WO2018036316A1 - 利用可穿戴设备监测游泳状态的方法及可穿戴设备 - Google Patents

利用可穿戴设备监测游泳状态的方法及可穿戴设备 Download PDF

Info

Publication number
WO2018036316A1
WO2018036316A1 PCT/CN2017/093724 CN2017093724W WO2018036316A1 WO 2018036316 A1 WO2018036316 A1 WO 2018036316A1 CN 2017093724 W CN2017093724 W CN 2017093724W WO 2018036316 A1 WO2018036316 A1 WO 2018036316A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
swimming
stroke
user
test data
Prior art date
Application number
PCT/CN2017/093724
Other languages
English (en)
French (fr)
Inventor
谢馥励
张一凡
Original Assignee
歌尔股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 歌尔股份有限公司 filed Critical 歌尔股份有限公司
Priority to US16/327,597 priority Critical patent/US11517789B2/en
Publication of WO2018036316A1 publication Critical patent/WO2018036316A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B24/00—Electric or electronic controls for exercising apparatus of preceding groups; Controlling or monitoring of exercises, sportive games, training or athletic performances
    • A63B24/0062—Monitoring athletic performances, e.g. for determining the work of a user on an exercise apparatus, the completed jogging or cycling distance
    • 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
    • A61B5/1116—Determining posture transitions
    • 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
    • A61B5/1121—Determining geometric values, e.g. centre of rotation or angular range of movement
    • 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
    • A61B5/1123—Discriminating type of movement, e.g. walking or running
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6802—Sensor mounted on worn items
    • A61B5/681—Wristwatch-type devices
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6813—Specially adapted to be attached to a specific body part
    • A61B5/6824—Arm or wrist
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7246—Details of waveform analysis using correlation, e.g. template matching or determination of similarity
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
    • 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
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2503/00—Evaluating a particular growth phase or type of persons or animals
    • A61B2503/10—Athletes
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B24/00—Electric or electronic controls for exercising apparatus of preceding groups; Controlling or monitoring of exercises, sportive games, training or athletic performances
    • A63B24/0062—Monitoring athletic performances, e.g. for determining the work of a user on an exercise apparatus, the completed jogging or cycling distance
    • A63B2024/0068—Comparison to target or threshold, previous performance or not real time comparison to other individuals
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B2208/00—Characteristics or parameters related to the user or player
    • A63B2208/03—Characteristics or parameters related to the user or player the user being in water
    • 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
    • A63B2220/40—Acceleration
    • 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
    • A63B2220/83—Special sensors, transducers or devices therefor characterised by the position of the sensor
    • A63B2220/836—Sensors arranged on the body of the user
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B2230/00—Measuring physiological parameters of the user
    • A63B2230/62—Measuring physiological parameters of the user posture
    • 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

Definitions

  • the present invention relates to the field of wearable device technologies, and in particular, to a method for monitoring a swimming state using a wearable device and a wearable device.
  • the invention provides a method for monitoring swimming state by using a wearable device and a wearable device, so as to solve the existing technical solution for monitoring and recognizing the swimming state in the field of intelligent sports watches, resulting in intelligent sports.
  • the user experience of the watch is not good.
  • a method for monitoring a swimming state using a wearable device comprising: setting a swimming mode in a wearable device, and storing pre-acquisition standard swimming motion data as corresponding template data,
  • the swimming mode is started according to an instruction issued by a user who is about to enter the water, and after the swimming mode is started, the control sensor collects swimming movement data of the user;
  • the test data for identifying the swimming state of the user is obtained from the swimming action data.
  • test data is matched with each template data.
  • the swimming state of the user is identified as the swimming state corresponding to the template data associated with the test data.
  • a wearable device in which a sensor and a swimming mode are provided, including:
  • a template storage unit configured to store pre-acquisition standard swimming motion data as corresponding template data
  • the test data collection unit is configured to start a swimming mode according to an instruction issued by a user who is about to enter the water when the monitoring process starts, the control sensor collects the swimming motion data of the user, and obtains test data for identifying the swimming state of the user from the swimming motion data.
  • the swimming state identifying unit is configured to match the test data with each template data, and when there is matching test data, identify the swimming state of the user as the swimming state corresponding to the template data associated with the test data.
  • the invention has the beneficial effects that the technical solution of the present invention monitors the swimming state by using the wearable device, sets the swimming mode in the wearable device, and stores the pre-acquisition standard swimming motion data as the corresponding template data, when a monitoring process starts.
  • the swimming mode is started according to the instruction issued by the user who is about to enter the water. After the swimming mode is started, the user's swimming motion data is collected to obtain test data, and the test data is matched with each template data to identify the swimming state of the user.
  • traditional sports watches which can simply record speed and other information to estimate the user's calorie consumption and other information
  • the present invention can accurately identify the swimming posture used by the user during swimming, and helps the user to better grasp his swimming movement. The state is also very helpful for improving the swimming posture.
  • the user only needs to switch the wearable device to the swimming mode before swimming in the water, and the recording and analysis of the swimming state data can be completed without any operation during the swimming process, and the use is more convenient. It also prevents the underwater button operation from damaging the waterproof performance of the watch.
  • FIG. 1 is a flow chart of a method for monitoring a swimming state using a wearable device according to an embodiment of the present invention
  • FIG. 2 is a flow chart of a method for monitoring a swimming state using a wearable device according to another embodiment of the present invention
  • FIG. 3 is a waveform diagram of a three-axis accelerometer data of a breaststroke in one embodiment of the present invention
  • FIG. 5 is a waveform diagram of an accelerometer data for intercepting one of the stroke periods of the freestyle swimming according to an embodiment of the present invention
  • FIG. 6 is a summation, first-order norm, second-order norm waveform diagram of 3-axis accelerometer data for freestyle and breaststroke, respectively, according to an embodiment of the present invention.
  • Figure 7 is a schematic diagram of the normalization of two time series according to an embodiment of the present invention.
  • FIG. 8 is a schematic diagram of a cost matrix in a DTW algorithm according to an embodiment of the present invention.
  • FIG. 9 is a structural block diagram of a wearable device according to an embodiment of the present invention.
  • the design concept of the present invention is to solve the problem that the conventional smart sports watch in the prior art cannot monitor the swimming state of the user during swimming.
  • the invention provides a scheme for implementing swimming state monitoring based on a wearable device, and collects standard four kinds of accelerometer data by using a Micro-Electro-Mechanical System (MEMS) sensor built in a wearable device.
  • MEMS Micro-Electro-Mechanical System
  • the preset template after the user opens the swimming mode set in the wearable device, during the swimming process, the user can perform monitoring and recognition of the swimming state without any operation, thereby facilitating the user to understand the state of the swimming movement and improving the user.
  • MEMS Micro-Electro-Mechanical System
  • a method for monitoring a swimming state by using a wearable device includes:
  • Step S11 setting a swimming mode in the wearable device, and storing pre-acquisition standard swimming motion data as corresponding template data.
  • the swimming mode is started according to an instruction issued by a user who is about to enter the water, after the swimming mode is started.
  • the control sensor collects the user's swimming motion data; here, the wearable device generally has a built-in three-axis acceleration sensor, can interact with the user through the screen, and is wearable.
  • the wearable device is specifically, for example, a smart watch.
  • Step S12 obtaining test data for identifying the swimming state of the user from the swimming action data
  • step S13 the test data is matched with each template data.
  • the swimming state of the user is identified as the swimming state corresponding to the template data associated with the test data.
  • the pre-acquisition of the standard swimming motion data in step S11 as the corresponding template data includes: pre-collecting the standard swimming posture data, generating template data from the collected standard swimming posture data, and storing the template data in the wearable device, the standard swimming
  • the pose data includes at least breaststroke stroke data and freestyle stroke data. Butterfly stroke data and backstroke stroke data.
  • the swimming state corresponding to the template data associated with the test data in step S13 includes: identifying the swimming posture of the user as the breaststroke swimming stroke, the freestyle swimming posture, the butterfly swimming posture or the template data associated with the test data. Backstroke.
  • the method for monitoring the swimming state by using the wearable device of the embodiment can automatically recognize the swimming state of the user (such as a swimming posture), and does not need any after the user switches the smart watch to the swimming mode.
  • FIG. 2 is a flowchart of a method for monitoring a swimming state by using a wearable device according to an embodiment of the present invention.
  • the method for monitoring a swimming state by using a wearable device according to the embodiment includes:
  • Step S21 collecting an accelerometer data of swimming
  • an accelerometer ie, an acceleration sensor in the MEMS sensor is used to collect the three-axis acceleration data of the swimmer, so as to facilitate subsequent matching according to the collected test data and the preset template data. Determine which type of stroke the athlete uses.
  • the swimming accelerometer data can be saved in the cache, and the data in the cache is processed after the end. It should be noted that the glimpse here refers to the length of each swimming pool in swimming.
  • Step S22 intercepting data of one of the stroke periods
  • step S21 Based on the acceleration data collected in step S21, data of a stroke cycle is intercepted. Usually when swimming in the pool, the swimmer will not change the swimming position during the swimming process. Therefore, it is only necessary to collect data of one stroke cycle in each swimming to realize the swimming stroke during this swimming. .
  • the swimming motion data of the stroke cycle is intercepted from the swimming movement data during the swimming process, and the test data includes: obtaining the total number of stroke periods in the swimming process of the user, and selecting the initial One stroke cycle and one stroke week beyond the last stroke cycle
  • the swimming action data of the period is used as test data; or, the number of total stroke cycles N during the swimming process is obtained by the user, and when N is even, the swimming stroke corresponding to the stroke period of N/2 or N/2+1 position is selected.
  • the motion data is used as test data. When N is an odd number, the swimming motion data corresponding to the stroke period at the (N+1)/2 position is selected as the test data.
  • the stroke period refers to the time taken by the athlete to complete a standard stroke operation while swimming.
  • a standard stroke action of breaststroke is: moving to the side-down-rear-inner-front direction for licking water.
  • the time spent by the athlete in completing such a stroke is the stroke cycle.
  • the first method is to first obtain the total number of stroke periods in the swimming process of the user, for example, to find a swim in the cache.
  • the acceleration data obtained the total number of stroke cycles during this swimming session was 12.
  • selection can be made in any of the stroke periods other than the first stroke cycle and the last stroke cycle as the test data to be identified.
  • the 5th stroke cycle is selected as the test data in the 2nd to 11th stroke cycles. It should be noted that the reason why the first stroke cycle and the last stroke cycle are excluded and the two stroke cycles are selected for identification is to improve the accuracy of recognition, because in each swimming The action at the beginning and the end is relatively large, which is not conducive to recognition.
  • the second method is to obtain the total number of stroke periods N in the swimming process of the user.
  • N is an even number
  • the swimming motion data corresponding to the stroke period corresponding to the N/2 or N/2+1 position is selected as the test data.
  • N is an odd number
  • the swimming motion data corresponding to the stroke period at the (N+1)/2 position is selected as the test data.
  • swimming motion data of N/2 ie, the sixth stroke period
  • N/2+1 the seventh stroke period
  • FIG. 3 is a waveform diagram of a three-axis accelerometer data of a breaststroke in one embodiment of the present invention
  • FIG. 4 is a waveform diagram of an accelerometer data of one stroke of a breaststroke in an embodiment of the present invention, which is combined with FIG. 3 and FIG. 4 to specify how to intercept a swimming stroke data of a stroke cycle.
  • the data between the two end points is taken as a stroke of the swimming stroke with the adjacent two waveform peaks as the end points.
  • Action data In practical application, the acceleration data appears periodically on the acquired acceleration data waveform diagram, but the time at which the cycle starts and ends is difficult to determine. Therefore, for the convenience of calculation, the adjacent two waveforms are taken on the acceleration data waveform diagram. The peak is the endpoint and the data between the two endpoints is intercepted as a stroke period.
  • Figure 3 shows the acceleration counts saved by a breaststroke in a 25-meter pool. According to the waveform, 12 strokes were made in this breaststroke.
  • FIG. 4 is a waveform diagram of an accelerometer data waveform of a stroke period taken from the acceleration data shown in FIG.
  • the X-axis in the three-axis acceleration data is selected in the embodiment (the X-axis is when the user normally wears the smart sports watch to the user's wrist, facing the user's finger.
  • the direction of the data of one stroke cycle as test data it can be understood that in other embodiments of the present invention, Y-axis or Z-axis data can also be selected, or three-axis data can also be selected as test data. No restrictions.
  • Step S23 performing smoothing filtering processing on the data
  • the method further includes: smoothing the original waveform. Specifically, the method in this embodiment performs smoothing filtering processing on the intercepted swimming motion data of a stroke period by using K-time neighboring mean filtering to filter out noise interference.
  • the K-time nearest neighbor mean filtering algorithm is a prior art.
  • the K-time neighbor neighboring mean data processing method is to set the K-orders of any point in each axis acceleration time series by setting the number K of the nearest neighbors in advance.
  • the mean of the composed sequence is taken as the value of this point after preprocessing.
  • special processing is performed, and as many neighbor data points as possible are taken as the objects of the equalization processing.
  • Step S24 data normalization processing
  • the data of the filtered swimming motion data needs to be normalized, and the filtered swimming motion data is normalized to Within the range of ⁇ 10.
  • the acceleration data during swimming is mainly concentrated in the range of plus or minus gravity acceleration. Therefore, in this embodiment, the filtered accelerometer data is normalized to a range of ⁇ 10.
  • Step S25 using a DTW algorithm to match the preset template data
  • the current common method is to combine the three-axis data.
  • the combination of the main methods are: summation, first-order norm and second-order norm.
  • the use of accelerometer data for stroke recognition requires that the difference in accelerometer data for different strokes should be as large as possible.
  • FIG. 5 is a waveform diagram of accelerometer data for intercepting one of the stroke periods in one of the embodiments of the present invention
  • FIG. 6 is a three-axis accelerometer corresponding to freestyle and breaststroke respectively according to an embodiment of the present invention
  • the summation of the data, the first-order norm, and the second-order norm waveform are shown in Fig. 6.
  • the left column represents freestyle and the right column represents breaststroke.
  • the first line represents the summation of the axis accelerometer data
  • the second line represents the first-order norm of the axis accelerometer data
  • the third line represents the second-order norm of the axis accelerometer data.
  • the method of calculating and comparing the three-axis acceleration values is adopted.
  • the data of each axis on the three axes is matched with the corresponding axial data in the template data, and the matching result of the test data of the group is obtained according to the matching results of the three axes. That is, using the Dynamic Time Warping (DTW) algorithm, the DTW minimum distance of each axial data of the normalized set of test data and the axial data of each template data is respectively calculated. The shortest distances of the DTWs in each axial direction are summed, and the template data with the smallest summed value is used as the template data associated with the set of test data.
  • DTW Dynamic Time Warping
  • the acquired three-axis data is separately analyzed from the three-axis data corresponding to a certain stroke in the template, and the shortest distance of the DTW in each axial direction is calculated, and the shortest distances of the three axial DTWs are summed.
  • the swimming posture with the highest matching degree with the template is the swimming posture adopted by the user.
  • the normalized data and the preset stroke template data are matched and calculated by the DTW algorithm.
  • the algorithm is based on the dynamic programming (DP) idea, and the time series is extended and shortened to calculate.
  • the similarity between the two time series effectively solves the template matching problem when the data duration is inconsistent.
  • the upper and lower two solid lines are two time series, the dashed line between them represents the similarity between two time series, and the DTW uses the sum of the distances between all these similar points, called the rectification path distance. (Warp Path Distance), the rounding path distance is used to measure two times The similarity between the sequences.
  • the two time series for calculating the similarity be X and Y, and the lengths are
  • the form of w k is (i, j), where i represents the i coordinate in X and j represents the j coordinate in Y.
  • Figure 8 shows the cost matrix D in a DTW algorithm and the shortest distance normalization path between two time series.
  • the solid line represents the normalization path.
  • Step S26 outputting a result
  • the basic data of the reaction swimming efficiency can be further calculated: the time of each use, the time of each stroke, the number of strokes per stroke, and the stroke position.
  • Other exercise data can be converted from these four data.
  • the identification and calculation results are output to the user, so that the user can understand the swimming state and help the user to adjust the swimming posture, thereby improving the user experience when using the smart watch.
  • the accelerometer data of the standard four strokes is collected by the accelerometer in the smart watch as a preset template, and after the user opens the swimming mode set in the wearable device, one of the stroke cycles is intercepted at the end of each swimming session.
  • the accelerometer data is matched with the pre-stored stroke template parameters by using the DTW dynamic time registration algorithm to identify the swimming stroke used by the user.
  • the user can be monitored and recognized without any operation, and the user experience is optimized, and the competitiveness of the smart watch is improved.
  • a wearable device 90 wherein the wearable device 90 is provided with a sensor processor and a memory, and further, according to the actual function of the wearable device, other hardware may be included. I won't go into details here.
  • Memory Stores machine executable instruction code. For example, an instruction code indicating a swimming mode
  • the processor is configured to communicate with the memory, read and execute the machine executable instruction code stored in the memory, and implement the steps of the method for monitoring the swimming state by using the wearable device disclosed in the first embodiment or the second embodiment of the present application.
  • the memory may be any electronic, magnetic, optical or other physical storage device, and may contain or store information such as executable instructions, data, and the like.
  • the machine-readable storage medium may be: RAM (Radom Access Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as a hard disk drive), solid state drive, any type of storage disk. (such as a disc, dvd, etc.), or a similar storage medium, or a combination thereof.
  • the wearable device 90 includes:
  • the template storage unit 902 is configured to store pre-acquisition standard swimming motion data as corresponding template data.
  • the test data collection unit 901 is configured to start a swimming mode according to an instruction issued by a user who is about to enter the water when the monitoring process starts, and the control sensor collects the swimming motion data of the user, and obtains a test for identifying the swimming state of the user from the swimming motion data. data,
  • the swimming state identification unit 903 is configured to match the test data with each template data. When there is successfully matched test data, the swimming state of the user is identified as the swimming state corresponding to the template data associated with the test data.
  • the template storage unit 902 is specifically configured to pre-collect standard swimming posture data, generate template data from the collected standard swimming posture data, and store the template data in the wearable device, and the standard swimming posture data includes at least the breaststroke.
  • swimming data, freestyle stroke data, butterfly stroke data, and backstroke stroke data are specifically configured to pre-collect standard swimming posture data, generate template data from the collected standard swimming posture data, and store the template data in the wearable device, and the standard swimming posture data includes at least the breaststroke.
  • the swimming state identification unit 903 is configured to identify the swimming stroke of the user as the breaststroke swimming stroke, the freestyle swimming posture, the butterfly swimming posture or the back swimming swimming posture corresponding to the template data associated with the test data;
  • the sensor is a three-axis acceleration sensor, and the test data acquisition unit 901 is specifically configured to control the triaxial acceleration sensor to collect the three-axis acceleration data of the user swimming action, and obtain test data;
  • the swimming state identification unit 903 is specifically configured to match the data of each axis on the three axes with the data of the corresponding axial direction in the template data for a set of test data, and obtain the group according to the matching results of the three axes. The matching result of the test data.
  • the test data collection unit 901 is specifically configured to intercept swimming motion data of a stroke period from the swimming motion data during the swimming process as a test data.
  • the test data collecting unit 901 is configured to obtain the total number of stroke cycles in the swimming process of the user, and select swimming motion data of any one of the stroke periods except the first one of the stroke periods and the last one of the stroke periods. As the test data; or, obtain the total number of stroke cycles N in the swimming process of the user, and when N is an even number, select the swimming motion data corresponding to the stroke period of the N/2 or N/2+1 position as the test data. When N is an odd number, the swimming motion data corresponding to the stroke period at the (N+1)/2 position is selected as the test data.
  • the test data acquisition unit 901 intercepts the data between the two endpoints as a stroke cycle in the axial data corresponding to the swimming forward direction, with the adjacent two waveform peaks as the end points. swimming action data.
  • the wearable device 90 further includes: a data processing unit, configured to perform smoothing filtering processing on the intercepted swimming motion data of one of the stroke periods by using K-time neighboring mean filtering to filter out noise interference; And normalizing the data of the filtered swimming motion data, and normalizing the filtered swimming motion data to a range of ⁇ 10.
  • the swimming state identification unit 903 is specifically configured to use a dynamic time-collimated DTW to calculate each axial data of the normalized set of test data respectively corresponding to each template data in the axial direction.
  • the shortest distance of the DTW of the data, the shortest distance of each axial DTW is summed, and the template data with the smallest sum obtained is used as the template data associated with the test data of the group.
  • the wearable device of the present embodiment can be used in the foregoing method for monitoring the swimming state by using the wearable device. Therefore, for the content that is not described in the working process of the wearable device in this embodiment, refer to the foregoing embodiment. The specific description will not be repeated here.
  • the method for monitoring a swimming state by using a wearable device by setting a swimming mode in the wearable device, and storing the pre-acquisition standard swimming motion data as corresponding template data, when a monitoring process is performed.
  • the swimming mode is started according to the instruction issued by the user who is about to enter the water.
  • the user's swimming motion data is collected to obtain test data, and the test data is matched with each template data to identify the swimming state of the user.
  • the present invention can more accurately recognize the swimming posture used by the user during swimming, and helps the user to better grasp his swimming state.
  • the user only needs to switch the smart watch to the swimming mode before swimming in the water, and the recording and analysis of the swimming state data can be completed without any operation during the swimming process, and the use is simpler and easier. Prevents underwater button operation from damaging the waterproof performance of the watch.

Abstract

一种利用可穿戴设备监测游泳状态的方法及可穿戴设备，方法包括：在可穿戴设备中设置游泳模式，并存储预先采集标准游泳动作数据作为对应的模板数据，当一次监测过程开始时根据即将入水的用户发出的指令启动游泳模式，游泳模式启动后，控制传感器采集用户的游泳动作数据（S11）；从游泳动作数据中得到用于识别用户游泳状态的测试数据（S12），将测试数据与每个模板数据进行匹配，当存在匹配成功的测试数据时，识别用户的游泳状态为测试数据关联的模板数据对应的游泳状态（S13）。用户佩戴可穿戴设备游泳时，只需将可穿戴设备切换到游泳模式后，无需任何操作即可自动识别用户的泳姿，进而帮助用户了解游泳运动信息，优化了用户体验。

Description

利用可穿戴设备监测游泳状态的方法及可穿戴设备 技术领域
本发明涉及可穿戴设备技术领域，具体涉及一种利用可穿戴设备监测游泳状态的方法及可穿戴设备。
背景技术
随着社会的进步，人们的运动时间越来越少，久坐的工作方式使许多人身体素质逐渐走下坡，肥胖和各种慢性病等严重影响了人们的生活质量。为了自身的健康，人们对运动也越来越重视，游泳作为一种非常有助于提高体能的有运动形式，成为了人们最喜爱的有氧运动形式之一，游泳可以改善心血管系统、提高肺活量、改善肌肉系统能力、改善体温调节能力。游泳也是一项需要全身参与的运动，能够提高人体肌肉的力量和协调性，特别是躯干、肩带和上肢的肌肉。因为在水中游泳需要克服较大的阻力，游泳又是周期性的运动，长期锻炼能够使肌肉力量、速度、耐力和关节的灵活性都得到提高。在人们越来越重视健康和运动的今天，游泳也成为大众热捧的有氧运动形式。
伴随着运动的热潮，智能运动手表也成为数码产品新的宠儿。运动状态识别作为解决运动监测和运动状态提醒的技术基础，是智能运动手表算法的核心，也是难点之一。目前在智能运动手表领域，还没有行之有效的对游泳运动状态进行监测和识别的技术方案，不能满足用户的使用需求，导致智能运动手表的体验不佳。
发明内容
本发明提供了一种利用可穿戴设备监测游泳状态的方法及可穿戴设备，以解决现有在智能运动手表领域，没有行之有效的对游泳运动状态进行监测和识别的技术方案，导致智能运动手表的用户体验不佳的问题。
根据本发明的一个方面，提供了一种利用可穿戴设备监测游泳状态的方法，该方法包括：在可穿戴设备中设置游泳模式，并存储预先采集标准游泳动作数据作为对应的模板数据，
当一次监测过程开始时，根据即将入水的用户发出的指令启动游泳模式，在游泳模式启动后，控制传感器采集用户的游泳动作数据；
从游泳动作数据中得到用于识别用户游泳状态的测试数据，
将测试数据与每个模板数据进行匹配，当存在匹配成功的测试数据时，识别用户的游泳状态为该测试数据关联的模板数据对应的游泳状态。
根据本发明的另一个方面，提供了一种可穿戴设备，可穿戴设备中设置有传感器和游泳模式，包括：
模板存储单元，用于存储预先采集标准游泳动作数据作为对应的模板数据，
测试数据采集单元，用于当一次监测过程开始时，根据即将入水的用户发出的指令启动游泳模式，控制传感器采集用户的游泳动作数据，从游泳动作数据中得到用于识别用户游泳状态的测试数据，
游泳状态识别单元，用于将测试数据与每个模板数据进行匹配，当存在匹配成功的测试数据时，识别用户的游泳状态为该测试数据关联的模板数据对应的游泳状态。
本发明的有益效果是：本发明的技术方案利用可穿戴设备监测游泳状态，通过在可穿戴设备中设置游泳模式，并存储预先采集标准游泳动作数据作为对应的模板数据，当一次监测过程开始时，根据即将入水的用户发出的指令启动游泳模式，在游泳模式启动后，采集用户的游泳动作数据得到测试数据，将测试数据与每个模板数据进行匹配，以识别出用户的游泳状态。与传统运动手表只能简单记录速度等信息来估算用户的热量消耗等信息相比，本发明能够准确识别用户一趟游泳过程中使用的泳姿，有助于用户更好的掌握自己的游泳运动状态，对于改善泳姿也有很大的帮助。并且，采用本发明的技术方案，用户只需在下水游泳之前，将可穿戴设备切换到游泳模式，在游泳过程中无需任何操作，即可完成对游泳状态数据的记录和分析，使用更加简便，也防止了水下按键操作破坏手表防水性能的发生。
附图说明
图1是本发明一个实施例的一种利用可穿戴设备监测游泳状态的方法的流程图；
图2是本发明另一个实施例的一种利用可穿戴设备监测游泳状态的方法的流程图；
图3是本发明一个实施例的一趟蛙泳的三轴加速度计数据波形图；
图4是本发明一个实施例的截取蛙泳其中一个划水周期的加速度计数据波形 图；
图5是本发明一个实施例的截取自由泳其中一个划水周期的加速度计数据波形图；
图6是本发明一个实施例的分别为自由泳和蛙泳时3轴加速度计数据的求和、一阶范数、二阶范数波形图。
图7是本发明一个实施例的两个时间序列的归整示意图；
图8是本发明一个实施例的DTW算法中的一个代价矩阵的示意图；
图9是本发明一个实施例的一种可穿戴设备的结构框图。
具体实施方式
本发明的设计构思在于：针对现有技术中普通智能运动手表不能监测用户游泳运动时的游泳状态的问题。本发明提出一种基于可穿戴设备实现游泳运动状态监测的方案，通过可穿戴设备内置的微机电系统(Micro-Electro-Mechanical System，简称MEMS)传感器采集标准的四种泳姿的加速度计数据作为预设模板，在用户开启可穿戴设备中设置的游泳模式后，游泳过程中，无需用户进行任何操作即可实现对用户游泳状态的监测识别，从而方便用户了解本次游泳运动的状态，提高用户体验。
实施例一
图1是本发明一个实施例的一种利用可穿戴设备监测游泳状态的方法的流程图，参见图1，本实施例的利用可穿戴设备监测游泳状态的方法包括：
步骤S11，在可穿戴设备中设置游泳模式，并存储预先采集标准游泳动作数据作为对应的模板数据，当一次监测过程开始时，根据即将入水的用户发出的指令启动游泳模式，在游泳模式启动后，控制传感器采集用户的游泳动作数据；这里的，可穿戴设备一般具有：内置三轴加速度传感器、能够与用户通过屏幕进行交互，且可穿戴的特性。可穿戴设备具体例如：智能手表。
步骤S12，从游泳动作数据中得到用于识别用户游泳状态的测试数据，
步骤S13，将测试数据与每个模板数据进行匹配，当存在匹配成功的测试数据时，识别用户的游泳状态为该测试数据关联的模板数据对应的游泳状态。
需要说明的是，步骤S11中的预先采集标准游泳动作数据作为对应的模板数据包括：预先采集标准泳姿数据，由采集到的标准泳姿数据生成模板数据并存储在可穿戴设备中，标准泳姿数据至少包括蛙泳泳姿数据、自由泳泳姿数据， 蝶泳泳姿数据以及仰泳泳姿数据。
步骤S13中的识别用户的游泳状态为该测试数据关联的模板数据对应的游泳状态包括：识别用户的泳姿为该测试数据关联的模板数据对应的蛙泳泳姿、自由泳泳姿，蝶泳泳姿或仰泳泳姿。
由图1所示的方法可知，本实施例的这种利用可穿戴设备监测游泳状态的方法，能够自动识别用户游泳状态(如泳姿)，在用户将智能手表切换到游泳模式后，无需任何操作即可自动识别泳姿，根据识别出的泳姿也能够帮助用户获取更详细的运动信息，例如，本次游泳的休息用时、游泳距离、划水数、SWOLF(SWOLF指数＝单趟计时+单趟划水次数，SWOLF数值越小，说明游泳的效率越高)等数据，有助于用户更精确地掌握自己的游泳运动状态和改善游泳效率。
实施例二
图2是本发明一个实施例的一种利用可穿戴设备监测游泳状态的方法的流程图，参见图2，本实施例的这种利用可穿戴设备监测游泳状态的方法包括：
流程开始；
步骤S21，采集一趟游泳的加速度计数据；
本实施例中，首先，采用MEMS传感器中的加速度计(即，加速度传感器)采集运动者一趟游泳的三轴加速度数据，以方便后续根据采集的测试数据与预设的模板数据的匹配结果来判定运动者采用的是哪种泳姿。
采集到用户游泳时的三轴加速度计数据后，可以将一趟游泳的加速度计数据保存在缓存中，一趟结束后对缓存中的数据进行处理。需要说明的是，这里的一趟，是指游泳中每游完一个泳池的长度。
步骤S22，截取其中一个划水周期的数据；
在步骤S21中采集到的加速度数据的基础上，截取一个划水周期的数据。通常在泳池游泳时，游泳者不会在一趟游泳过程中更换泳姿，因此，每趟游泳中只要采集其中一个划水周期的数据即可实现对这一趟游泳过程中的泳姿进行识别。
本实施例中，从用户一趟游泳过程中的游泳动作数据中截取一个划水周期的游泳动作数据，作为测试数据，包括：获取用户一趟游泳过程中的总划水周期数，选取除最初的一个划水周期和最末的一个划水周期之外的任一个划水周 期的游泳动作数据作为测试数据；或者，获取用户一趟游泳过程中的总划水周期数N，当N为偶数时，选取N/2或N/2+1位置对应的划水周期的游泳动作数据作为测试数据，当N为奇数时，选取(N+1)/2位置上对应的划水周期的游泳动作数据作为测试数据。
需要说明的是，划水周期是指运动者在游泳时，完成一个标准划水动作所花费的时间。以蛙泳为例，蛙泳的一个标准划水动作是：向侧-下-后-内-前方向移动进行蹬夹水。运动者在完成这样的划水动作时所花费时间即为划水周期。
本实施例中，在选择一个划水周期的加速度数据时，提供了两种实现方式，方式一是先获取用户一趟游泳过程中的总划水周期数，例如，查找缓存中一趟游泳的加速度数据得到这一趟游泳过程中总的划水周期数为12。然后，可以在除最初的一个划水周期和最末的一个划水周期之外的任一个划水周期中进行选择，作为待识别的测试数据。例如，在第2个至第11个划水周期中选择第5个划水周期作为测试数据。需要说明的是，之所以将最初的一个划水周期和最末的一个划水周期排除在外而不选择这两个划水周期进行识别，是为了提高识别的准确度，因为在每趟游泳中，最开始和快结束时的动作相对变形较大，不利于识别。
方式二是，获取用户一趟游泳过程中的总划水周期数N，当N为偶数时，选取N/2或N/2+1位置对应的划水周期的游泳动作数据作为测试数据，当N为奇数时，选取(N+1)/2位置上对应的划水周期的游泳动作数据作为测试数据。
举例而言，总划水周期N为12，那么可以选择N/2(即第6个划水周期)或N/2+1(第7个划水周期)的游泳动作数据作为测试数据。
图3是本发明一个实施例的一趟蛙泳的三轴加速度计数据波形图，图4是本发明一个实施例的截取蛙泳其中一个划水周期的加速度计数据波形图，下面结合图3和图4来具体说明如何截取一个划水周期的游泳动作数据。
本实施例中，在游泳前进方向对应的轴向数据中(例如X轴上的数据)，以相邻的两个波形峰值为端点，截取两个端点之间的数据作为一个划水周期的游泳动作数据。实际应用时，在采集到的加速度数据波形图上，加速度数据呈现周期性，但是周期开始和结束的时间难以确定，故，为计算方便，才取加速度数据波形图上以相邻的两个波形峰值为端点，截取两个端点之间的数据作为一个划水周期。以蛙泳为例，图3为25米的泳池中一趟蛙泳所保存的加速度计数 据波形示意，在这趟蛙泳中共进行了12次划水。而在每趟游泳中，最开始和快结束时的动作相对变形较大，不利于识别，从图3中也可以明显看出，中间的动作较为规范，因此，截取数据选择为总划水周期数中间的一个划水周期。图4即为从图3中所示的加速度数据中截取到的一个划水周期的加速度计数据波形示意图。
需要说明的是，为了简化数据计算量、提高计算效率，本实施例中只选择了三轴加速度数据中X轴向(X轴向是用户正常佩戴智能运动手表到用户腕部时，朝向用户手指的方向)一个划水周期的数据作为测试数据，可以理解，在本发明的其他实施例中，也可以选择Y轴或Z轴数据，或者也可以同时选择三轴的数据作为测试数据，对此不作限制。
步骤S23，对数据进行平滑滤波处理；
在本实施例中，为了避免噪声的影响，提高泳姿识别的精度。方法还包括：对原始波形进行平滑滤波处理。具体的，本实施例的方法是利用K时间近邻均值滤波对截取的一个划水周期的游泳动作数据进行平滑滤波处理，以滤除噪声干扰。
K时间近邻均值滤波算法是现有技术，K时间近邻均值化数据处理的方法就是通过事先设定时间最近邻的个数K，在各轴加速度时间序列中，把任意一点左右各K个元素所组成的序列的均值作为预处理后该点的值。而对于时间序列中前K个数据点和最后K个数据点，做特殊处理，取尽可能多的邻居数据点作为均值化处理的对象。
步骤S24，数据归一化处理；
为了避免不同用户的速度、力量等差别带来的影响以及为数据处理方便，本实施例中需要对滤波后的游泳动作数据进行数据归一化处理，将滤波后的游泳动作数据归一化到±10的范围内。
经过试验和统计分析发现，游泳时的加速度数据主要集中在正负一倍重力加速度范围内，因此，本实施例中将滤波后的加速度计数据归一化到±10的范围内。
步骤S25，采用DTW算法与预设模板数据进行匹配；
对于三轴加速度计数据的处理，目前常用的办法是将三轴数据合一后进行 分析，合一的办法主要有：求和，一阶范数和二阶范数。而采用加速度计数据进行泳姿识别则需要不同泳姿的加速度计数据差异度越大越好。
以自由泳和蛙泳为例，图5是本发明一个实施例的截取自由泳其中一个划水周期的加速度计数据波形图；图6是本发明一个实施例的分别为自由泳和蛙泳对应的三轴加速度计数据的求和、一阶范数、二阶范数波形图，图6为自由泳和蛙泳3轴加速度计的求和、一阶范数、二阶范数的对比。在图6中，左列表示自由泳，右列表示蛙泳。第一行表示轴加速度计数据的求和，第二行表示轴加速度计数据的一阶范数，第三行表示轴加速度计数据的二阶范数。可以看出，对加速度计数据求二阶范数后，其数值变化范围只在0.02左右，无法进行泳姿匹配计算和划水周期识别。同样，一阶范数也存在这一问题，数值变化范围较小，局部峰值过多。而求和计算虽然相对而言保存了比较完整的加速度值变化范围，但其两种泳姿数据的差异度并不如三轴加速度数据分别对比那样丰富。
因此，本实施例中采用采用对三轴加速度值分别进行计算、对比的办法。对一组测试数据，将三轴上每一轴向的数据，分别与模板数据中对应轴向的数据进行匹配，根据三个轴向的匹配结果得到该组测试数据的匹配结果。即，利用动态时间归准DTW(Dynamic Time Warping，简称DTW)算法，分别计算归一化后的一组测试数据的每个轴向数据与每个模板数据对应轴向的数据的DTW最短距离，将各个轴向的DTW最短距离进行求和运算，将求和所得值最小的模板数据作为该组测试数据关联的模板数据。具体的，将获取的三轴数据与模板中某种泳姿对应的三轴数据分别进行分析，计算得到各个轴向的DTW最短距离，并将三个轴向的DTW最短距离进行求和。求和值越小，则匹配度越高，在完成与四种泳姿模板的匹配后，找到与模板匹配度最高的泳姿即为用户采用的泳姿。
具体实现时，是将归一化后的数据与预设的泳姿模板数据采用DTW算法进行匹配计算，该算法基于动态规划(Dynamic Programming，DP)思想，通过把时间序列进行延伸和缩短来计算两个时间序列之间的相似性，有效地解决了数据时长不一致时的模板匹配问题。如图7所示，上下两条实线为两个时间序列，之间的虚线表示两个时间序列间的相似点，DTW使用所有这些相似点之间的距离的和，称为归整路径距离(Warp Path Distance)，归整路径距离用来衡量两个时 间序列之间的相似性。
DTW算法具体过程如下
令要计算相似度的两个时间序列为X和Y，长度分别为|X|和|Y|。归整路径(Warp Path)的形式为W＝w1,w2,...wk，其中max(|X|,|Y|)≤K≤|X|+|Y|。wk的形式为(i,j),其中i表示的是X中的i坐标，j表示的是Y中的j坐标。归整路径W必须从w1＝(1,1)开始，到wk＝(|X|,|Y|)结束，以保证X和Y中的每个坐标都在W中出现。并且i和j必须是单调增加的，以保证图7中的虚线不会相交，最后得到的归整路径是距离最短的一个归整路径
D(i,j)＝Dist(i,j)+min[D(i-1,j),D(i,j-1),D(i-1,j-1)]
图8展示了一个DTW算法中的代价矩阵D和两个时间序列之间的最短距离归整路径，在图8中实线表示归整路径。通过DTW算法，将归一化之后的一个划水周期的加速度计数据波形与预设模板进行匹配计算，得到的最短路径值对应的泳姿确定为用户所采用的泳姿。
需要说明的是，DTW算法为现有技术，本实施例中没有描述的算法细节可以参见现有技术中的说明，这里不再赘述。
步骤S26，输出结果；
根据DTW算法计算用户游泳所使用的泳姿后，可以进一步计算得到反应游泳运动效率的基础数据：每趟用时、每趟划水用时、每趟划水数、泳姿。其他运动数据(包括SWOLF值，划水速率等)均可通过这四个数据换算得出。将识别和计算得到的结果输出给用户，从而方便用户了解自己的游泳状态，并可以帮助用户调整泳姿，提高了用户使用智能手表时的体验。
流程结束。
至此，通过智能手表中的加速度计采集标准的四种泳姿的加速度计数据作为预设模板，在用户开启可穿戴设备中设置的游泳模式后，每趟游泳结束时，截取其中一次划水周期的加速度计数据作为测试数据，采用DTW动态时间归准算法，与预存的泳姿模板参数进行匹配，从而识别得到用户这一趟游泳所使用的泳姿。游泳过程中，无需用户进行任何操作即可实现对用户游泳状态的监测识别，也优化了用户体验，提高了智能手表的竞争力。
实施例三
在本发明的再一个实施例中，还提供了一种可穿戴设备90，可穿戴设备90中设置有传感器处理器和存储器，此外，根据该可穿戴设备的实际功能，还可以包括其他硬件，对此不再赘述。
存储器：存储机器可执行指令代码。例如，指示游泳模式的指令代码；
处理器：与存储器通信，读取和执行存储器中存储的机器可执行指令代码，实现本申请上述实施例一或实施例二公开的利用可穿戴设备监测游泳状态的方法的步骤。
这里，存储器可以是任何电子、磁性、光学或其它物理存储装置，可以包含或存储信息，如可执行指令、数据，等等。例如，机器可读存储介质可以是：RAM(Radom Access Memory，随机存取存储器)、易失存储器、非易失性存储器、闪存、存储驱动器(如硬盘驱动器)、固态硬盘、任何类型的存储盘(如光盘、dvd等)，或者类似的存储介质，或者它们的组合。
从功能上划分，可穿戴设备90包括：
模板存储单元902，用于存储预先采集标准游泳动作数据作为对应的模板数据，
测试数据采集单元901，用于当一次监测过程开始时，根据即将入水的用户发出的指令启动游泳模式，控制传感器采集用户的游泳动作数据，从游泳动作数据中得到用于识别用户游泳状态的测试数据，
游泳状态识别单元903，用于将测试数据与每个模板数据进行匹配，当存在匹配成功的测试数据时，识别用户的游泳状态为该测试数据关联的模板数据对应的游泳状态。
在本发明的一个实施例中，模板存储单元902，具体用于预先采集标准泳姿数据，由采集到的标准泳姿数据生成模板数据并存储在可穿戴设备中，标准泳姿数据至少包括蛙泳泳姿数据、自由泳泳姿数据，蝶泳泳姿数据以及仰泳泳姿数据，
游泳状态识别单元903，用于识别用户的泳姿为该测试数据关联的模板数据对应的蛙泳泳姿、自由泳泳姿，蝶泳泳姿或仰泳泳姿；
传感器为三轴加速度传感器，测试数据采集单元901，具体用于控制三轴加速度传感器采集用户游泳动作的三轴加速度数据，得到测试数据；
游泳状态识别单元903，具体用于对一组测试数据，将三轴上每一轴向的数据，分别与模板数据中对应轴向的数据进行匹配，根据三个轴向的匹配结果得到该组测试数据的匹配结果。
在本发明的一个实施例中，测试数据采集单元901，具体用于从用户一趟游泳过程中的游泳动作数据中截取一个划水周期的游泳动作数据作为测试数据。
测试数据采集单元901，用于获取用户一趟游泳过程中的总划水周期数，选取除最初的一个划水周期和最末的一个划水周期之外的任一个划水周期的游泳动作数据作为测试数据；或者，获取用户一趟游泳过程中的总划水周期数N，当N为偶数时，选取N/2或N/2+1位置对应的划水周期的游泳动作数据作为测试数据，当N为奇数时，选取(N+1)/2位置上对应的划水周期的游泳动作数据作为测试数据。
在本发明的一个实施例中，测试数据采集单元901，在游泳前进方向对应的轴向数据中，以相邻的两个波形峰值为端点，截取两个端点之间的数据作为一个划水周期的游泳动作数据。
在本发明的一个实施例中，可穿戴设备90还包括：数据处理单元，用于利用K时间近邻均值滤波对截取的一个划水周期的游泳动作数据进行平滑滤波处理，以滤除噪声干扰；以及，对滤波后的游泳动作数据进行数据归一化处理，将滤波后的游泳动作数据归一化到±10的范围内。
在本发明的一个实施例中游泳状态识别单元903，具体用于采用动态时间归准DTW，分别计算归一化后的一组测试数据的每个轴向数据与每个模板数据对应轴向的数据的DTW最短距离，将各个轴向的DTW最短距离进行求和运算，将求和所得值最小的模板数据作为该组测试数据关联的模板数据。
需要说明的是，本实施例的这种可穿戴设备可以用于前述利用可穿戴设备监测游泳状态的方法中，因而本实施例中对可穿戴设备的工作过程没有描述的内容可以参见前述实施例的具体说明，在此不再赘述。
综上所述，本发明技术方案的这种利用可穿戴设备监测游泳状态的方法，通过在可穿戴设备中设置游泳模式，并存储预先采集标准游泳动作数据作为对应的模板数据，当一次监测过程开始时，根据即将入水的用户发出的指令启动游泳模式，在游泳模式启动后，采集用户的游泳动作数据得到测试数据，将测试数据与每个模板数据进行匹配，以识别出用户的游泳状态。与传统运动手表 只能简单记录速度等信息来估算用户的热量消耗等信息相比，本发明能够更加准确的识别用户一趟游泳过程中使用的泳姿，有助于用户更好的掌握自己的游泳运动状态，对于改善泳姿也有很大得帮助。并且，采用本发明的技术方案，用户只需在下水游泳之前，将智能手表切换到游泳模式，在游泳过程中无需任何操作，即可完成对游泳状态数据的记录和分析，使用更加简便，也防止了水下按键操作破坏手表防水性能的发生。
以上所述仅为本发明的较佳实施例而已，并非用于限定本发明的保护范围。凡在本发明的精神和原则之内所作的任何修改、等同替换、改进等，均包含在本发明的保护范围内。

Claims (15)

  1. 一种利用可穿戴设备监测游泳状态的方法，其中，该方法包括：在可穿戴设备中设置游泳模式，并存储预先采集标准游泳动作数据作为对应的模板数据，
    当一次监测过程开始时，根据即将入水的用户发出的指令启动游泳模式，在游泳模式启动后，控制传感器采集用户的游泳动作数据；
    从所述游泳动作数据中得到用于识别用户游泳状态的测试数据，
    将所述测试数据与每个模板数据进行匹配，当存在匹配成功的测试数据时，识别用户的游泳状态为该测试数据关联的模板数据对应的游泳状态。
  2. 根据权利要求1所述的方法，其中，所述预先采集标准游泳动作数据作为对应的模板数据包括：
    预先采集标准泳姿数据，由采集到的标准泳姿数据生成模板数据并存储在可穿戴设备中，
    所述标准泳姿数据至少包括蛙泳泳姿数据、自由泳泳姿数据，蝶泳泳姿数据以及仰泳泳姿数据，
    所述识别用户的游泳状态为该测试数据关联的模板数据对应的游泳状态包括：
    识别用户的泳姿为该测试数据关联的模板数据对应的蛙泳泳姿、自由泳泳姿，蝶泳泳姿或仰泳泳姿。
  3. 根据权利要求2所述的方法，其中，所述控制传感器采集用户的游泳动作数据包括：
    控制利用三轴加速度传感器采集用户游泳动作的三轴加速度数据，得到测试数据；
    将所述测试数据与每个模板数据进行匹配包括：
    对一组测试数据，将三轴上每一轴向的数据，分别与模板数据中对应轴向的数据进行匹配，根据三个轴向的匹配结果得到该组测试数据的匹配结果。
  4. 根据权利要求3所述的方法，其中，所述控制传感器采集用户的游泳动作数据包括：
    控制传感器采集得到用户一趟游泳过程中的游泳动作数据；
    从用户一趟游泳过程中的游泳动作数据中截取一个划水周期的游泳动作数据作为测试数据。
  5. 根据权利要求4所述的方法，其中，所述从用户一趟游泳过程中的游泳动作数据中截取一个划水周期的游泳动作数据作为测试数据包括：
    获取用户一趟游泳过程中的总划水周期数，选取除最初的一个划水周期和最末的一个划水周期之外的任一个划水周期的游泳动作数据作为测试数据；
    或者，获取用户一趟游泳过程中的总划水周期数N，当N为偶数时，选取N/2或N/2+1位置对应的划水周期的游泳动作数据作为测试数据，当N为奇数时，选取(N+1)/2位置上对应的划水周期的游泳动作数据作为测试数据。
  6. 根据权利要求4所述的方法，其中，所述从用户一趟游泳过程中的游泳动作数据中截取一个划水周期的游泳动作数据作为测试数据包括：
    在游泳前进方向对应的轴向数据中，以相邻的两个波形峰值为端点，截取两个端点之间的数据作为一个划水周期的游泳动作数据。
  7. 根据权利要求4所述的方法，其中，该方法还包括：利用K时间近邻均值滤波对截取的一个划水周期的游泳动作数据进行平滑滤波处理，以滤除噪声干扰；
    以及，对滤波后的游泳动作数据进行数据归一化处理，将滤波后的游泳动作数据归一化到±10的范围内。
  8. 根据权利要求3所述的方法，其中，所述对一组测试数据，将三轴上每一轴向的数据，分别与模板数据中对应轴向的数据进行匹配，根据三个轴向的匹配结果得到该组测试数据的匹配结果包括：
    采用动态时间归准DTW，分别计算归一化后的一组测试数据的每个轴向数据与每个模板数据对应轴向的数据的DTW最短距离，将各个轴向的DTW最短距离进行求和运算，将求和所得值最小的模板数据作为该组测试数据关联的模板数据。
  9. 一种可穿戴设备，其中，所述可穿戴设备中设置有传感器和游泳模式，包括：
    模板存储单元，用于存储预先采集标准游泳动作数据作为对应的模板数据，
    测试数据采集单元，用于当一次监测过程开始时，根据即将入水的用户发出的指令启动游泳模式，控制传感器采集用户的游泳动作数据，从所述游泳动作数据中得到用于识别用户游泳状态的测试数据，
    游泳状态识别单元，用于将所述测试数据与每个模板数据进行匹配，当存在匹配成功的测试数据时，识别用户的游泳状态为该测试数据关联的模板数据对应的游泳状态。
  10. 根据权利要求9所述的可穿戴设备，其中，所述模板存储单元，具体用于预先采集标准泳姿数据，由采集到的标准泳姿数据生成模板数据并存储在可穿戴设备中，所述标准泳姿数据至少包括蛙泳泳姿数据、自由泳泳姿数据，蝶泳泳姿数据以及仰泳泳姿数据，
    所述游泳状态识别单元，用于识别用户的泳姿为该测试数据关联的模板数据对应的蛙泳泳姿、自由泳泳姿，蝶泳泳姿或仰泳泳姿；
    所述传感器为三轴加速度传感器，所述测试数据采集单元，具体用于控制三轴加速度传感器采集用户游泳动作的三轴加速度数据，得到测试数据；
    所述游泳状态识别单元，具体用于对一组测试数据，将三轴上每一轴向的数据，分别与模板数据中对应轴向的数据进行匹配，根据三个轴向的匹配结果得到该组测试数据的匹配结果。
  11. 根据权利要求9所述的可穿戴设备，其中，所述测试数据采集单元，具体用于控制传感器采集得到用户一趟游泳过程中的游泳动作数据；从用户一趟游泳过程中的游泳动作数据中截取一个划水周期的游泳动作数据作为测试数据。
  12. 根据权利要求11所述的可穿戴设备，其中，所述测试数据采集单元具体用于获取用户一趟游泳过程中的总划水周期数，选取除最初的一个划水周期和最末的一个划水周期之外的任一个划水周期的游泳动作数据作为测试数据；
    或者，获取用户一趟游泳过程中的总划水周期数N，当N为偶数时，选取N/2或N/2+1位置对应的划水周期的游泳动作数据作为测试数据，当N为奇数时，选取(N+1)/2位置上对应的划水周期的游泳动作数据作为测试数据。
  13. 根据权利要求10所述的可穿戴设备，其中，所述测试数据采集单元在游泳前进方向对应的轴向数据中，以相邻的两个波形峰值为端点，截取两个端点之间的数据作为一个划水周期的游泳动作数据。
  14. 根据权利要求11所述的可穿戴设备，其中，还包括，数据处理单元，用于利用K时间近邻均值滤波对截取的一个划水周期的游泳动作数据进行平滑滤波处理，以滤除噪声干扰；以及，对滤波后的游泳动作数据进行数据归一化处理，将滤波后的游泳动作数据归一化到±10的范围内。
  15. 根据权利要求10所述的可穿戴设备，其中，所述游泳状态识别单元，用于采用动态时间归准DTW，分别计算归一化后的一组测试数据的每个轴向数据与每个模板数据对应轴向的数据的DTW最短距离，将各个轴向的DTW最短距离进行求和运算，将求和所得值最小的模板数据作为该组测试数据关联的模板数据。
PCT/CN2017/093724 2016-08-25 2017-07-20 利用可穿戴设备监测游泳状态的方法及可穿戴设备 WO2018036316A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/327,597 US11517789B2 (en) 2016-08-25 2017-07-20 Method for monitoring swimming state by means of wearable device, and wearable device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610725326.XA CN106175781B (zh) 2016-08-25 2016-08-25 利用可穿戴设备监测游泳状态的方法及可穿戴设备
CN201610725326.X 2016-08-25

Publications (1)

Publication Number Publication Date
WO2018036316A1 true WO2018036316A1 (zh) 2018-03-01

Family

ID=57524767

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/093724 WO2018036316A1 (zh) 2016-08-25 2017-07-20 利用可穿戴设备监测游泳状态的方法及可穿戴设备

Country Status (3)

Country Link
US (1) US11517789B2 (zh)
CN (1) CN106175781B (zh)
WO (1) WO2018036316A1 (zh)

Families Citing this family (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106175781B (zh) 2016-08-25 2019-08-20 歌尔股份有限公司 利用可穿戴设备监测游泳状态的方法及可穿戴设备
US11896368B2 (en) * 2016-08-31 2024-02-13 Apple Inc. Systems and methods for determining swimming metrics
KR102252269B1 (ko) * 2016-08-31 2021-05-14 애플 인크. 수영 분석 시스템 및 방법
CN106730651A (zh) * 2017-03-27 2017-05-31 广东小天才科技有限公司 一种游泳监测模块切换的方法和装置
CN107329823B (zh) * 2017-05-16 2020-04-03 泉州味盛食品有限公司 一种用于运动检测的三轴加速度数据的柔性滤波装置
CN107469326B (zh) * 2017-07-04 2020-08-07 广东乐心医疗电子股份有限公司 一种用于可穿戴设备的游泳监测方法与装置及可穿戴设备
CN108133160B (zh) * 2017-12-21 2021-09-28 儒安物联科技集团有限公司 基于rfid的游泳安全监控系统
CN108460322A (zh) * 2017-12-28 2018-08-28 惠州市德赛工业研究院有限公司 一种泳姿识别方法及应用
CN108245869B (zh) * 2017-12-29 2020-03-06 北京顺源开华科技有限公司 一种游泳信息检测方法、装置及电子设备
CN110548273B (zh) * 2018-05-31 2021-06-22 华为技术有限公司 一种设置运动模式的方法和装置
CN108837476B (zh) * 2018-06-08 2020-09-04 歌尔科技有限公司 一种检测游泳起点的方法、装置和智能穿戴设备
CN108837477A (zh) * 2018-06-08 2018-11-20 青岛真时科技有限公司 一种检测游泳划水数的方法、装置和智能穿戴设备
CN108837480B (zh) * 2018-06-29 2020-05-01 安徽省徽腾智能交通科技有限公司 游泳监测系统及训练方法
CN109758154B (zh) * 2019-01-09 2022-02-25 北京卡路里信息技术有限公司 一种运动状态确定方法、装置、设备及存储介质
CN111803902B (zh) * 2019-04-10 2022-08-12 北京卡路里信息技术有限公司 泳姿识别方法、装置、可穿戴设备及存储介质
US11937904B2 (en) 2019-09-09 2024-03-26 Apple Inc. Detecting the end of cardio machine activities on a wearable device
CN110731762B (zh) * 2019-09-18 2022-02-08 平安科技（深圳）有限公司 基于相似度预处理脉搏波的方法、装置、计算机系统及可读存储介质
CN111265841B (zh) * 2020-02-05 2021-07-27 北京卡路里信息技术有限公司 一种游泳圈数确定方法、装置、设备及存储介质
CN112057834A (zh) * 2020-09-10 2020-12-11 青岛大学 一种基于传感器的康复动作标准评判方法
CN112274902B (zh) * 2020-10-16 2022-06-10 深圳卡路里科技有限公司 游泳数据获取方法及装置、游泳数据采集设备的控制方法
CN112587901B (zh) * 2020-11-24 2022-05-31 安徽华米健康科技有限公司 游泳姿势识别方法、装置、系统和存储介质
CN113713358B (zh) * 2021-06-28 2023-12-26 深圳市奋达智能技术有限公司 基于多传感器融合的游泳监测方法、设备、存储介质和程序产品
CN114602155B (zh) * 2022-05-11 2023-02-21 荣耀终端有限公司 游泳信息统计方法、计算机可读存储介质和电子设备

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060161053A1 (en) * 2005-01-19 2006-07-20 Polar Electro Oy System for monitoring an individual's physiological state, and a computer software product
CN101518442A (zh) * 2008-07-05 2009-09-02 杭州义盛祥通信技术有限公司 运动量化腕表及运动计量分析方法
CN101520815A (zh) * 2008-07-05 2009-09-02 杭州义盛祥通信技术有限公司 个人运动管理系统及管理方法
CN103908259A (zh) * 2014-04-03 2014-07-09 深圳市德凯瑞科技有限公司 一种智能穿戴设备及人体运动的监测与识别方法
CN104200234A (zh) * 2014-07-11 2014-12-10 杭州微纳科技有限公司 人体动作建模和识别方法
US20150164377A1 (en) * 2013-03-13 2015-06-18 Vaidhi Nathan System and method of body motion analytics recognition and alerting
CN105159441A (zh) * 2015-07-28 2015-12-16 东华大学 基于自主式运动识别技术的私人教练智能手环
CN105892674A (zh) * 2016-04-26 2016-08-24 哈尔滨工程大学 一种基于智能手环的泳姿识别方法和基于智能手环的运动方案
CN106175781A (zh) * 2016-08-25 2016-12-07 歌尔股份有限公司 利用可穿戴设备监测游泳状态的方法及可穿戴设备

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101158442A (zh) 2007-09-28 2008-04-09 唐建业 铝包竹型材
US20140099615A1 (en) * 2012-10-04 2014-04-10 Topcor, Llc Auditory feedback device and method for swimming speed
GB2529082B (en) * 2013-03-15 2016-08-17 Suunto Oy Device and method for monitoring swimming performance
CN105184325B (zh) * 2015-09-23 2021-02-23 歌尔股份有限公司 一种移动智能终端
US10339371B2 (en) 2015-09-23 2019-07-02 Goertek Inc. Method for recognizing a human motion, method for recognizing a user action and smart terminal
CN105786182B (zh) * 2016-02-26 2018-12-07 深圳还是威健康科技有限公司 一种基于手势控制周围设备的方法及装置

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060161053A1 (en) * 2005-01-19 2006-07-20 Polar Electro Oy System for monitoring an individual's physiological state, and a computer software product
CN101518442A (zh) * 2008-07-05 2009-09-02 杭州义盛祥通信技术有限公司 运动量化腕表及运动计量分析方法
CN101520815A (zh) * 2008-07-05 2009-09-02 杭州义盛祥通信技术有限公司 个人运动管理系统及管理方法
US20150164377A1 (en) * 2013-03-13 2015-06-18 Vaidhi Nathan System and method of body motion analytics recognition and alerting
CN103908259A (zh) * 2014-04-03 2014-07-09 深圳市德凯瑞科技有限公司 一种智能穿戴设备及人体运动的监测与识别方法
CN104200234A (zh) * 2014-07-11 2014-12-10 杭州微纳科技有限公司 人体动作建模和识别方法
CN105159441A (zh) * 2015-07-28 2015-12-16 东华大学 基于自主式运动识别技术的私人教练智能手环
CN105892674A (zh) * 2016-04-26 2016-08-24 哈尔滨工程大学 一种基于智能手环的泳姿识别方法和基于智能手环的运动方案
CN106175781A (zh) * 2016-08-25 2016-12-07 歌尔股份有限公司 利用可穿戴设备监测游泳状态的方法及可穿戴设备

Also Published As

Publication number Publication date
US11517789B2 (en) 2022-12-06
CN106175781B (zh) 2019-08-20
CN106175781A (zh) 2016-12-07
US20190184233A1 (en) 2019-06-20

Similar Documents

Publication Publication Date Title
WO2018036316A1 (zh) 利用可穿戴设备监测游泳状态的方法及可穿戴设备
US20180249908A1 (en) Multi-state performance monitoring system
WO2018040757A1 (zh) 可穿戴设备及利用其监测运动状态的方法
Mason et al. What can we learn from competition analysis at the 1999 pan pacific swimming championships?
WO2019033586A1 (zh) 基于智能手表的游泳运动分析方法和智能手表
CN108245869B (zh) 一种游泳信息检测方法、装置及电子设备
CN113516175B (zh) 刷牙区域的识别方法、装置、刷牙设备和存储介质
Jensen et al. Classification of kinematic swimming data with emphasis on resource consumption
CN105450911B (zh) 图像处理装置、图像处理方法
CN109753868B (zh) 运动动作的评估方法及装置、智能手环
CN113713358B (zh) 基于多传感器融合的游泳监测方法、设备、存储介质和程序产品
CN102739957A (zh) 能够确定被摄体的运动的图像处理装置及图像处理方法
CN111265841B (zh) 一种游泳圈数确定方法、装置、设备及存储介质
CN109784178A (zh) 基于手势识别的参数调节方法、装置及健身设备
CN110909621A (zh) 一种基于视觉的健身指导系统
CN106178466A (zh) 一种健身用度量分析方法及系统
Toyoda et al. Post-data augmentation to improve deep pose estimation of extreme and wild motions
CN111803902B (zh) 泳姿识别方法、装置、可穿戴设备及存储介质
CN112274902B (zh) 游泳数据获取方法及装置、游泳数据采集设备的控制方法
CN108837476B (zh) 一种检测游泳起点的方法、装置和智能穿戴设备
CN109302532B (zh) 一种基于智能手机加速度传感器的身份认证方法
US20220241672A1 (en) Methods and systems for swim analysis
WO2022053943A1 (en) System and method for coaching a swimmer
CN110705496B (zh) 一种基于九轴传感器的游泳姿势识别方法
EP4128016A1 (en) Motion tracking of a toothcare appliance

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: 17842739

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17842739

Country of ref document: EP

Kind code of ref document: A1