CN110471529A

CN110471529A - Act methods of marking and device

Info

Publication number: CN110471529A
Application number: CN201910725545.1A
Authority: CN
Inventors: 姚丽峰
Original assignee: Beijing Calorie Information Technology Co Ltd
Current assignee: Beijing Calorie Information Technology Co Ltd
Priority date: 2019-08-07
Filing date: 2019-08-07
Publication date: 2019-11-19

Abstract

This application discloses a kind of movement methods of marking and devices.Wherein, this method comprises: obtaining target action information, the target action information includes the mobile angular velocity information and/or acceleration information of target object；The type of corresponding target action is determined according to the target action information；Corresponding first code of points is determined based on the type of the target action；It scores according to first code of points the target action, obtains the first appraisal result, wherein first appraisal result is used to evaluate the similarity degree of the target action and standard operation.Present application addresses user by itself experience determine body-building movement whether standard, the lower technical problem of accuracy.

Description

Act methods of marking and device

Technical field

This application involves physical training skill fields, in particular to a kind of movement methods of marking and device.

Background technique

With the continuous enhancing of people's health consciousness, body-building group constantly increases, and increasingly it is hoped that there will be equipment or sides by people Method automatic identification and can record oneself all data during body-building.But current most of equipment are only capable of automatically recording On foot, the state run, manually opened can only record other movements, therefore, need a product that can be automatically recorded in strongly The movement carried out when body-building brings more conveniences to user.

In the prior art, user, to the assurance that itself body-building acts, is mostly based on subjective feeling, shortage pair in body-building User action whether the identification of standard, by the standard that user itself impression acts, body-building accuracy of action is lower.

For above-mentioned problem, currently no effective solution has been proposed.

Summary of the invention

The embodiment of the present application provides a kind of movement methods of marking and device, true by itself impression at least to solve user Determine body-building movement whether standard, the lower technical problem of accuracy.

According to the one aspect of the embodiment of the present application, a kind of movement methods of marking is provided, comprising: obtain target action letter Breath, the target action information include the mobile angular velocity information and/or acceleration information of target object；It is dynamic according to the target The type of corresponding target action is determined as information；Corresponding first code of points is determined based on the type of the target action； It scores according to first code of points the target action, obtains the first appraisal result, wherein first scoring As a result for evaluating the similarity degree of the target action and standard operation.

Optionally, after the type that corresponding target action is determined according to the target action information, the method is also wrapped Include: the target action is split as by the type based on the target action information and the target action according to preset rules Multiple sub- action sections；Corresponding second code of points of every sub- action section is determined based on the type of the multiple sub- action section；It presses It scores according to second code of points every sub- action section, obtains the second appraisal result, wherein described second comments Point result is used to evaluate the similarity degree of the sub- action section and the movement of standard.

Optionally, the preset rules include: by the angular velocity information in the target action information corresponding First extreme point of one magnitude of angular velocity, and/or the acceleration information corresponding first in the target action information is added The secondary extremal point of velocity amplitude is as splitting node, wherein first extreme point and the secondary extremal point include part most Big value and/or local minimum.

Optionally, the preset rules include: using the third extreme point of the fisrt feature track as fractionation node.It can Selection of land determines that the type of corresponding target action includes: to carry out to the target action information according to the target action information First pretreatment, obtains the first movement locus of corresponding first dimension of the target action information；Based on first movement Track determines motion characteristic information, and the motion characteristic information includes at least one of: first movement locus is corresponding Mean value, variance, cumulative arc length, the movement duration of curve；The type of the target action is determined according to the motion characteristic information.

Optionally, the first pretreatment is carried out to the target action information, obtains the target action information corresponding the First movement locus of dimension includes: to carry out the second pretreatment to the target action information, and second pretreatment includes At least one of: filtering, denoising, amendment；It is dynamic that the target is generated based on the second pretreated target action information Make the second movement locus of corresponding second dimension of information, wherein second dimension is greater than first dimension；By described Two movement locus are converted to first movement locus.

Optionally, determine that the type of the target action includes: by the motion characteristic according to the motion characteristic information Information input preset model is analyzed, and the type of the target action is obtained, wherein, the preset model is to pass through multiple groups number It is obtained according to training, every group of data in the multi-group data include: sample characteristics information and for marking the sample The label of the type of sample action corresponding to characteristic information.

Optionally, before the type based on the target action determines corresponding first code of points, the method is also wrapped It includes: establishing the type of the target action and the corresponding relationship of first code of points, first code of points includes: institute The higher scoring of similarity for stating the corresponding standard operation of target action is higher；And/or the storage corresponding relationship.

Optionally, carrying out scoring to the target action according to first code of points includes: to obtain the angular speed Corresponding second magnitude of angular velocity of information and/or corresponding second acceleration value of the acceleration information；Calculate second jiao of speed The first of angle value and/or the second acceleration value standard magnitude of angular velocity corresponding with the standard operation and/or normal acceleration value Difference；The first object score of the target action is determined according to first difference.

Optionally, carrying out scoring to the target action according to first code of points includes: to calculate the movement spy Reference ceases the second difference of standard feature information corresponding with the standard operation；The target is determined according to second difference Second target fractional of movement.

Optionally, second movement locus is converted to first movement locus includes: by the second movement rail Mark is converted to the third movement locus of corresponding first dimension of the target action information；Third movement locus is divided It cuts, obtains multiple first movement locus.

Optionally, after determining the type of corresponding target action according to the target action information, the method is also It include: the type for obtaining health and fitness facilities；Judge whether the type of the health and fitness facilities is corresponding with the type of the target action；Base Determine that corresponding first code of points includes: the type and the target in the health and fitness facilities in the type of the target action In the corresponding situation of the type of movement, corresponding first code of points is determined based on the type of the target action.

Optionally, before the type for obtaining health and fitness facilities, the method also includes: the connection for receiving the health and fitness facilities is asked It asks；It is attached based on the connection request and the health and fitness facilities.

Optionally, before the type for obtaining health and fitness facilities, the method also includes: detection target device and the fitness equipment Whether the distance of tool is less than pre-determined distance, if so, the connection of starting and the health and fitness facilities.

According to the one aspect of the embodiment of the present application, a kind of movement scoring apparatus is provided, comprising: obtain module, be used for Target action information is obtained, the target action information includes the mobile angular velocity information and/or acceleration information of target object； First determining module, for determining the type of corresponding target action according to the target action information；Second determining module is used Corresponding first code of points is determined in the type based on the target action；Grading module, for according to first scoring Rule scores to the target action, obtains the first appraisal result, wherein first appraisal result is described for evaluating The similarity degree of target action and standard operation.

According to the one aspect of the embodiment of the present application, a kind of storage medium is provided, the storage medium includes storage Program, wherein equipment where controlling the storage medium in described program operation executes above-mentioned movement methods of marking.

According to the one aspect of the embodiment of the present application, a kind of processor is provided, the processor is used to run program, In, described program executes above-mentioned movement methods of marking when running.

In the embodiment of the present application, using target action information is obtained, the target action information includes that target object moves Dynamic angular velocity information and/or acceleration information；The type of corresponding target action is determined according to the target action information；Base Corresponding first code of points is determined in the type of the target action；According to first code of points to the target action It scores, obtains the first appraisal result, wherein first appraisal result is for evaluating the target action and standard operation Similarity degree mode, movement is identified automatically with acceleration by the angular speed based on user's particular body portion, And to the mode that the movement of user is given a mark, reaches and the type of action of user has been determined automatically, and user has been moved It gives a mark, so that user is determined the technical effect of movement accuracy by itself actual workout data, and then solve User experienced by itself determine body-building movement whether standard, accuracy lower technical problem.

Detailed description of the invention

The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings:

Fig. 1 is the flow diagram according to the optional movement methods of marking of one kind of the embodiment of the present application；

Fig. 2 is the structural schematic diagram according to the optional movement points-scoring system of one kind of the embodiment of the present application；

Fig. 3 is the flow chart of data processing schematic diagram according to the optional movement points-scoring system of one kind of the embodiment of the present application；

Fig. 4 is the structural schematic diagram according to the optional movement scoring apparatus of one kind of the embodiment of the present application.

Specific embodiment

In order to make those skilled in the art more fully understand application scheme, below in conjunction in the embodiment of the present application Attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only The embodiment of the application a part, instead of all the embodiments.Based on the embodiment in the application, ordinary skill people Member's every other embodiment obtained without making creative work, all should belong to the model of the application protection It encloses.

It should be noted that the description and claims of this application and term " first " in above-mentioned attached drawing, " Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way Data be interchangeable under appropriate circumstances, so as to embodiments herein described herein can in addition to illustrating herein or Sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that cover Cover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to Step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, product Or other step or units that equipment is intrinsic.

Firstly, the part noun or term that occur during the embodiment of the present application is described are suitable for following solution It releases:

PCA: principal component analysis is a kind of Multivariable Statistical Methods, it is one of most common dimension reduction method, by orthogonal There may be the variable datas of correlation to be converted to one group of linearly incoherent variable by one group for transformation, and the variable after conversion is claimed For principal component.

According to the embodiment of the present application, a kind of movement methods of marking embodiment is provided, it should be noted that in the stream of attached drawing The step of journey illustrates can execute in a computer system such as a set of computer executable instructions, although also, flowing Logical order is shown in journey figure, but in some cases, it can be to be different from shown or described by sequence execution herein The step of.

Fig. 1 is according to the flow diagram of the movement methods of marking of the embodiment of the present application, as shown in Figure 1, this method is at least Include the following steps:

Step S102, obtains target action information, target action information include the mobile angular velocity information of target object and/ Or acceleration information；

In some optional embodiments of the application, above-mentioned target action information can be by being set to body-building user wearing Wearable device in sensor acquisition, wherein accelerometer and gyroscope are provided in wearable device, target object can Think the arm or leg or foot of user.Target action information is received by terminal or wearable device.Optionally, on Stating target action information can also be acceleration information and rotation speed or other expression rotations or the data of direction change；Gyro Instrument can also be substituted with other detection directions or the sensor of rotation, and magnetometer is also provided in wearable device.

Step S104 determines the type of corresponding target action according to target action information；

Optionally, the corresponding movement of the target action information received is target action, and the type of target action can be Russian swivel, push-up, volume abdomen, sit-ups, deep-knee-bend, burpee, burpee split squat, the kicking of support swivel, high lift leg, Running, the outward turning of dumbbell shoulder, 8 calligraphy and painting ring of small dumbbell, reversed flying bird of small dumbbell etc. of bending.

In some optional embodiments of the application, corresponding target action is determined according to the target action information Type can be realized in the following manner:

By the mobile angular velocity information of target object, and/or acceleration information, and/or the classification of motion characteristic information input Model obtains target action type；Wherein, disaggregated model is obtained by model training, utilizes known action and/or characteristic information Data carry out the model training disaggregated model that is trained or optimized；Using trained disaggregated model to unknown User data feature, such as: the mobile angular velocity information of target object, and/or acceleration information, and/or motion characteristic information into Row Classification and Identification determines the type of action of user；The model includes but is not limited to decision tree, random forest etc..

Step S106 determines corresponding first code of points based on the type of target action；

In some optional embodiments of the application, different type of action, corresponding standard operation is different, same Movement from different standard operation ratios, have different similarities, therefore, need to be determined in the type based on target action corresponding Before first code of points, the type of target action is determined.After the type for determining target action, according to the type of target action Determine corresponding first code of points.

Step S108 scores to target action according to the first code of points, obtains the first appraisal result, wherein the Similarity degree of one appraisal result for evaluation goal movement and standard operation, and/or act for evaluation goal in movement Score and good bad equal distribution situation on each key takeaway.

Wherein, standard operation can be the pre-set specification movement of administrator, specifically, scoring target action When, it can be according to the corresponding angular velocity information of target action and/or acceleration information standard angular speed corresponding with standard operation The similarity of information and/or normal acceleration information scores.

Optionally, after the type that corresponding target action is determined according to the target action information, the method is also needed Following steps: the type based on the target action information and the target action are executed, by the target action according to default Rule is split as multiple sub- action sections；Determine that every sub- action section corresponding second is commented based on the type of the multiple sub- action section Divider is then；It scores according to second code of points every sub- action section, obtains the second appraisal result, wherein Second appraisal result is used to evaluate the similarity degree of the sub- action section and the movement of standard.

Optionally, the preset rules can be with are as follows: the angular velocity information in the target action information is corresponding First extreme point of the first magnitude of angular velocity, and/or by the acceleration information corresponding first in the target action information The secondary extremal point of acceleration value is as fractionation node, wherein first extreme point and the secondary extremal point include part Maximum value and/or local minimum.

Wherein, splitting node is the split point that target action is split as to sub- action section, and the first extreme point can be first The local minimum of the local maximum of magnitude of angular velocity or the first magnitude of angular velocity.Secondary extremal point can be the first acceleration value Local maximum or the first acceleration value local minimum.

Optionally, the preset rules can be with are as follows: using the third extreme point of the fisrt feature track as splitting node.

Specifically, fisrt feature track is curve, and the trough or crest location in curve can be third extreme point.

Specifically, above-mentioned different sub- action section can be the different crucial postures in same body-building movement, such as: In one running movement, user lifts leg, pedals ground, posture；Wherein, above-mentioned preset angle speed interval and predetermined acceleration section can Think coach or administrator's setting；Corresponding first magnitude of angular velocity of above-mentioned sub- action section can actually be somebody's turn to do for body-building user When sub- action section, magnitude of angular velocity that wearable device detects.Corresponding first acceleration value of above-mentioned sub- action section can be strong Body user is when actually doing the sub- action section, acceleration value that wearable device detects.

Optionally, the corresponding predetermined acceleration section of different postures is different from preset angle speed interval.Based on described The target action is split as multiple sub- action sections according to preset rules by the type of target action information and the target action When, each angular velocity information and/or the acceleration information in target action information can be obtained, it is corresponding to search each angular velocity information The first magnitude of angular velocity and/or corresponding first acceleration value of each acceleration information where preset angle speed interval and/or Predetermined acceleration section.Such as: in running movement, sub- action section can be lift leg action section or leg thrust section.Such as: lift leg Acting corresponding preset angle speed interval is (a, b), and the corresponding preset angle speed interval of leg thrust is (c, d)；When by target When movement is split as multiple sub- action sections according to preset rules, if detecting, the first magnitude of angular velocity in target action information belongs to Section (a, b) then splits target action corresponding with first magnitude of angular velocity in lift leg action section, if detecting, target is dynamic Make the first magnitude of angular velocity in information and belong to section (c, d) then to pedal the fractionation of corresponding with first magnitude of angular velocity target action In ground action section.

In addition, corresponding second code of points of every sub- action section can be different, and such as: the lift leg movement in running movement Section standard operation corresponding from leg thrust section is different, their the second code of points is also different.

By to user's same body-building movement in, the determination of corresponding second appraisal result of different gestures point, convenient for for User shows that user in the accuracy of each posture point, can be improved with each posture of timely correction user as user's progress The efficiency of exercise guide.

Optionally, determine that the type of corresponding target action can carry out reality in the following manner according to target action information It is existing: the first pretreatment being carried out to target action information, obtains the first movement locus of corresponding first dimension of target action information； Determine that motion characteristic information, motion characteristic information may include at least one of based on the first movement locus: the first movement rail Mean value, variance, cumulative arc length, the movement duration of the corresponding curve of mark；The type of target action is determined according to motion characteristic information.

Optionally, motion characteristic information can also be special for acceleration or the corresponding characteristic value of gyro data and its fusion Sign.

In some optional embodiments of the application, the above method also needs to execute following steps: the movement to identifying Marking deciding grade and level is carried out, and/or error correction or error prompting etc. are carried out to user action.The method of movement marking deciding grade and level, including but not It is limited to, carries out score value prediction with the matching analysis of standard operation or according to expert/coach's marking trained model of result Deng.

In some optional embodiments of the application, above-mentioned first pretreatment includes but is not limited to data filtering, track Generation, data conversion etc.；Optionally, initial actuating track can will be generated according to collected target action information, to first initiating It is filtered and removes dryness and/or data conversion as track, obtain the obvious data of motion characteristic under specific dimension, it is specific at this Dimension both can be the first dimension, obtain the first movement locus；Data conversion, including but not limited to, PCA (Principal Component Analysis, principal component analysis), IPCA (Incremental Principal Component, increment it is main at Analysis) etc..Wherein, the first dimension can be two dimension, and the first movement locus can be by the angular speed in target action information Information and acceleration information carry out the track after dimensionality reduction conversion.

Optionally, the corresponding curve of the first movement locus can be the first movement locus curve.

Optionally, the first pretreatment is carried out to target action information, obtains corresponding first dimension of target action information First movement locus can be obtained in the following manner: carry out the second pretreatment to target action information, described second is pre- Processing can be at least one of: filtering, denoising, amendment are not limited to this three kinds；Based on the second pretreated target action Information generates the second movement locus of corresponding second dimension of target action information, wherein the second dimension is greater than the first dimension；It will Second movement locus is converted to the first movement locus.

In some optional embodiments of the application, it can be to incite somebody to action that the second movement locus, which is converted to the first movement locus, Second movement locus is converted to the first movement locus by PCA or IPCA.

In some optional embodiments of the application, the second movement locus, which is converted to the first movement locus, to be passed through Following manner is realized: second movement locus is converted to corresponding first dimension of the target action information Third movement locus；Third movement locus is split, multiple first movement locus are obtained.It is alternatively possible to right When the target action information carries out the first pretreatment, the data of the obtained target action information are handled to obtain just In the data for carrying out movement segmentation, the movement at this, which is divided into, is divided into multiple first movement locus for third movement locus.

Second movement locus is converted to the third movement of corresponding first dimension of the target action information After track, method also needs to execute following steps: third movement locus is split, for data prediction is obtained the Three movement locus carry out track segmentation, are partitioned into corresponding first movement locus of multiple sub- movements；Data segmentation, is used to basis The characteristic point information of third movement locus is split third movement locus, determines the beginning and end of a movement；Third The characteristic point of movement locus, including but not limited to, the extreme points such as wave crest, trough, data endpoint etc.；The method of data segmentation, packet It includes but is not limited to, peak valley detection, end-point detection；The method of end-point detection, including but not limited to, correlation method, variance method, spectrum distance from Method, spectrum entropy method, can zero than with can entropy ratio the end-point detecting methods such as end-point detection, wavelet transformation, EMD.

Optionally, determine that the type of target action can be realized in the following manner according to motion characteristic information: will Motion characteristic information input preset model is analyzed, and the type of target action is obtained, wherein, preset model is to pass through multiple groups number It is obtained according to training, every group of data in multi-group data include: sample characteristics information and for marker samples characteristic information The label of the type of corresponding sample action.

Optionally, before determining corresponding first code of points based on the type of target action, method also needs to execute following Step: establishing the type of target action and the corresponding relationship of the first code of points, and the first code of points includes: target action and its The similarity of corresponding standard operation is higher, and scoring is higher；Store corresponding relationship.The method for establishing corresponding relationship, including but It is not limited to, determines the first code of points, using artificial intelligence such as machine learning, deep learning models by manual types such as coaches Mode determine prediction model score.

In some optional embodiments of the application, different target action types corresponds to the first different scoring rule Then, the type of target action and the corresponding relationship of the first code of points need to be established.

Optionally, target action is scored and can be realized in the following manner according to the first code of points: obtained Take corresponding second magnitude of angular velocity of angular velocity information and/or corresponding second acceleration value of acceleration information；Calculate second jiao of speed First difference of angle value and/or the second acceleration value standard magnitude of angular velocity corresponding with standard operation and/or normal acceleration value； The first object score of target action is determined according to the first difference.

Specifically he, scores and can realize in the following manner to target action according to the first code of points: obtaining Take corresponding second magnitude of angular velocity of angular velocity information；Calculate the second magnitude of angular velocity standard magnitude of angular velocity corresponding with standard operation First difference；The first object score of target action is determined according to the first difference.

Specifically, target action is scored and can be realized in the following manner according to the first code of points: obtained Take corresponding second acceleration value of acceleration information；Calculate the second acceleration value normal acceleration value corresponding with standard operation First difference；The first object score of target action is determined according to the first difference.Wherein, second magnitude of angular velocity and described the One magnitude of angular velocity can be identical, and second acceleration value can be identical with first acceleration value.

Wherein, the first difference is bigger, indicates that target action differs bigger with standard operation, first object score is lower.

Optionally, target action is scored and can be realized in the following manner according to the first code of points: meter Calculate the second difference of motion characteristic information standard feature information corresponding with standard operation；Target action is determined according to the second difference The second target fractional.

Wherein, the second difference is bigger, indicates that target action differs bigger with standard operation, the second target fractional is lower

Optionally, after first object score or the second target fractional has been determined, terminal or wearable device pair can be passed through First object score or the second target fractional are shown.

Motion characteristic information can be at least one of: the mean value of the corresponding curve of the first movement locus, adds up at variance Arc length, movement duration.Wherein, motion characteristic information can be the equal of the corresponding curve of the first movement locus in preset time period Value, variance, cumulative arc length, movement duration can be the preset time period, and cumulative arc length can be corresponding for the first movement locus Total arc length of curve.

Optionally, the second movement locus being converted to the first movement locus can realize in the following manner: by Two movement locus are converted to the third movement locus of corresponding first dimension of target action information；Third movement locus is divided It cuts, obtains multiple first movement locus.In some optional embodiments of the application, when target action includes multiple movements When, the type of target action may include multiple types.

Optionally, after the type for determining corresponding target action according to target action information, method also need execute with Lower step: the type of health and fitness facilities is obtained；Judge whether the type of health and fitness facilities is corresponding with the type of target action；Based on target The type of movement determines that corresponding first code of points includes: in the type of health and fitness facilities and the corresponding feelings of the type of target action Under condition, corresponding first code of points is determined based on the type of target action.

It is wearable when user for body-building process needs to use health and fitness facilities in some optional embodiments of the application Wearable device realizes the accurate record of workout data using the interactive function between wearable device and intelligent body-building instrument. Communication and/or data transmission can be carried out between wearable device and health and fitness facilities.When user and health and fitness facilities privileged site connect When touching, health and fitness facilities can issue signal to wearable device, and health and fitness facilities inform the corresponding instrument type of wearable device and type Number etc. information or wearable device enter AD HOC and start type and model of automatic identification health and fitness facilities etc..

Health and fitness facilities are common health and fitness facilities, when not being equipped with communication/data transmission module, using wearable device to user Held health and fitness facilities carry out automatic identification, obtain the information such as type and the model of health and fitness facilities, and it is dynamic further to obtain user for body-building The more accurate identification made and further evaluation.

Optionally, before the type for obtaining health and fitness facilities, the above method also needs to execute following steps: receiving health and fitness facilities Connection request；It is attached based on connection request and health and fitness facilities.

Optionally, before the type for obtaining health and fitness facilities, the above method also need execute following steps: detection target device with Whether the distance of the health and fitness facilities is less than pre-determined distance, if so, starting and the connections of the health and fitness facilities, such as: NFC or The connection type of person RFID.Wherein, target device can be wearable device, such as: wrist-watch, bracelet etc..

Wherein, the connection request of health and fitness facilities, the link order that can be sent for health and fitness facilities, or wearable to set When for detecting the health and fitness facilities for being less than pre-determined distance at a distance from wearable device, or when wearable device detects user When contacting health and fitness facilities, the link order of generation.

The interaction of wearable device and health and fitness facilities is suitable for more people's scenes and/or single scene.Intelligent body-building instrument, It include the function of common health and fitness facilities for the health and fitness facilities for being equipped with communication module.Communication module, including but not limited to, near field Communication module, bluetooth communication module, wifi module.Near field communication module, including but not limited to, NFC, RFID.Preferably, it uses The intelligence of health and fitness facilities is realized in NFC near field communication module.Wherein, the health and fitness facilities connecting with wearable device can be One, or multiple.

Specifically, under either more people's scenes or single scene, the communication modes such as near-field communication can be used to realize Communication/data transmission between wearable device and health and fitness facilities.When user's body and/or body a part and health and fitness facilities and Or its certain privileged site communicate/interact when contacting, and can effectively avoid more people's scenes, and/or near-field communication and close but not False triggering caused by beginning to use etc. etc..Specifically, the fitness equipment that user touches can be detected by wearable device Vibration judges whether user contacts fitness equipment.It, can be by working as when wearable device and health and fitness facilities carry out near field connection Wearable device detects starting connection when the distance apart from health and fitness facilities is less than pre-determined distance, or when health and fitness facilities detect Starting connection when its distance apart from wearable device is less than pre-determined distance.

Optionally, wearable device, includes at least communication module, and the preferred near field communication module of communication module can effectively drop The power demands of low communication module communicate/data transmission for carrying out with intelligent body-building instrument.

Optionally, wearable device includes at least one with the accelerometer compared with high sampling rate range, the number of accelerometer It can be adjusted according to specifically used scene according to sample rate, the range of data sampling rate is more than or equal to 2Hz.For example, in common mould Under formula, lower sample rate can be set, for saving power consumption and/or reducing data transmission/data processing pressure, in body-building Under this AD HOC in room, can be set allows wearable device to identify health and fitness facilities compared with high sampling rate.Lower data are adopted Sample rate is any value of 0~50Hz；The relatively high sampling rate value being arranged is not less than 50Hz.

In some optional embodiments of the application, the application is applicable to gymnasium (including household, public), The function of automatic identification can be opened with automatic or manual, or long-term open of default is not turned off, that is to say, that the automatic identification What the function of equipment was preferably realized under the scene of gymnasium.If user present position is not gymnasium, can directly close The function of automatic identification.

Optionally, the type of target action can be identified by the accelerometer of wearable device, if in order to more accurate, It may also be identified in conjunction with other sensors such as gyroscope, magnetometers.

In some optional embodiments of the application, when health and fitness facilities do not include communication module, such as common dumbbell, Horizontal bar.Wearable device can trigger the connection with health and fitness facilities, Huo Zheyong by the way that whether detection user contacts health and fitness facilities Family carries wearable device and does preset movement to trigger the connection of wearable device and health and fitness facilities.

In some optional embodiments of the application, when health and fitness facilities include communication module, for example including communication mould The dumbbell of block, the horizontal bar including communication module.Wearable device can by detection wearable device apart from health and fitness facilities away from From whether pre-determined distance is less than, if so, the connection of triggering wearable device and health and fitness facilities or user carry wearable set It is standby to do preset movement to trigger the connection of wearable device and health and fitness facilities or health and fitness facilities and send connection request and extremely wear Wear equipment.When the health and fitness facilities that wearable device detects are multiple, then select apart from nearest fitness equipment instrument to be the company of need The instrument connect can be connected or be controlled automatically after identification.

When there is multiple health and fitness facilities including communication module, if detecting distance is identical or when being not enough to distinguish, Ke Yitong Directly contact is crossed, the contact between equipment, such as NFC mode is can be, is also possible to the direct contact arrangement of user, or pass through Confirmed in conjunction with the mode of movement.

Optionally, after scoring target action, the method also needs to execute with following steps: whether judging scoring Lower than preset threshold, if so, sending prompt information to user, the prompt information includes carrying out error correction or mistake to target action False alert etc..The scoring includes but is not limited to the whole score of target action, and/or, target action is in each crucial scoring Score on point.According to target action in the score performance of each scoring point, determine which target action need to correct in terms of Or reinforce etc..Each scoring point, including but not limited to, each key takeaway of the movement.Each scoring point exists May exist some differences between difference movement.

When health and fitness facilities are equipped with communication/data transmission module, to avoid data error caused by identification mistake etc., it is preferable that It is transmitted by health and fitness facilities by data and the information such as the type of health and fitness facilities and model is issued into wearable device, it is strong to inform that it is held The information such as body instrument carry out finer action recognition convenient for wearable device.

Optionally, intelligent body-building instrument can also carry out communicating/data transmission with terminals such as mobile phones, be controlled by mobile phone terminal is unified The communication of system and scheduling wearable device and health and fitness facilities and/or data transmission procedure and its transmission data.Health and fitness facilities include But it is not limited to, dumbbell, small dumbbell, barbell etc..

In some optional embodiments of the application, health and fitness facilities can be connect with the wearable device of user, and will The type of health and fitness facilities is sent to wearable device, the type of the target action and type of health and fitness facilities is corresponding can be mesh The identical body-building movement corresponding with the type of health and fitness facilities of the type of mark movement, such as: the type of running movement, with treadmill Type, all corresponding running.

The wearable device can be bracelet, wrist-watch etc..

In the embodiment of the present application, using target action information is obtained, target action information includes that target object is mobile Angular velocity information and/or acceleration information；The type of corresponding target action is determined according to target action information；It is dynamic based on target The type of work determines corresponding first code of points；It scores according to the first code of points target action, obtains first and comment Divide result, wherein mode of first appraisal result for evaluation goal movement and the similarity degree of standard operation, by based on use The angular speed of family particular body portion automatically identifies movement with acceleration, and to the side that the movement of user is given a mark Formula has reached and has been determined automatically to the type of action of user, and gives a mark to the movement of user, passes through user certainly The actual workout data of body determines the technical effect of movement accuracy, and then it is dynamic by itself experiencing determining body-building to solve user Make whether standard, the lower technical problem of accuracy.

Fig. 2 is according to the structural schematic diagram of the movement points-scoring system of the embodiment of the present application, as shown in Fig. 2, the system is at least Include: data acquisition module 22, data preprocessing module 24, data segmentation module 26, characteristic extracting module 28, model training/ Classification of motion identification module 210, movement marking/error correction or cue module 212；Wherein:

Data acquisition module 22, for carrying out data acquisition using wearable device, the data of acquisition include accelerating degree According to the data with rotation speed or other expression rotations or direction change；Wearable device includes: accelerometer and gyroscope；Top Spiral shell instrument can also be substituted with other detection directions or the sensor of rotation；

Optionally, it is to acquire that data acquisition module 22, which executes the target action information in the step S102, Fig. 1 in Fig. 1, Data.

Data preprocessing module 24, the data for obtaining data acquisition module 22, which are handled to obtain, to be convenient for moving Make the data divided, processing includes but is not limited to data filtering, track generation, data conversion etc.；It more preferably, will be collected Data generate motion profile, are filtered and remove dryness and/or data conversion to track, it is more apparent to obtain motion characteristic under specific dimension Data；Data conversion, including but not limited to, PCA, increment principal component analysis iPCA etc..

Data segmentation module 26, the data for obtaining data preprocessing module 24 carry out data segmentation, are partitioned into one It is a to act corresponding data；Data segmentation, for being split according to the characteristic point information of data to action data, determines one The beginning and end of movement；The characteristic point of data, including but not limited to, the extreme points such as wave crest, trough, data endpoint etc.；Data The method of segmentation, including but not limited to, peak valley detection, end-point detection；The method of end-point detection, including but not limited to, correlation method, Variance method, spectrum Furthest Neighbor, spectrum entropy method, can zero than with can entropy ratio the end-point detecting methods such as end-point detection, wavelet transformation, EMD.

Characteristic extracting module 28 obtains after the data segmentation that data segmentation module 26 obtains data preprocessing module 24 Multiple to act corresponding data, the corresponding data of the single action that characteristic extracting module 28 is used to obtain carry out feature extraction, Feature, including but not limited to, the mean value that is obtained by track data, variance, cumulative arc length, movement duration and/or acceleration or The corresponding characteristic value of gyro data and its fusion feature；

Model training/classification of motion identification module 210, for being trained to the corresponding data of each movement and its feature Or Classification and Identification；Model training mainly carries out model training using the data of known action and its feature and is trained or excellent The disaggregated model of change；Model training/classification of motion identification module 210 be also used for trained disaggregated model to unknown User data feature carry out Classification and Identification and determine the type of action of user；Model includes but is not limited to decision tree, random forest Deng.

Act marking/error correction or cue module 212, act marking/error correction or cue module 212, for model training/ The movement that classification of motion identification module 210 identifies carries out marking deciding grade and level, and/or carries out error correction or error prompting to user action Deng, be optional module, can determine according to actual needs whether include the module.The method of movement marking deciding grade and level, including but it is unlimited In with the matching analysis of standard operation or according to trained model progress score value prediction of expert/coach's marking result etc..

Optionally, the step S104 in Fig. 1 can be mentioned by data preprocessing module 24, data segmentation module 26, feature Modulus block 28, model training/classification of motion identification module 210 are realized.Optionally, the step S108 in Fig. 1 can be beaten by movement Point/error correction or cue module 212 realize.

Fig. 3 is according to the flow chart of data processing schematic diagram of the optional movement points-scoring system of one kind of the embodiment of the present application, such as Shown in Fig. 3, the movement points-scoring system in above-mentioned Fig. 2 can also include wearable device 32, mobile phone terminal 34 and server 36.

Wherein, mobile phone terminal 34 can also be other intelligent terminals.

The wearable device 32 is used to collect the sensing data during user for body-building, and/or carries out action recognition Deng processing, data by sensing data and/or after processing are transferred to the client etc. of mobile phone terminal 34, mobile phone etc. eventually Further detailed results are carried out on end and abundant form is shown.

Specifically, data acquisition module 22, data preprocessing module 24, data segmentation module 26, characteristic extracting module 28, Model training/classification of motion identification module 210, movement marking/error correction or cue module 212 are all set to wearable device 32 In.

Optionally, the mobile phone terminal 34 and its application can also through data transmit etc. update wearable device algorithm and/ Or type of action etc.；The wearable device 32 can also be served only for collecting the sensing data during user for body-building, directly It is transferred to the mobile phone terminal 34 and its application, the pretreatment of data and Classification and Identification etc. are analyzed on mobile phone terminal 34；

Specifically, data acquisition module 22 is set in wearable device 32, data preprocessing module 24, data divide mould Block 26, characteristic extracting module 28, model training/classification of motion identification module 210, movement marking/error correction or cue module 212 are set It is placed in mobile phone terminal 34.

Optionally, data acquisition module 22 is set in wearable device 32, data preprocessing module 24, data divide mould With wearable device 32, model training/classification of motion identification module 210, movement are beaten for block 26, the setting of characteristic extracting module 28 Point/error correction or cue module 212 be set in mobile phone terminal 34.

The server 36 is the optional module of the movement points-scoring system, for being transmitted by data by wearable device 32 and/or the upper data of the application of mobile phone terminal 34 carry out data storage and further analysis etc., analysis result can also be returned to Mobile phone application etc. show and/or be updated to mobile phone application；

The wearable device 32 can also directly carry out data biography with the server 36 without the mobile phone application It is defeated, carry out data transmission with algorithm update etc..

Specifically, data acquisition module 22 is set in wearable device 32, data preprocessing module 24, data divide mould Block 26, characteristic extracting module 28, model training/classification of motion identification module 210, movement marking/error correction or cue module 212 are set It is placed in server 36.

According to the embodiment of the present application, additionally provide it is a kind of for implementing the movement scoring apparatus of above-mentioned movement methods of marking, As shown in figure 4, the device includes: to obtain module 42, the first determining module 44, the second determining module 46, grading module 48；Its In:

Module 42 is obtained, for obtaining target action information, target action information includes the mobile angular speed of target object Information and/or acceleration information；

First determining module 44, for determining the type of corresponding target action according to target action information；

Second determining module 46 determines corresponding first code of points for the type based on target action；

Grading module 48 obtains the first appraisal result for scoring according to the first code of points target action, In, similarity degree of first appraisal result for evaluation goal movement and standard operation.

Specifically, the first determining module 44 is used to carry out target action information the first pretreatment, obtains target action letter Cease the first movement locus of corresponding first dimension；Motion characteristic information, motion characteristic information are determined based on the first movement locus Including at least one of: mean value, variance, cumulative arc length, the movement duration of the corresponding curve of the first movement locus；According to movement Characteristic information determines the type of target action.

Optionally, after the type that corresponding target action is determined according to the target action information, described device is also used In: the target action is split as by the type based on the target action information and the target action according to preset rules Multiple sub- action sections；Corresponding second code of points of every sub- action section is determined based on the type of the multiple sub- action section；It presses It scores according to second code of points every sub- action section, obtains the second appraisal result, wherein described second comments Point result is used to evaluate the similarity degree of the sub- action section and the movement of standard.

Optionally, the preset rules can be with are as follows: presets belonging to corresponding first magnitude of angular velocity of different sub- action sections Predetermined acceleration section belonging to angular speed section and/or corresponding first acceleration value is different.

Specifically, the first determining module 44 is used to carry out the target action information the second pretreatment, and described second is pre- Processing can be at least one of: filtering, denoising, amendment；It is generated based on the second pretreated target action information Second movement locus of corresponding second dimension of the target action information, wherein second dimension is greater than first dimension Degree；Second movement locus is converted into first movement locus.

Optionally, the first determining module 44 obtains target for analyzing motion characteristic information input preset model The type of movement, wherein preset model is obtained by multi-group data training, and every group of data in multi-group data include: The label of sample characteristics information and the type for sample action corresponding to marker samples characteristic information.

Optionally, before determining corresponding first code of points based on the type of target action, device is also used to establish mesh The type of mark movement and the corresponding relationship of the first code of points, the first code of points can be with are as follows: the corresponding mark of target action The similarity of quasi- movement is higher, and scoring is higher；Store corresponding relationship.

Specifically, grading module 48 is for obtaining corresponding second magnitude of angular velocity of angular velocity information and/or acceleration information Corresponding second acceleration value；Calculate the second magnitude of angular velocity and/or the second acceleration value standard angle speed corresponding with standard operation First difference of angle value and/or normal acceleration value；The first object score of target action is determined according to the first difference.

Specifically, grading module 48 is used to carry out scoring to target action according to the first code of points to include: calculating action Second difference of characteristic information standard feature information corresponding with standard operation；The second of target action is determined according to the second difference Target fractional.

Optionally, the first determining module 44 is used to being converted to second movement locus into the target action information corresponding First dimension third movement locus；Third movement locus is split, multiple first movement locus are obtained.

Optionally, after the type for determining corresponding target action according to target action information, device is also used to: being obtained The type of health and fitness facilities；Judge whether the type of health and fitness facilities is corresponding with the type of target action；Type based on target action Determine that corresponding first code of points includes: to be based under the type of health and fitness facilities and the corresponding situation of the type of target action The type of target action determines corresponding first code of points.

Optionally, before the type for obtaining health and fitness facilities, the connection that described device is also used to receive the health and fitness facilities is asked It asks；It is attached based on the connection request and the health and fitness facilities.

Optionally, before the type for obtaining health and fitness facilities, described device is also used to detect target device and the fitness equipment Whether the distance of tool is less than pre-determined distance, if so, the connection of starting and the health and fitness facilities.

It should be noted that the preferred embodiment of embodiment illustrated in fig. 4 may refer to the phase of the embodiment of the method in Fig. 1 Description is closed, details are not described herein again.

According to the other side of the embodiment of the present application, a kind of storage medium is additionally provided, storage medium includes storage Program, optionally, in the present embodiment, storage medium is arranged to store the program code for executing following steps: obtaining Target action information, target action information include the mobile angular velocity information and/or acceleration information of target object；According to target Action message determines the type of corresponding target action；Corresponding first code of points is determined based on the type of target action；It presses It scores according to the first code of points target action, obtains the first appraisal result, wherein the first appraisal result is for evaluating mesh The similarity degree of mark movement and standard operation.

According to the other side of the embodiment of the present application, a kind of processor is additionally provided, processor is used to run program, In, it can be with the program code of following steps in the movement methods of marking of executing application when program is run: obtaining target action Information, target action information include the mobile angular velocity information and/or acceleration information of target object；According to target action information Determine the type of corresponding target action；Corresponding first code of points is determined based on the type of target action；It is commented according to first Divider then scores to target action, obtains the first appraisal result, wherein the first appraisal result for evaluation goal movement with The similarity degree of standard operation, and/or for evaluation goal score of the movement on each key takeaway of movement and good bad etc. Distribution situation.

Above-mentioned the embodiment of the present application serial number is for illustration only, does not represent the advantages or disadvantages of the embodiments.

In above-described embodiment of the application, all emphasizes particularly on different fields to the description of each embodiment, do not have in some embodiment The part of detailed description, reference can be made to the related descriptions of other embodiments.

In several embodiments provided herein, it should be understood that disclosed technology contents can pass through others Mode is realized.Wherein, the apparatus embodiments described above are merely exemplary, such as the division of unit, can be one kind Logical function partition, there may be another division manner in actual implementation, such as multiple units or components can combine or can To be integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual Coupling, direct-coupling or communication connection can be through some interfaces, the indirect coupling or communication connection of unit or module, It can be electrical or other forms.

Unit may or may not be physically separated as illustrated by the separation member, shown as a unit Component may or may not be physical unit, it can and it is in one place, or may be distributed over multiple units On.It can some or all of the units may be selected to achieve the purpose of the solution of this embodiment according to the actual needs.

It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list Member both can take the form of hardware realization, can also realize in the form of software functional units.

It, can if integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product To be stored in a computer readable storage medium.Based on this understanding, the technical solution of the application substantially or Say that all or part of the part that contributes to existing technology or the technical solution can embody in the form of software products Out, which is stored in a storage medium, including some instructions are used so that a computer equipment (can be personal computer, server or network equipment etc.) executes all or part of step of each embodiment method of the application Suddenly.And storage medium above-mentioned includes: USB flash disk, read-only memory (ROM, Read-Only Memory), random access memory The various media that can store program code such as (RAM, Random Access Memory), mobile hard disk, magnetic or disk.

The above is only the preferred embodiments of the application, it is noted that those skilled in the art are come It says, under the premise of not departing from the application principle, several improvements and modifications can also be made, these improvements and modifications also should be regarded as The protection scope of the application.

Claims

1. a kind of movement methods of marking characterized by comprising

Target action information is obtained, the target action information includes the mobile angular velocity information and/or acceleration of target object Information；

The type of corresponding target action is determined according to the target action information；

Corresponding first code of points is determined based on the type of the target action；

It scores according to first code of points the target action, obtains the first appraisal result, wherein described first Appraisal result is used to evaluate the similarity degree of the target action and standard operation.

2. the method according to claim 1, wherein determining that corresponding target is dynamic according to the target action information The type of work includes:

First pretreatment is carried out to the target action information, obtains the first of corresponding first dimension of the target action information Movement locus；

Motion characteristic information is determined based on first movement locus, and the motion characteristic information includes at least one of: institute State mean value, variance, cumulative arc length, the movement duration of the corresponding curve of the first movement locus；

The type of the target action is determined according to the motion characteristic information.

3. being obtained according to the method described in claim 2, being pre-processed it is characterized in that, carrying out first to the target action information The first movement locus to corresponding first dimension of the target action information includes:

Second pretreatment is carried out to the target action information, second pretreatment includes at least one of: filtering, goes It makes an uproar, correct；

The of corresponding second dimension of the target action information is generated based on the second pretreated target action information Two movement locus, wherein second dimension is greater than first dimension；

Second movement locus is converted into first movement locus.

4. according to the method described in claim 2, it is characterized in that, determining the target action according to the motion characteristic information Type include:

The motion characteristic information input preset model is analyzed, the type of the target action is obtained, wherein, it is described pre- If model is obtained by multi-group data training, every group of data in the multi-group data include: sample characteristics information with And the label for marking the type of sample action corresponding to the sample characteristics information.

5. the method according to claim 1, wherein the type based on the target action determines corresponding first Before code of points, the method also includes:

The type of the target action and the corresponding relationship of first code of points are established, first code of points includes: The similarity of the corresponding standard operation of the target action is higher, and scoring is higher；

Store the corresponding relationship.

6. the method according to claim 1, wherein determining that corresponding target is dynamic according to the target action information After the type of work, the method also includes:

Based on the type of the target action information and the target action, the target action is split as according to preset rules Multiple sub- action sections；

Corresponding second code of points of every sub- action section is determined based on the type of the multiple sub- action section；

It scores according to second code of points every sub- action section, obtains the second appraisal result, wherein described Second appraisal result is used to evaluate the similarity degree of the sub- action section and the movement of standard.

7. according to the method described in claim 6, it is characterized in that, the preset rules include: by the target action information In corresponding first magnitude of angular velocity of the angular velocity information the first extreme point, and/or will be in the target action information The secondary extremal point of corresponding first acceleration value of the acceleration information is as splitting node, wherein first extreme point It include local maximum and/or local minimum with the secondary extremal point.

8. according to the method described in claim 6, it is characterized in that, the preset rules include: by the fisrt feature track Third extreme point as split node.

9. according to the method described in claim 5, it is characterized in that, according to first code of points to the target action into Row scores

Obtain corresponding second magnitude of angular velocity of the angular velocity information and/or corresponding second acceleration of the acceleration information Value；

Calculate second magnitude of angular velocity and/or second acceleration value standard magnitude of angular velocity corresponding with the standard operation And/or the first difference of normal acceleration value；

The first object score of the target action is determined according to first difference.

10. according to the method described in claim 5, it is characterized in that, according to first code of points to the target action Carrying out scoring includes:

Calculate the second difference of motion characteristic information standard feature information corresponding with the standard operation；

The second target fractional of the target action is determined according to second difference.

11. according to the method described in claim 3, being moved it is characterized in that, second movement locus is converted to described first Include: as track

Second movement locus is converted to the third movement locus of corresponding first dimension of the target action information；

Third movement locus is split, multiple first movement locus are obtained.

12. the method according to claim 1, wherein determining corresponding mesh according to the target action information After the type of mark movement, the method also includes:

Obtain the type of health and fitness facilities；

Judge whether the type of the health and fitness facilities is corresponding with the type of the target action；

Determine that corresponding first code of points includes: based on the type of the target action

In the type situation corresponding with the type of the target action of the health and fitness facilities, the class based on the target action Type determines corresponding first code of points.

13. according to the method for claim 12, which is characterized in that before the type for obtaining health and fitness facilities, the method is also Include:

Receive the connection request of the health and fitness facilities；

It is attached based on the connection request and the health and fitness facilities.

14. according to the method for claim 12, which is characterized in that before the type for obtaining health and fitness facilities, the method is also Include:

Whether detection target device is less than pre-determined distance at a distance from the health and fitness facilities, if so, starting and the fitness equipment The connection of tool.

15. a kind of movement scoring apparatus characterized by comprising

Module is obtained, for obtaining target action information, the target action information includes the mobile angular speed letter of target object Breath and/or acceleration information；

First determining module, for determining the type of corresponding target action according to the target action information；

Second determining module, for determining corresponding first code of points based on the type of the target action；

Grading module obtains the first appraisal result for scoring according to first code of points the target action, Wherein, first appraisal result is used to evaluate the similarity degree of the target action and standard operation.

16. a kind of storage medium, which is characterized in that the storage medium includes the program of storage, wherein run in described program When control the storage medium where equipment perform claim require any one of 1 to 14 described in movement methods of marking.

17. a kind of processor, which is characterized in that the processor is for running program, wherein right of execution when described program is run Benefit require any one of 1 to 14 described in movement methods of marking.