103997572A (en) | Step counting method and device based on data of acceleration sensor of mobile phone | |
CN107393260B (en) | Sedentariness reminding method and device and wrist type sedentariness reminder | |
CN103941874B (en) | Recognition methodss and equipment | |
WO2010090867A2 (en) | Multi-state performance monitoring system | |
CN106389097B (en) | A kind of intelligent massaging gloves | |
CN105758404A (en) | Real-time positioning method and system of intelligent equipment | |
WO2018082190A1 (en) | Ecg signal processing method and apparatus | |
CN108837476A (en) | A kind of method, apparatus and intelligent wearable device of detection swimming starting point |
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 | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20191115 Address after: 266104 Laoshan Qingdao District North House Street investment service center room, Room 308, Shandong Applicant after: GOERTEK TECHNOLOGY Co.,Ltd. Address before: 266061, No. 3, building 18, Qinling Mountains Road, Laoshan District, Shandong, Qingdao 401 Applicant before: QINGDAO ZHENSHI TECHNOLOGY CO.,LTD. |
|
TA01 | Transfer of patent application right | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181123 |
|
RJ01 | Rejection of invention patent application after publication |