CN109753868B - Method and device for evaluating movement actions and intelligent bracelet - Google Patents
Method and device for evaluating movement actions and intelligent bracelet Download PDFInfo
- Publication number
- CN109753868B CN109753868B CN201811355425.9A CN201811355425A CN109753868B CN 109753868 B CN109753868 B CN 109753868B CN 201811355425 A CN201811355425 A CN 201811355425A CN 109753868 B CN109753868 B CN 109753868B
- Authority
- CN
- China
- Prior art keywords
- motion
- data
- action
- motion trail
- trail data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000033001 locomotion Effects 0.000 title claims abstract description 409
- 230000009471 action Effects 0.000 title claims abstract description 225
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000011156 evaluation Methods 0.000 claims abstract description 57
- 238000010606 normalization Methods 0.000 claims abstract description 15
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims description 118
- 238000006243 chemical reaction Methods 0.000 claims description 29
- 238000000513 principal component analysis Methods 0.000 claims description 22
- 238000012549 training Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 17
- 230000000386 athletic effect Effects 0.000 claims description 11
- 239000013598 vector Substances 0.000 claims description 9
- 238000012847 principal component analysis method Methods 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 5
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 10
- 230000000694 effects Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 235000008694 Humulus lupulus Nutrition 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 2
- 230000037147 athletic performance Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009469 supplementation Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a method and a device for evaluating sports actions and an intelligent bracelet. Wherein the method comprises the following steps: collecting motion trail data I of a target object, and carrying out normal normalization preprocessing on the motion trail data I to obtain motion trail data II; determining a motion action corresponding to the motion trail data II, and calling a reference action of the motion action from a reference action library; determining the similarity between the motion trail data II and the motion trail data III of the reference motion; and evaluating the motion action of the target object according to the similarity to obtain evaluation data. The invention solves the technical problems that the wearable equipment in the related technology can not realize the identification of the complex actions of the user, and further can not evaluate the complex actions, so that the movement of the user is not scientific enough.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method and a device for evaluating motion actions and an intelligent bracelet.
Background
The current mass body building is more diversified and refined, a few gym enthusiasts can obtain the guidance of professional coaches in gymnasiums, but most amateur gym enthusiasts learn the professional training action courses through micro videos. Therefore, most lovers can exercise without professional guidance, and the exercise action is not standard, so that the satisfactory exercise effect cannot be achieved. Without specialized instruction, the trainer may be injured or lose the enthusiasm of training without achieving the training effect. Based on this problem, wearable devices have been developed, and so far, the action recognition based on the wearable device includes: game simple gesture recognition, daily walking steps count times, daily simple action recognition (standing, running, sleeping, falling … …) and golf swing recognition. The recognition of the above-mentioned movements is mostly based on the detection of the characteristics of the sensor raw data (acceleration and angular velocity data), different characteristics are required to be manually designed for different movements, a simple movement does not need a high recognition accuracy, and the characteristic design of the simple movement is also easier. However, when more complex actions are to be identified, it becomes a difficult problem to design appropriate features to distinguish between the complex actions.
Aiming at the problems that the wearable equipment in the related technology cannot recognize the complex actions of the user and further cannot evaluate the complex actions, so that the motions of the user are not scientific enough, no effective solution is proposed at present.
Disclosure of Invention
The embodiment of the invention provides a method and a device for evaluating sports actions and an intelligent bracelet, which at least solve the technical problems that in the related art, a wearable device cannot recognize complex actions of a user, and further cannot evaluate the complex actions, so that the sports of the user is not scientific enough.
According to an aspect of an embodiment of the present invention, there is provided a method for evaluating a sport action, including: collecting motion trail data I of a target object, and carrying out normal normalization preprocessing on the motion trail data I to obtain motion trail data II; determining a motion action corresponding to the motion trail data II, and calling a reference action of the motion action from a reference action library; determining the similarity between the motion trail data II and the motion trail data III of the reference action; and evaluating the motion action of the target object according to the similarity to obtain evaluation data.
Optionally, before the reference action of the motion action is invoked, the method for evaluating a motion action further comprises: collecting a plurality of groups of historical motion trail data in a historical time period and historical motion actions corresponding to each group of historical motion trail data in the plurality of groups of historical motion trail data; training the collected historical motion actions corresponding to each set of historical motion track data in the historical motion track data to obtain reference actions corresponding to each set of motion track data in the historical motion track data; and storing the reference actions corresponding to each group of motion trail data in the plurality of groups of historical motion trail data to obtain the reference action library.
Optionally, before determining the similarity between the second motion trajectory data and the third motion trajectory data of the reference motion, the method for evaluating the motion further includes: and performing coordinate transformation on the motion trail data II by using a Principal Component Analysis (PCA).
Optionally, performing coordinate transformation on the motion trail data two by using principal component analysis PCA includes: determining each group of motion trail data in the motion trail data II to carry out trail centralization treatment; generating a data matrix corresponding to the motion trail data according to the motion trail data obtained by trail centralization; generating a covariance matrix according to the data matrix, and obtaining a characteristic vector and a characteristic value of the covariance matrix; generating an intermediate matrix according to the eigenvectors of the covariance matrix; and obtaining second motion trail data subjected to coordinate conversion according to the intermediate matrix.
Optionally, generating a covariance matrix from the data matrix includes: generating a covariance matrix according to the data matrix through a first formula, wherein the first formula is as follows:c represents the covariance matrix, n represents the number of samples of the motion trail data two, X represents the data matrix, X T Representing a transpose of the data matrix.
Optionally, the intermediate matrix is a matrix of 3*3.
Optionally, obtaining the motion trail data two after coordinate conversion according to the intermediate matrix includes: obtaining motion trail data II after coordinate conversion through a second formula based on the intermediate matrix, wherein the second formula is as follows: y=px, Y represents the motion trajectory data two after coordinate conversion, P represents the intermediate matrix, and X represents the data matrix.
Optionally, after the reference motion corresponding to each set of motion trajectory data in the plurality of sets of historical motion trajectory data is stored to obtain the reference motion library, the motion evaluation method further includes: updating the reference action library; wherein updating the reference action library comprises: collecting a plurality of groups of sample data corresponding to the new actions; fitting the plurality of groups of sample data in a preset mode to obtain a reference action corresponding to the new action; and adding the reference action corresponding to the new action to the reference action library.
Optionally, after the motion action of the target object is evaluated according to the similarity, the evaluation method of the motion action further includes: and transmitting the evaluation data to the target object.
According to another aspect of the embodiment of the present invention, there is also provided an apparatus for evaluating a sport action, including: the processing unit is used for collecting motion trail data I of the target object, and carrying out normal normalization preprocessing on the motion trail data I to obtain motion trail data II; the invoking unit is used for determining the motion action corresponding to the motion trail data II and invoking the reference action of the motion action from a reference action library; the determining unit is used for determining the similarity between the motion trail data II and the motion trail data III of the reference motion; and the evaluation unit is used for evaluating the motion action of the target object according to the similarity to obtain evaluation data.
Optionally, the evaluation device of the athletic movement further includes: the acquisition unit is used for acquiring a plurality of groups of historical motion trail data in a historical time period and historical motion actions corresponding to each group of historical motion trail data in the plurality of groups of historical motion trail data before the reference action of the motion actions is called; the training unit is used for training the collected historical motion actions corresponding to each set of historical motion track data in the plurality of sets of historical motion track data to obtain reference actions corresponding to each set of motion track data in the plurality of sets of historical motion track data; and the storage unit is used for storing the reference actions corresponding to each group of motion trail data in the plurality of groups of historical motion trail data to obtain the reference action library.
Optionally, the evaluation device of the athletic movement further includes: and the conversion unit is used for carrying out coordinate conversion on the motion track data II by utilizing a principal component analysis method PCA before determining the similarity between the motion track data II and the motion track data III of the reference motion.
Optionally, the conversion unit includes: the determining module is used for determining each group of motion trail data in the motion trail data II to carry out trail centralization treatment; the first generation module is used for generating a data matrix corresponding to the motion trail data according to the motion trail data obtained through trail centralization; the second generation module is used for generating a covariance matrix according to the data matrix and obtaining a characteristic vector and a characteristic value of the covariance matrix; the third generation module is used for generating an intermediate matrix according to the eigenvectors of the covariance matrix; and the acquisition module is used for acquiring the motion trail data II after coordinate conversion according to the intermediate matrix.
Optionally, the second generating module includes: the generating submodule is used for generating a covariance matrix according to the data matrix through a first formula, wherein the first formula is as follows: C represents the covariance matrix, n represents the number of samples of the motion trail data two, X represents the data matrix, X T Representing a transpose of the data matrix.
Optionally, the intermediate matrix is a matrix of 3*3.
Optionally, the acquiring module includes: the obtaining submodule is used for obtaining the motion trail data II after coordinate conversion through a second formula based on the intermediate matrix, wherein the second formula is as follows: y=px, Y represents the motion trajectory data two after coordinate conversion, P represents the intermediate matrix, and X represents the data matrix.
Optionally, the evaluation device of the athletic movement further includes: the updating unit is used for updating the reference motion library after the reference motion corresponding to each group of motion trail data in the plurality of groups of historical motion trail data is stored to obtain the reference motion library; wherein the updating unit includes: the acquisition module is used for acquiring a plurality of groups of sample data corresponding to the new actions; the fitting module is used for fitting the plurality of groups of sample data in a preset mode to obtain a reference action corresponding to the new action; and the adding module is used for adding the reference action corresponding to the new action to the reference action library.
Optionally, the evaluation device of the athletic movement further includes: and the sending unit is used for sending the evaluation data to the target object after evaluating the motion action of the target object according to the similarity to obtain the evaluation data.
According to another aspect of the embodiments of the present invention, there is also provided a smart band using the method for evaluating a sports motion described in any one of the above.
According to another aspect of the embodiments of the present invention, there is also provided a storage medium including a stored program, wherein the program performs the method of evaluating a sport action of any one of the above.
According to another aspect of the embodiments of the present invention, there is also provided a processor for executing a program, wherein the program executes the method for evaluating a sports motion according to any one of the above.
In the embodiment of the invention, firstly, motion trail data I of a target object is acquired, and normal normalization preprocessing is carried out on the motion trail data I to obtain motion trail data II; then determining a motion action corresponding to the motion trail data II, and calling a reference action of the motion action from a reference action library; then determining the similarity between the motion trail data II and the motion trail data III of the reference motion; and evaluating the motion action of the target object according to the similarity to obtain evaluation data. According to the method for evaluating the motion actions, disclosed by the embodiment of the invention, the purpose of evaluating the motion actions by calculating the similarity of the motion actions corresponding to the motion track data and the reference actions according to the acquired motion track data can be achieved, the technical effect of improving the motion experience of a user is achieved, and the technical problem that the wearable equipment in the related art cannot recognize the complex actions of the user, and further cannot evaluate the complex actions, so that the motion of the user is not scientific is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
FIG. 1 is a flow chart of a method of evaluating athletic activity in accordance with an embodiment of the application;
FIG. 2 is a preferred flow chart of a method of evaluating athletic activity in accordance with an embodiment of the application;
FIG. 3 is a schematic diagram of misalignment of motion trajectory data according to an embodiment of the present application;
FIG. 4 is a schematic diagram of the alignment of motion trajectory data according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a set of actions for multiple acquisitions of opening and closing beats in accordance with an embodiment of the present application;
FIG. 6 is a schematic diagram of the results after fitting 3 coordinate axes, respectively, according to an embodiment of the present application;
FIG. 7 is a schematic diagram of a standard template action for open and close hops in accordance with an embodiment of the application; and
fig. 8 is a schematic view of an evaluation device of a sport action according to an embodiment of the present application.
Detailed Description
In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Aiming at the problems, the embodiment of the invention provides an evaluation method based on the motion action of the intelligent bracelet, which can guide a user to perform correct training, for example, when a target object (namely, the user) starts to move after wearing the intelligent bracelet, the intelligent bracelet can be started, in the course of course training, the intelligent bracelet can automatically collect motion track data of the user, the motion state of the user is recorded according to the motion track data, for example, the counting of the training action and the standard degree of the training action, the monitoring data (action counting and action scoring) of the intelligent bracelet can be uploaded to a mobile phone end in real time, and the mobile phone end can prompt the user whether the current action is reasonable in real time, so that the user can obtain good course feedback information in time. That is, the intelligent bracelet can monitor the action standard of the user in real time and remind the user of the action standard degree in real time. It should be noted that, besides the above-mentioned smart bracelet, other smart devices convenient for users to carry may also use the method for evaluating exercise actions provided by the embodiment of the present invention. The following detailed description will be given by taking the smart device as a smart bracelet in connection with the following embodiments.
Example 1
According to an embodiment of the present invention, there is provided a method embodiment of a method of evaluating athletic activity, it being noted that the steps illustrated in the flowchart of the figures may be performed in a computer system, such as a set of computer-executable instructions, and that, although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in an order other than that illustrated herein.
Fig. 1 is a flowchart of a method of evaluating a sport action according to an embodiment of the present invention, as shown in fig. 1, the method of evaluating a sport action including the steps of:
step S102, motion trail data I of the target object are collected, and normal normalization preprocessing is carried out on the motion trail data I to obtain motion trail data II.
In step S102, the collected motion trail data of the target object may be subjected to normal normalization preprocessing, so that the motion trail data may be subjected to normalization processing, and a data base may be provided for subsequent evaluation of motion actions of the target object.
Step S104, determining the motion corresponding to the motion trail data II, and calling the reference motion of the motion from the reference motion library.
Step S106, the similarity between the second motion trail data and the third motion trail data of the reference motion is determined.
Step S108, the motion action of the target object is evaluated according to the similarity, and evaluation data are obtained.
The evaluation data may include, but is not limited to, the following: the number of executions of the athletic maneuver, and the standard degree of the athletic maneuver.
Through the steps, the first motion trail data of the target object can be acquired first, and the first motion trail data is subjected to normal normalization pretreatment; then determining a motion action corresponding to the motion trail data II, and calling a reference action of the motion action from a reference action library; then determining the similarity between the motion trail data II and the motion trail data III of the reference motion; and evaluating the motion action of the target object according to the similarity to obtain evaluation data. Compared with the prior art that the wearable equipment cannot recognize the complex motion of the user and further cannot evaluate the complex motion, so that the motion of the user is not scientific, the motion evaluation method of the embodiment of the invention can calculate the similarity of the motion corresponding to the motion track data and the reference motion according to the acquired motion track data to evaluate the motion, thereby achieving the technical effect of improving the motion experience of the user, and further solving the technical problem that the wearable equipment in the prior art cannot recognize the complex motion of the user and evaluate the complex motion, so that the motion of the user is not scientific.
As an alternative embodiment, before the reference action of the exercise action is invoked, the evaluation method of the exercise action may further include: collecting a plurality of groups of historical motion trail data in a historical time period and historical motion actions corresponding to each group of historical motion trail data in the plurality of groups of historical motion trail data; training the collected historical motion actions corresponding to each set of historical motion track data in the plurality of sets of historical motion track data to obtain reference actions corresponding to each set of motion track data in the plurality of sets of historical motion track data; and storing the reference actions corresponding to each group of motion trail data in the plurality of groups of historical motion trail data to obtain a reference action library.
The following describes in detail the method for evaluating the motion provided by the embodiment of the present invention with reference to the accompanying drawings.
FIG. 2 is a preferred flow chart of a method of evaluating athletic performance in accordance with an embodiment of the invention, wherein the left hand branch in FIG. 2 generates a branch for an ideal performance (i.e., a reference performance), and the intelligent bracelet is used to repeatedly collect multiple standard performances, and an ideal standard template performance is generated for one athletic performance via modules 1 and 2 shown in FIG. 2; in fig. 2, after the user motion trail data collected by the intelligent bracelet is branched to the right side and passes through the module 1, the processed motion trail data and ideal template data are imported into the module 3, the similarity of the two motion trail data is calculated, and finally the counting and scoring of the user action are given.
Preferably, before determining the similarity between the second motion trajectory data and the third motion trajectory data of the reference motion, the method for evaluating the motion may further include: and performing coordinate transformation on the motion trail data II by using a Principal Component Analysis (PCA).
Specifically, in the embodiment of the invention, three algorithms are integrated to realize the recognition of multiple motion actions, and the 3 algorithms are respectively a module 1, a module 2 and a module 3.
For module 1 in fig. 2: and converting the 3D motion trail data coordinate system by using a principal component analysis method PCA.
For example, users have different wearing habits when wearing smart bracelets (e.g., some users are used to wearing bracelets in the left hand and some users are used to wearing bracelets in the right hand). During the training process, the user may move or turn around, so the motion track has random translational and rotational changes in the 3D coordinate system for the same motion track made by the user.
In the above implementation, performing coordinate transformation on the motion trajectory data two by using the principal component analysis PCA may include: determining each group of motion trail data in the motion trail data II to carry out trail centralization treatment; generating a data matrix corresponding to the motion trail data according to the motion trail data obtained by trail centralization; generating a covariance matrix according to the data matrix, and obtaining a characteristic vector and a characteristic value of the covariance matrix; generating an intermediate matrix according to the eigenvectors of the covariance matrix; and obtaining motion trail data II after coordinate conversion according to the intermediate matrix.
Aiming at the problems, the principal component analysis PCA can be used for finding out the characteristic vectors and the corresponding characteristic values of 3 principal components in the 3D motion trail data, and the 3D coordinate system is reestablished by taking the 3 characteristic vectors as coordinate axes, so that the alignment of the same motion trail data is realized, and the similarity calculation of the motion trail data is convenient.
The main component analysis PCA mainly comprises the following implementation steps:
1. assuming that 3D motion trail data x (0 … … n), y (0 … … n) and z (0 … … n) are obtained, mean values of x, y and z are calculated respectively, and the mean values are subtracted respectively to perform normalization processing, so that trail centering processing is performed on the motion trail data.
Preferably, generating the covariance matrix from the data matrix comprises: generating a covariance matrix according to the data matrix through a first formula, wherein the first formula is as follows:c represents covariance matrix, n represents sample number of motion trail data two, X represents data matrix, X T Representing the transposed matrix of the data matrix. That is, assuming that a data matrix (i.e., a normalized data matrix) corresponding to motion trajectory data obtained after trajectory centering processing is X, the number of samples is n, and the manner of obtaining the covariance matrix is as follows: />Then, eigenvectors and eigenvalues of the covariance matrix are found. The permutation eigenvectors generate an intermediate matrix P.
Preferably, the intermediate matrix is a matrix of 3*3. I.e. the intermediate matrix P is a matrix of 3*3.
As an alternative embodiment, obtaining the motion trajectory data two after coordinate conversion according to the intermediate matrix may include: obtaining motion trail data II after coordinate conversion through a second formula based on the intermediate matrix, wherein the second formula is as follows: y=px, Y represents the motion trajectory data two after coordinate conversion, P represents the intermediate matrix, and X represents the data matrix.
Module 2: the motion trajectory data of the plurality of standard motions after the fitting process generates an ideal standard template motion, for example, fig. 3 is a schematic diagram of the alignment processing of the motion trajectory data, wherein the motion trajectory data of the two dumbbell lifting motions is first collected, and as can be seen in fig. 3, the two original motion trajectory data do not coincide in the coordinate system. Fig. 4 is a schematic diagram of the alignment of motion trajectory data according to an embodiment of the present invention, and as shown in fig. 4, after two motion trajectory data are processed by using principal component analysis PCA, it is found that the two processed motion trajectory data are substantially aligned in a coordinate system.
For example, in the recognition algorithm, an ideal standard template action is required as a reference action for each motion action, and when a new action is added, standard actions are required to be acquired, wherein fig. 5 is a schematic diagram of a group of actions acquired for opening and closing runout multiple times according to an embodiment of the present invention. And then a common regression method (support vector product, gaussian process, polynomial fitting and the like) is used for fitting the plurality of 3D motion trail data samples. Fig. 6 is a schematic diagram of results obtained after 3 coordinate axes are respectively fitted according to an embodiment of the present invention, and then an ideal standard template action (i.e. a reference action) is generated by using the curve fitted by the 3 coordinate axes in fig. 6, and fig. 7 is a schematic diagram of a standard template action of opening and closing hops according to an embodiment of the present invention.
Module 3: and calculating the similarity between the motion trail data corresponding to the ideal template action (namely the reference action) and the processed user motion trail data.
Specifically, a common similarity calculation method (for example, euclidean distance, DTW, or cosine distance) may be used to calculate the similarity between the motion corresponding to the motion trajectory data after the PCA processing by the principal component analysis method and the reference motion. According to the calculated similarity, the training action times of the user and the standard actions of each movement action can be counted, and the effective information is timely fed back to the user, so that the movement efficiency of the user is improved.
As an alternative embodiment, after evaluating the motion of the target object according to the similarity, the method for evaluating the motion may further include: the evaluation data is transmitted to the target object.
The sending of the evaluation data to the target object may be through sending the evaluation data to a terminal device (for example, a mobile phone) of the target object, through which the user can be prompted in real time whether the current action is reasonable, so that the user can obtain good training course feedback information in time.
As an optional embodiment, after storing the reference motion corresponding to each set of motion trajectory data in the plurality of sets of historical motion trajectory data and obtaining the reference motion library, the method for evaluating the motion may further include: updating a reference action library; wherein updating the reference action library comprises: collecting a plurality of groups of sample data corresponding to the new actions; fitting a plurality of groups of sample data in a preset mode to obtain a reference action corresponding to the new action; and adding the reference action corresponding to the new action to the reference action library. Therefore, the continuous perfection and supplementation of the reference action library can be realized, the reference is better provided for the user training, and the user experience is improved.
The method for evaluating the motion actions provided by the embodiment of the invention has good universality, does not depend on the design of manual characteristics any more, can identify the newly-added track actions by only collecting the standard template actions of different motion actions and calculating the track similarity, and has great improvement on the accuracy and the accuracy of identification and also effectively improves the fitness satisfaction of users.
According to another aspect of the embodiments of the present invention, there is also provided a smart band using the method for evaluating a motion of any one of the above. Compared with the track recognition method in the prior art, more features are extracted based on the original data of acceleration and angular velocity, and then different recognition features are manually designed according to different actions, and the recognition of the motion actions is performed by combining the module 1 (principal component analysis method), the module 2 (ideal action fitting method) and the module 3 (track similarity calculation method) in the intelligent bracelet provided by the embodiment of the invention. It should be noted that, the method applied to the smart bracelet can also be applied to track recognition of all wearable devices, and the newly added motion can be recognized only by collecting standard template actions of different motion actions and calculating the similarity without manually designing features for each different action. For block 3 more similarity calculation methods, e.g. cosine distance, maximum common subsequence, etc. methods may be tried.
Example 2
According to the embodiment of the invention, the evaluation device for the sports motion is provided, and the evaluation device for the sports motion can be used for executing the evaluation method for the sports motion provided by the embodiment of the invention. The following describes an evaluation device for exercise actions provided by the embodiment of the present invention.
Fig. 8 is a schematic view of an evaluation apparatus of a sports motion according to an embodiment of the present invention, as shown in fig. 8, the evaluation apparatus of a sports motion may include: the processing unit 81, the retrieving unit 83, the determining unit 85, the evaluating unit 87. The evaluation device of the movement is described in detail below.
The processing unit 81 is configured to collect motion trajectory data one of the target object, and perform normal normalization preprocessing on the motion trajectory data one to obtain motion trajectory data two.
The retrieving unit 83 is connected to the processing unit 81, and is configured to determine a motion corresponding to the motion trajectory data two, and retrieve a reference motion of the motion from the reference motion library.
And a determining unit 85, connected to the retrieving unit 83, for determining the similarity between the second motion trajectory data and the third motion trajectory data of the reference motion.
And an evaluation unit 87, connected to the determination unit 85, for evaluating the motion of the target object according to the similarity, to obtain evaluation data.
It should be noted that the processing unit 81 in this embodiment may be used to perform step S102 in the embodiment of the present invention, the retrieving unit 83 in this embodiment may be used to perform step S104 in the embodiment of the present invention, the determining unit 85 in this embodiment may be used to perform step S106 in the embodiment of the present invention, and the evaluating unit 87 in this embodiment may be used to perform step S108 in the embodiment of the present invention. The above modules are the same as examples and application scenarios implemented by the corresponding steps, but are not limited to what is disclosed in the above embodiments.
In the embodiment of the invention, the processing unit 81 can be utilized to collect the first motion trail data of the target object, and perform normal normalization preprocessing on the first motion trail data to obtain the second motion trail data; then, determining a motion action corresponding to the motion trail data II by using a calling unit 83, and calling a reference action of the motion action from a reference action library; then the determining unit 85 determines the similarity between the first motion trail data and the second motion trail data of the reference motion; and the evaluation unit 87 is used for evaluating the motion action of the target object according to the similarity to obtain evaluation data. Compared with the prior art that the wearable equipment cannot recognize the complex motion of the user and further cannot evaluate the complex motion, so that the motion of the user is not scientific, the motion evaluation device of the embodiment of the invention can calculate the similarity of the motion corresponding to the motion track data and the reference motion according to the acquired motion track data to evaluate the motion, thereby achieving the technical effect of improving the motion experience of the user, and further solving the technical problem that the wearable equipment in the prior art cannot recognize the complex motion of the user and evaluate the complex motion, so that the motion of the user is not scientific.
As an alternative embodiment, the evaluation device of the sport action may further comprise: the acquisition unit is used for acquiring a plurality of groups of historical motion trail data in a historical time period and historical motion actions corresponding to each group of historical motion trail data in the plurality of groups of historical motion trail data before the reference action of the motion actions is called; the training unit is used for training the collected historical motion actions corresponding to each set of historical motion track data in the plurality of sets of historical motion track data to obtain reference actions corresponding to each set of motion track data in the plurality of sets of historical motion track data; the storage unit is used for storing the reference actions corresponding to each group of motion trail data in the plurality of groups of historical motion trail data to obtain a reference action library.
As an alternative embodiment, the evaluation device of the sport action may further comprise: and the conversion unit is used for carrying out coordinate conversion on the motion track data II by utilizing a principal component analysis method PCA before the similarity between the motion track data II and the motion track data III of the reference motion is determined.
As an alternative embodiment, the conversion unit may include: the determining module is used for determining each group of motion trail data in the second motion trail data to carry out trail centralization treatment; the first generation module is used for generating a data matrix corresponding to the motion trail data according to the motion trail data obtained through trail centralization; the second generation module is used for generating a covariance matrix according to the data matrix and obtaining a characteristic vector and a characteristic value of the covariance matrix; the third generation module is used for generating an intermediate matrix according to the eigenvectors of the covariance matrix; and the acquisition module is used for acquiring motion trail data II after coordinate conversion according to the intermediate matrix.
As an alternative embodiment, the second generating module may include: the generating submodule is used for generating a covariance matrix according to a data matrix through a first formula, wherein the first formula is as follows:c represents covariance matrix, n represents sample number of motion trail data two, X represents data matrix, X T Representing the transposed matrix of the data matrix.
Preferably, the intermediate matrix is a matrix of 3*3.
As an alternative embodiment, the acquiring module may include: the acquisition sub-module is used for obtaining motion trail data II after coordinate conversion through a second formula based on the intermediate matrix, wherein the second formula is as follows: y=px, Y represents the motion trajectory data two after coordinate conversion, P represents the intermediate matrix, and X represents the data matrix.
As an alternative embodiment, the evaluation device of the sport action may further comprise: the updating unit is used for updating the reference action library after the reference action corresponding to each group of motion trail data in the plurality of groups of historical motion trail data is stored to obtain the reference action library; wherein the updating unit includes: the acquisition module is used for acquiring a plurality of groups of sample data corresponding to the new actions; the fitting module is used for fitting the plurality of groups of sample data in a preset mode to obtain a reference action corresponding to the new action; and the adding module is used for adding the reference action corresponding to the new action to the reference action library.
As an alternative embodiment, the evaluation device of the sport action may further comprise: and the sending unit is used for sending the evaluation data to the target object after evaluating the motion action of the target object according to the similarity to obtain the evaluation data.
The evaluation device of the above-mentioned sport action includes a processor and a memory, the above-mentioned processing unit 81, retrieving unit 83, determining unit 85, evaluation unit 87, etc. are stored in the memory as program units, and the above-mentioned program units stored in the memory are executed by the processor to realize the corresponding functions.
The processor includes a kernel, and the kernel fetches a corresponding program unit from the memory. The kernel can be provided with one or more than one, and the motion action of the target object is evaluated according to the similarity by adjusting kernel parameters to obtain evaluation data.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), which includes at least one memory chip.
According to another aspect of the embodiments of the present invention, there is also provided a storage medium including a stored program, wherein the program performs the method of evaluating a sports motion of any one of the above.
According to another aspect of the embodiments of the present invention, there is further provided a processor, configured to execute a program, where the program executes the method for evaluating the motion of any one of the above steps.
In an embodiment of the present invention, there is also provided an apparatus including a processor, a memory, and a program stored in the memory and executable on the processor, the processor implementing the following steps when executing the program: collecting motion trail data I of a target object, and carrying out normal normalization preprocessing on the motion trail data I to obtain motion trail data II; determining a motion action corresponding to the motion trail data II, and calling a reference action of the motion action from a reference action library; determining the similarity between the motion trail data II and the motion trail data III of the reference motion; and evaluating the motion action of the target object according to the similarity to obtain evaluation data.
There is also provided in an embodiment of the invention a computer program product adapted to perform, when executed on a data processing apparatus, a program initialized with the method steps of: collecting motion trail data I of a target object, and carrying out normal normalization preprocessing on the motion trail data I to obtain motion trail data II; determining a motion action corresponding to the motion trail data II, and calling a reference action of the motion action from a reference action library; determining the similarity between the motion trail data II and the motion trail data III of the reference motion; and evaluating the motion action of the target object according to the similarity to obtain evaluation data.
The foregoing embodiment numbers of the present application are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, for example, may be a logic function division, and may be implemented in another manner, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-only memory (ROM), a random access memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (11)
1. A method of evaluating athletic activity, comprising:
collecting motion trail data I of a target object, and carrying out normal normalization preprocessing on the motion trail data I to obtain motion trail data II;
determining a motion action corresponding to the motion trail data II, and calling a reference action of the motion action from a reference action library;
performing coordinate transformation on the motion trail data II by using a Principal Component Analysis (PCA);
determining the similarity between the motion trail data II and the motion trail data III of the reference action;
evaluating the motion action of the target object according to the similarity to obtain evaluation data;
wherein, before the reference action of the motion action is invoked, the method further comprises: collecting a plurality of groups of historical motion trail data in a historical time period and historical motion actions corresponding to each group of historical motion trail data in the plurality of groups of historical motion trail data; training the collected historical motion actions corresponding to each set of historical motion track data in the historical motion track data to obtain reference actions corresponding to each set of motion track data in the historical motion track data; storing reference actions corresponding to each group of motion trail data in the plurality of groups of historical motion trail data to obtain the reference action library;
Wherein, after the reference motion corresponding to each set of motion trail data in the plurality of sets of historical motion trail data is stored to obtain the reference motion library, the method further comprises: updating the reference action library; wherein updating the reference action library comprises: collecting a plurality of groups of sample data corresponding to the new actions; fitting the plurality of groups of sample data in a preset mode to obtain a reference action corresponding to the new action; adding a reference action corresponding to the new action to the reference action library;
the coordinate conversion of the motion trail data two by using the principal component analysis PCA comprises the following steps:
determining each group of motion trail data in the motion trail data II to carry out trail centralization treatment;
generating a data matrix corresponding to the motion trail data according to the motion trail data obtained by trail centralization;
generating a covariance matrix according to the data matrix, and obtaining a characteristic vector and a characteristic value of the covariance matrix;
generating an intermediate matrix according to the eigenvectors of the covariance matrix;
obtaining motion trail data II after coordinate conversion through a second formula based on the intermediate matrix, wherein the second formula is as follows: y=px, Y represents the motion trajectory data two after coordinate conversion, P represents the intermediate matrix, and X represents the data matrix.
2. The method of claim 1, wherein generating a covariance matrix from the data matrix comprises:
generating a covariance matrix according to the data matrix through a first formula, wherein the first formula is as follows:
c represents the covariance matrix, n represents the number of samples of the motion trail data two, X represents the data matrix, X T Representing a transpose of the data matrix.
3. The method of claim 2, wherein the intermediate matrix is a matrix of 3*3.
4. The method according to claim 1, further comprising, after evaluating the motion of the target object based on the similarity, obtaining evaluation data: and transmitting the evaluation data to the target object.
5. An evaluation device for exercise motion, comprising:
the processing unit is used for collecting motion trail data I of the target object, and carrying out normal normalization preprocessing on the motion trail data I to obtain motion trail data II;
the invoking unit is used for determining the motion action corresponding to the motion trail data II and invoking the reference action of the motion action from a reference action library;
The conversion unit is used for carrying out coordinate conversion on the motion trail data II by utilizing a principal component analysis method PCA before determining the similarity between the motion trail data II and the motion trail data III of the reference motion;
the determining unit is used for determining the similarity between the motion trail data II and the motion trail data III of the reference motion;
the evaluation unit is used for evaluating the motion action of the target object according to the similarity to obtain evaluation data;
wherein, still include: the acquisition unit is used for acquiring a plurality of groups of historical motion trail data in a historical time period and historical motion actions corresponding to each group of historical motion trail data in the plurality of groups of historical motion trail data before the reference action of the motion actions is called; the training unit is used for training the collected historical motion actions corresponding to each set of historical motion track data in the plurality of sets of historical motion track data to obtain reference actions corresponding to each set of motion track data in the plurality of sets of historical motion track data; the storage unit is used for storing the reference actions corresponding to each group of motion trail data in the plurality of groups of historical motion trail data to obtain the reference action library;
Wherein, still include: the updating unit is used for updating the reference motion library after the reference motion corresponding to each group of motion trail data in the plurality of groups of historical motion trail data is stored to obtain the reference motion library; wherein the updating unit includes: the acquisition module is used for acquiring a plurality of groups of sample data corresponding to the new actions; the fitting module is used for fitting the plurality of groups of sample data in a preset mode to obtain a reference action corresponding to the new action; the adding module is used for adding the reference action corresponding to the new action to the reference action library;
wherein the conversion unit includes: the determining module is used for determining each group of motion trail data in the motion trail data II to carry out trail centralization treatment; the first generation module is used for generating a data matrix corresponding to the motion trail data according to the motion trail data obtained through trail centralization; the second generation module is used for generating a covariance matrix according to the data matrix and obtaining a characteristic vector and a characteristic value of the covariance matrix; the third generation module is used for generating an intermediate matrix according to the eigenvectors of the covariance matrix; the acquisition module is used for acquiring second motion trail data subjected to coordinate conversion according to the intermediate matrix;
Wherein, the acquisition module includes: the obtaining submodule is used for obtaining the motion trail data II after coordinate conversion through a second formula based on the intermediate matrix, wherein the second formula is as follows: y=px, Y represents the motion trajectory data two after coordinate conversion, P represents the intermediate matrix, and X represents the data matrix.
6. The apparatus of claim 5, wherein the second generating module comprises:
the generating submodule is used for generating a covariance matrix according to the data matrix through a first formula, wherein the first formula is as follows:c represents the covariance matrix, n represents the number of samples of the motion trail data two, X represents the data matrix, X T Representing a transpose of the data matrix.
7. The apparatus of claim 6, wherein the intermediate matrix is a matrix of 3*3.
8. The apparatus as recited in claim 5, further comprising: and the sending unit is used for sending the evaluation data to the target object after evaluating the motion action of the target object according to the similarity to obtain the evaluation data.
9. A smart bracelet, characterized in that it uses the method of evaluation of the movement actions of any one of claims 1 to 4.
10. A storage medium comprising a stored program, wherein the program performs the method of evaluating a athletic activity of any one of claims 1 to 4.
11. A processor, characterized in that the processor is adapted to run a program, wherein the program runs to perform the method of evaluating a sport action according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811355425.9A CN109753868B (en) | 2018-11-14 | 2018-11-14 | Method and device for evaluating movement actions and intelligent bracelet |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811355425.9A CN109753868B (en) | 2018-11-14 | 2018-11-14 | Method and device for evaluating movement actions and intelligent bracelet |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109753868A CN109753868A (en) | 2019-05-14 |
| CN109753868B true CN109753868B (en) | 2023-09-29 |
Family
ID=66402512
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811355425.9A Active CN109753868B (en) | 2018-11-14 | 2018-11-14 | Method and device for evaluating movement actions and intelligent bracelet |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109753868B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751060B (en) * | 2019-09-29 | 2021-02-19 | 西安交通大学 | A Portable Motion Pattern Real-time Recognition System Based on Multi-source Signals |
| CN110786825B (en) * | 2019-09-30 | 2022-06-21 | 浙江凡聚科技有限公司 | Spatial perception detuning training system based on virtual reality visual and auditory pathway |
| CN111757254B (en) * | 2020-06-16 | 2022-09-13 | 北京软通智慧科技有限公司 | Skating motion analysis method, device and system and storage medium |
| CN111800693A (en) * | 2020-06-30 | 2020-10-20 | 韶关市启之信息技术有限公司 | Bluetooth headset-based motion posture judgment method and system |
| CN111773651A (en) * | 2020-07-06 | 2020-10-16 | 湖南理工学院 | A system and method for monitoring and evaluating badminton training based on big data |
| CN113065979A (en) * | 2021-03-22 | 2021-07-02 | 贵州电网有限责任公司 | Load report improving and self-checking method for dispatching automation system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107029408A (en) * | 2017-05-03 | 2017-08-11 | 盐城工学院 | Method of motion analysis, device and electronic equipment |
| EP3557549A1 (en) * | 2018-04-19 | 2019-10-23 | PKE Holding AG | Method for evaluating a motion event |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101711488B1 (en) * | 2015-01-28 | 2017-03-03 | 한국전자통신연구원 | Method and System for Motion Based Interactive Service |
-
2018
- 2018-11-14 CN CN201811355425.9A patent/CN109753868B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107029408A (en) * | 2017-05-03 | 2017-08-11 | 盐城工学院 | Method of motion analysis, device and electronic equipment |
| EP3557549A1 (en) * | 2018-04-19 | 2019-10-23 | PKE Holding AG | Method for evaluating a motion event |
Non-Patent Citations (5)
| Title |
|---|
| Detecting abnormal fish trajectories using clustered and labeled data;Cigdem Beyan et al.;《IEEE Xplore》;全文 * |
| Hiroki Nagashima et al..Human-motion analysis of grasping/manipulating motion including time-variable function using principal component analysis.《 IEEE Xplore》.2013,全文. * |
| 一种基于可穿戴平台的武术动作量化评估方法;王漠等;《物联网技术》;20180817(第08期);全文 * |
| 基于混合高斯模型和主成分分析的轨迹分析行为识别方法;田国会 等;《电子学报》;20160115(第01期);第146页第4节 * |
| 基于运动历史轨迹的行为检测;陈坤 等;《仪表技术》;20150315(第03期);第38页第1节至第40页第4节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109753868A (en) | 2019-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109753868B (en) | Method and device for evaluating movement actions and intelligent bracelet | |
| US11615648B2 (en) | Practice drill-related features using quantitative, biomechanical-based analysis | |
| Fan et al. | Sensor fusion basketball shooting posture recognition system based on CNN | |
| DE102015207415A1 (en) | Method and apparatus for associating images in a video of a person's activity with an event | |
| CN103970271A (en) | Daily activity identifying method with exercising and physiology sensing data fused | |
| WO2014178794A1 (en) | Method and system for characterizing sporting activity | |
| US11620783B2 (en) | 3D avatar generation and robotic limbs using biomechanical analysis | |
| CN105848737B (en) | Analysis device, recording medium, and analysis method | |
| CN115331314B (en) | A method and system for evaluating exercise effect based on APP screening function | |
| Taghavi et al. | Tennis stroke detection using inertial data of a smartwatch | |
| CN111353345B (en) | Method, apparatus, system, electronic device, and storage medium for providing training feedback | |
| CN114259720A (en) | Motion recognition method and device, terminal equipment, motion monitoring system | |
| CN118690233A (en) | Rope skipping action recognition system and method based on deep learning | |
| Li et al. | [Retracted] Deep Learning Algorithm‐Based Target Detection and Fine Localization of Technical Features in Basketball | |
| CN112784699A (en) | Method and system for realizing posture evaluation guidance of sports coach | |
| CN111639232A (en) | Resource recommendation method and device, storage medium and electronic equipment | |
| CN106310609B (en) | A method and device for analyzing badminton | |
| Zaidi et al. | Mae Mai Muay Thai Layered Classification Using CNN and LSTM Models | |
| CN114358043A (en) | Motion recognition evaluation method, motion recognition evaluation device, and storage medium | |
| CN114492520A (en) | Action recognition method and device | |
| CN111460868A (en) | Action recognition error correction method, system, electronic device and storage medium | |
| US20130225294A1 (en) | Detecting illegal moves in a game using inertial sensors | |
| Andreeva et al. | Utilizing Gyroscope Data for Classifying Types of Fencer Movements in an Assistive Coaching System | |
| CN109453501A (en) | Ball training data processing method, equipment | |
| HK40069175A (en) | Motion recognition method and apparatus, terminal device, and sport monitoring system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20221202 Address after: L1202, China Resources Building, No. 2666, Keyuan South Road, Haizhu Community, Yuehai Street, Nanshan District, Shenzhen, Guangdong 518000 Applicant after: Shenzhen calorie Sports Technology Co.,Ltd. Address before: 518067 501K, Floor 5, Building 5, Nanhai Yiku, No. 6 Xinghua Road, Shenzhen, Guangdong Applicant before: SHENZHEN KALULI TECHNOLOGY Co.,Ltd. Applicant before: BEIJING CALORIE INFORMATION TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |