CN108242260B - Fitness monitoring method and device - Google Patents
Fitness monitoring method and device Download PDFInfo
- Publication number
- CN108242260B CN108242260B CN201611216325.9A CN201611216325A CN108242260B CN 108242260 B CN108242260 B CN 108242260B CN 201611216325 A CN201611216325 A CN 201611216325A CN 108242260 B CN108242260 B CN 108242260B
- Authority
- CN
- China
- Prior art keywords
- action
- data
- category
- motion
- classification model
- 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
Images
Landscapes
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention provides a fitness monitoring method and a fitness monitoring device, wherein the fitness monitoring method comprises the following steps: the method comprises the steps of obtaining physiological index data and motion data corresponding to an action of a user during body building, determining the action type of the action according to the physiological index data and the motion data, and reminding the user based on the action type of the action. According to the scheme of the invention, different reminders can be performed under different action types based on the physiological index data and the motion data of the user corresponding to the body-building action, so that whether the body-building action is standard or not can be comprehensively and reliably monitored.
Description
Technical Field
The invention relates to the technical field of wearable equipment, in particular to a fitness monitoring method and device.
Background
Due to the limitations of time, space, cost and the like, people rarely choose to exercise with a trainer in a gym at present when choosing the exercise mode, and the exercise mode often chosen is to simulate the action of the trainer in a video or a picture and combine with the explanation to exercise based on a video type or a picture type application program.
In this case, since the human being may have a deviation in understanding the received information, when the exerciser exercises based on the application program, the exerciser may consider his/her own exercise to be standard, but the exerciser may actually not be standard, resulting in poor exercise effect. In addition, the non-standard body-building actions not only can not achieve good body-building effect, but also can easily hurt the body-building person.
In the traditional fitness effect monitoring process, wearable equipment such as displacement sensors worn on forearms and hind arms and smart bracelets are often used for roughly monitoring the motion conditions of a fitness person, and whether fitness actions are standard or not is not comprehensively and reliably monitored.
Disclosure of Invention
The invention aims to provide a fitness monitoring method and a fitness monitoring device, which can comprehensively and reliably monitor whether the actions of a user are standard or not during fitness.
To achieve the above object, in one aspect, the present invention provides a fitness monitoring method, comprising:
acquiring physiological index data and motion data corresponding to an action of a user during body building;
determining an action type to which the action belongs according to the physiological index data and the motion data, wherein the action type is related to whether the action is standard or not;
and reminding the user based on the action category to which the action belongs.
Preferably, the step of determining the action category to which the action belongs according to the physiological index data and the motion data includes:
processing the physiological index data and the motion data by utilizing at least one action category classification model, and determining a possible action category to which the action belongs under each action category classification model;
and counting the possible action categories to which the actions belong, and selecting the possible action category with the most counting times as the action category to which the actions belong.
Preferably, before the step of obtaining the physiological index data and the exercise data corresponding to an action of the user during the exercise, the exercise monitoring method further includes:
determining at least one action category training data set and a basic classification model of the user, wherein each action category training data set comprises a plurality of groups of data divided into two action categories, each group of data is physiological index data and motion data of the user under one action category, and the basic classification model is used for reflecting the incidence relation among the physiological index data, the motion data and the action categories;
and training the basic classification model by using the at least one action class training data set to obtain the at least one action class classification model.
Preferably, the step of reminding the user based on the action category to which the action belongs includes:
determining a reminding mode corresponding to the action type based on the action type to which the action belongs;
and reminding the user according to the determined reminding mode.
Preferably, the action belongs to the action category of action amplitude standard, action amplitude not meeting standard or action amplitude exceeding standard.
Preferably, the physiological index data includes at least one of the following data: pulse frequency, heart rate variability value, body temperature value and myoelectric value.
Preferably, the motion data includes at least one of the following data: motion displacement, motion angle, and motion arc.
In another aspect, the present invention also provides a fitness monitoring device comprising:
the acquisition module is used for acquiring physiological index data and motion data corresponding to an action of a user during fitness;
the determining module is used for determining an action type to which the action belongs according to the physiological index data and the motion data, and the action type is related to whether the action is standard or not;
and the reminding module is used for reminding the user based on the action category to which the action belongs.
Preferably, the determining module includes:
the processing unit is used for processing the physiological index data and the motion data by utilizing at least one action category classification model and determining a possible action category to which the action belongs under each action category classification model;
and the counting unit is used for counting the possible action categories to which the actions belong and selecting the possible action category with the most counting times as the action category to which the actions belong.
Preferably, the fitness monitoring device further comprises:
the determination module is used for determining at least one action category training data set and a basic classification model of the user, each action category training data set comprises a plurality of groups of data divided into two action categories, each group of data is physiological index data and motion data of the user of one action category, and the basic classification model is used for reflecting the incidence relation among the physiological index data, the motion data and the action categories;
and the training module is used for training the basic classification model by utilizing the at least one action category training data set to obtain the at least one action category classification model.
Preferably, the reminding module includes:
the determining unit is used for determining a reminding mode corresponding to the action type based on the action type to which the action belongs;
and the reminding unit is used for reminding the user according to the determined reminding mode.
Preferably, the action belongs to the action category of action amplitude standard, action amplitude not meeting standard or action amplitude exceeding standard.
Preferably, the physiological index data includes at least one of the following data: pulse frequency, heart rate variability value, body temperature value and myoelectric value.
Preferably, the motion data includes at least one of the following data: motion displacement, motion angle, and motion arc.
According to the fitness monitoring method, the physiological index data and the motion data corresponding to one action of the user during fitness are obtained, the action type of the action is determined according to the physiological index data and the motion data, the user is reminded based on the action type of the action, different reminders can be performed under different action types based on the physiological index data and the motion data of the user corresponding to the fitness action, and whether the fitness action is standard or not is comprehensively and reliably monitored.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive labor.
Fig. 1 shows a flow chart of a fitness monitoring method according to a first embodiment of the invention.
Fig. 2 shows a flow chart of a fitness monitoring method according to a second embodiment of the invention.
Fig. 3 shows a schematic configuration of a fitness monitoring device according to a third embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
First embodiment
Referring to fig. 1, a first embodiment of the present invention provides a fitness monitoring method, which includes the following steps 101 to 103, which are described in detail below.
Step 101: the method comprises the steps of obtaining physiological index data and motion data corresponding to an action of a user during body building.
In an embodiment of the present invention, the physiological index data includes, but is not limited to, at least one of the following data: pulse frequency, heart rate variability value, body temperature value, myoelectric value and the like. The motion data includes, but is not limited to, at least one of the following: motion displacement, motion angle, motion radian and the like.
When the physiological index data of the user is obtained, the physiological index data can be obtained by means of a sports bracelet and the like worn by the user. The sports bracelet can monitor the pulse, heart rate variability, body temperature, myoelectric numerical value and the like of a corresponding user in real time. When acquiring the motion data of the user, the motion data may be acquired by various sensors or the like worn by the user at key parts of the body. Through the sensor worn by the user, the movement displacement (such as the relative displacement between the key parts of the body), the movement angle (such as the angle between the key parts of the body), the movement radian and the like of the corresponding user can be monitored in real time.
Step 102: and determining an action type to which the action belongs according to the physiological index data and the motion data, wherein the action type is related to whether the action is standard or not.
In the embodiment of the invention, the action category to which the action belongs is determined according to the physiological index data and the motion data corresponding to the action. Generally, the action category is related to whether the corresponding action is standard or not, and can reflect whether the corresponding action achieves the expected exercise effect or not. The effect of the movement is optimal only in the case of the action criterion. If the movement is not standard, not only the expected exercise effect is not achieved, but also the danger is brought to the user, such as muscle injury and the like.
Specifically, the action category to which the action belongs may be an action amplitude standard, an action amplitude not meeting the standard or an action amplitude exceeding the standard, or may be an action in place, an action not in place, or an action too violent, and the present invention does not limit the action.
Step 103: and reminding the user based on the action category to which the action belongs.
In the embodiment of the present invention, the manner of reminding the user may be vibration, ring tone, etc., and the present invention does not limit the manner.
Based on the action category to which the action belongs, the reminding of the user may specifically be:
determining a reminding mode corresponding to the action type based on the action type to which the action belongs;
and reminding the user according to the determined reminding mode.
Therefore, different action types adopt different reminding modes, so that a user can know the current action state in time, correct action correction is made under the condition that the action is not standard, and monitoring of the fitness effect and fitness guidance are realized.
For example, the corresponding relationship between the preset action type and the reminding mode is as follows:
standard action amplitude to no vibration;
the action amplitude does not reach the standard and vibrates with the first vibration intensity;
the action amplitude exceeds the standard and vibrates with a second vibration intensity; wherein the second vibration intensity is greater than the first vibration intensity.
Under the condition, when the action type is determined to be the action amplitude standard, no vibration prompt is provided, so that a user can know the action amplitude standard of the current action without correction; when the action type is determined to be that the action amplitude does not meet the standard, vibration reminding is carried out by using the first vibration strength, so that a user can know that the action amplitude of the current action does not meet the standard, and the action amplitude needs to be increased; when the action type is determined to be that the action amplitude exceeds the standard, vibration reminding is carried out at a second vibration intensity, so that a user can know that the action amplitude of the current action exceeds the standard, and the action amplitude needs to be reduced appropriately.
According to the fitness monitoring method, the physiological index data and the motion data corresponding to one action of the user during fitness are obtained, the action category to which the action belongs is determined according to the physiological index data and the motion data, the user is reminded based on the action category to which the action belongs, different reminders can be performed under different action categories based on the physiological index data and the motion data corresponding to the fitness action of the user, and whether the fitness action is standard or not can be comprehensively and reliably monitored.
Second embodiment
Referring to fig. 2, a second embodiment of the present invention provides a fitness monitoring method, which includes the following steps 201 to 206, which are described in detail below.
Step 201: at least one action category training dataset and a base classification model for the user are determined.
In the embodiment of the invention, each action type training data set comprises a plurality of groups of data divided into two action types, and each group of data is physiological index data and motion data of the user under one action type. When the action category training data set is determined, the physiological index data and the motion data of the user in each action category can be recorded through the observation of a fitness trainer, so that the data set for training the basic classification model is formed.
The basic classification model is used for reflecting the incidence relation among the physiological index data, the motion data and the action category, and the basic classification model can analyze the unknown mode of the input based on the known classification knowledge so as to determine the category of the input. Specifically, for the basic classification model, the physiological index data and the motion data are input, and the corresponding action category is output. The basic classification model is pre-established, and may be a Support Vector Machine (SVM) classification model for performing class two classification using an RBF kernel function, although other Machine learning models may also be used to perform the identification of the action class in the embodiment of the present invention.
When the basic classification model is pre-established, if the physiological index data and the motion data include more data parameters, in order to reduce data redundancy, a Principal Component Analysis (PCA) algorithm may be used to reduce the dimensions of the data parameters, and store corresponding covariance matrix and dimensions for subsequent model training.
For example, the physiological index data includes 4 data parameters, i.e. pulse frequency, heart rate variability value, body temperature value and myoelectric value, and the motion data includes 3 data parameters, i.e. motion displacement, motion angle and motion radian, then the basic classification model will involve 7 data parameters, i.e. 7 dimensions, and the PCA algorithm can be used to reduce the dimensions, e.g. to 6 dimensions, because the dimensions involved are more. The specific dimension reduction process may adopt the conventional method, and is not described herein again.
Step 202: and training the basic classification model by using the at least one action class training data set to obtain at least one action class classification model.
In the embodiment of the invention, when the basic classification model is trained by using the action category training data set, the processing of data parameters, such as a dimension reduction process, when the basic classification model is established in advance is referred to, so as to ensure the accuracy of the trained classification model.
Step 203: and acquiring physiological index data and motion data corresponding to an action of the user during fitness.
In an embodiment of the present invention, the physiological index data includes, but is not limited to, at least one of the following data: pulse frequency, heart rate variability value, body temperature value, myoelectric value and the like. The motion data includes, but is not limited to, at least one of the following: motion displacement, motion angle, motion radian and the like.
Step 204: and processing the physiological index data and the motion data by utilizing the at least one action category classification model, and determining a possible action category to which the action belongs under each action category classification model.
In the embodiment of the invention, the action type classification model is used for carrying out two types of classification, and the action type to which the action belongs can be determined by processing the physiological index data and the motion data corresponding to the action.
Wherein the possible action category to which the action belongs is related to whether the action is a criterion or not. The possible action category can be action amplitude standard, action amplitude not meeting standard or action amplitude exceeding standard, or can be action in place, action not in place or action too hard.
Step 205: and counting the possible action categories to which the actions belong, and selecting the possible action category with the most counting times as the action category to which the actions belong.
In the embodiment of the invention, by means of at least one action category classification model and by adopting a voting statistic mode to determine the action category to which the action belongs, the accuracy of judging the action type can be improved, and the monitoring effect of fitness is improved.
For example, there are 3 motion classes, class 1, class 2, and class 3, and there are 3 motion class classification models, class 1, class 2, and class 3; if the possible action category to which the action D belongs is determined to be category 1 by using the classification model 1, the possible action category to which the action D belongs is determined to be category 2 by using the classification model 2, and the possible action category to which the action D belongs is determined to be category 1 by using the classification model 3, then "category 1" may be selected as the action category to which the action D belongs.
Step 206: and reminding the user based on the action category to which the action belongs.
In the embodiment of the present invention, the manner of reminding the user may be vibration, ring tone, etc., and the present invention does not limit the manner.
Based on the action category to which the action belongs, the reminding of the user may specifically be:
determining a reminding mode corresponding to the action type based on the action type to which the action belongs;
and reminding the user according to the determined reminding mode.
Therefore, different action types adopt different reminding modes, so that a user can know the current action state in time, correct action correction is made under the condition that the action is not standard, and monitoring of the fitness effect and fitness guidance are realized.
Hereinafter, a dumbbell will be taken as an example to explain the fitness monitoring method according to an embodiment of the present invention.
The fitness monitoring method of the embodiment mainly comprises three steps, which are detailed as follows:
the method comprises the following steps: establishing a basic classification model
And establishing an incidence relation model, namely a basic classification model, with the action category based on the physiological index data and the motion data of the experimental exerciser. The physiological index data comprise pulse times, heart rate variability values, body temperature values and myoelectricity values, and are acquired by means of a sports bracelet worn by an experimental exerciser when a dumbbell is lifted. The exercise data comprise exercise displacement, exercise angles and exercise radians, and are acquired by means of sensors worn on key parts by an experimental exerciser when lifting a dumbbell. The action categories are classified into three categories, namely action in place (action category 1), action not in place (action category 2), and action too hard (action category 3).
When the basic classification model is established, firstly, the physical index data and the motion data of the experimental exerciser in the states of in-place action, not in-place action or violent action when the experimental exerciser lifts the dumbbell are recorded through the observation of a fitness coach, so that 3 data samples are obtained, and the action numbers corresponding to the data samples are the same.
Then, 3 SVM classification models, namely basic classification models, are established according to the three types of action categories, the three types of action models are input as pulse times (X1), heart rate variability values (X2), body temperature values (X3), myoelectric values (X4), movement displacement (X5), movement angles (X6) and movement radians (X7) of the experimental exerciser when the experimental exerciser holds the dumbbell, and the three types of action models are output as action categories. The SVM classification model is a two-class classification model.
When the SVM classification model is established, SVM algorithm is used for determining classification model parameters. Specifically, the physiological index data and the motion data in each data sample can be normalized in the manner of Xi'＝(Xi-Ximin)/(Ximax-Ximin)，XimaxIs physiological index data or motion data XiMaximum value of (A), XiminIs XiMinimum value of (d); reducing the dimension, namely 7 dimensions, because 7 data parameters are involved, adopting a PCA algorithm to map the 7-dimensional data parameters into data parameters with smaller dimensions, such as 6-dimensional data parameters, and storing the involved covariance matrix C and the dimension 6 for subsequent model training to use in order to reduce data redundancy; carrying out grid optimization of classification parameters, using an SVM classification model of an RBF kernel function to relate to a penalty coefficient C and a self-contained parameter gamma, and determining the optimal C and gamma through cross validation of 3 data samples; and finally, carrying out two-class classification by using an RBF kernel function, substituting the best C and gamma, setting probability estimation, and determining classification model parameters.
Step two: training motion class classification model
In this particular embodiment, an action class classification model is trained for exerciser A. Under the supervision of a fitness coach, pulse times (X1), heart rate variability values (X2), body temperature values (X3), myoelectric values (X4), movement displacement (X5), movement angles (X6) and movement radians (X7) of n times of exercise A in dumbbell lifting are recorded, and action categories are classified, as shown in the following training data set.
And training a motion class classification model between any two classes of motion classes, namely training three SVM classification models, wherein the three SVM classification models respectively correspond to the motion class 1 and the motion class 2, the motion class 1 and the motion class 3, and the motion class 2 and the motion class 3.
Thus, the training data set is divided into three action category training data sets, which are:
action class training dataset 1, corresponding SVM1 (action class 1 and action class 2)
Action class training dataset 2, corresponding SVM2 (action class 1 and action class 3)
Action class training dataset 3, corresponding SVM3 (action class 2 and action class 3)
When training the action category classification model, firstly, referring to the establishment process of the basic classification model, the data in the action category training data set 1, the action category training data set 2 and the action category training data set 3 are normalized and dimension reduced respectively, then the SVM1 is trained by using the processed action category training data set 1 to obtain the action category classification model 1, the SVM2 is trained by using the processed action category training data set 2 to obtain the action category classification model 2, and the SVM3 is trained by using the processed action category training data set 3 to obtain the action category classification model 3.
Step three: fitness monitoring using motion class classification models
In the process of body-building monitoring, the physiological index data and the exercise data, namely the monitoring data (X), corresponding to the action alpha when the exerciser A lifts the dumbbell are obtained1a、X2a、X3a、X4a、X5a、X6a、X7a) The monitoring data is normalized and dimension reduced, and the processed monitoring data is analyzed by utilizing action class classification models 1, 2 and 3 respectively, so as to determine the possible action class to which the action alpha belongs under each action class classification model. If the possible action category to which the action α belongs is the action category 1 under the action category classification model 1, the possible action category to which the action α belongs is the action category 2 under the action category classification model 2, and the possible action category to which the action α belongs is the action category 2 under the action category classification model 3, the action category to which the action α belongs is determined to be the action category 2.
If there is no vibration alert in the action type 1, a relatively slow vibration alert in the action type 2, and a relatively fast vibration alert in the action type 3 are preset, the action type to which the action α belongs is determined to be the action type 2, and then the relatively fast vibration alert is performed.
In addition, it should be noted that, after the user takes a fitness period (half a month/one month), the circumference of the fitness area (for example, the circumference of the arm, the hardness of the muscle, etc.) may be measured to detect the fitness effect. If the body-building effect does not reach the target, for example, the muscle hardness is not enough, the action type training data set can be corrected under the help of a body-building coach, and the association relationship between the physiological index data and/or the motion data and the action type can be adjusted, so that the body-building guidance effect is improved. For example, if the exercise effect is considered to be the best when the heart rate variability value is s1 in the previous action category training data set, but the exercise effect is not reached, the heart rate variability value can be adjusted, for example, to s2, and then the exercise effect is detected. Finally, by continuously adjusting the action category training data set, the expected fitness effect is achieved when the SVM model trained by the training data set is used for fitness monitoring.
According to the fitness monitoring method provided by the second embodiment of the invention, through at least one trained action category classification model, the physiological index data and the motion data corresponding to the action are processed, the possible action category to which the action belongs under each action category classification model is determined, the determined possible action categories are counted, the possible action category with the largest number of times of counting is selected as the action category to which the action belongs, and the user is reminded based on the selected action category, so that the comprehensive and reliable monitoring on whether the fitness action is standard can be realized, the accuracy of judging the action type can be improved, and the fitness monitoring effect is improved.
Third embodiment
Referring to fig. 3, a third embodiment of the present invention provides a fitness monitoring device corresponding to the fitness monitoring method shown in fig. 1, including:
the acquiring module 31 is used for acquiring physiological index data and motion data corresponding to an action of a user during fitness;
a determining module 32, configured to determine, according to the physiological index data and the motion data, an action category to which the action belongs, where the action category is related to whether the action is a standard or not;
and the reminding module 33 is configured to remind the user based on the action category to which the action belongs.
In an embodiment of the present invention, the determining module 32 includes:
the processing unit is used for processing the physiological index data and the motion data by utilizing at least one action category classification model and determining a possible action category to which the action belongs under each action category classification model;
and the counting unit is used for counting the possible action categories to which the actions belong and selecting the possible action category with the most counting times as the action category to which the actions belong.
Specifically, this body-building monitoring devices still includes:
the determination module is used for determining at least one action category training data set and a basic classification model of the user, each action category training data set comprises a plurality of groups of data divided into two action categories, each group of data is physiological index data and motion data of the user of one action category, and the basic classification model is used for reflecting the incidence relation among the physiological index data, the motion data and the action categories;
and the training module is used for training the basic classification model by utilizing the at least one action category training data set to obtain the at least one action category classification model.
Specifically, the reminding module 33 includes:
the determining unit is used for determining a reminding mode corresponding to the action type based on the action type to which the action belongs;
and the reminding unit is used for reminding the user according to the determined reminding mode.
The action category to which the action belongs is an action amplitude standard, an action amplitude which does not meet the standard or an action amplitude which exceeds the standard.
Wherein the physiological indicator data includes, but is not limited to, at least one of the following: pulse frequency, heart rate variability value, body temperature value and myoelectric value.
Wherein the motion data includes, but is not limited to, at least one of: motion displacement, motion angle, and motion arc.
The fitness monitoring device according to the third embodiment of the invention determines the action category to which the action belongs by acquiring the physiological index data and the motion data corresponding to the action of the user during fitness, and reminds the user based on the action category to which the action belongs, so that different reminders can be performed under different action categories based on the physiological index data and the motion data corresponding to the fitness action of the user, and the comprehensive and reliable monitoring of whether the fitness action is standard can be realized.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (10)
1. A method of fitness monitoring, comprising:
acquiring physiological index data and motion data corresponding to an action of a user during body building;
determining an action type to which the action belongs according to the physiological index data and the motion data, wherein the action type is related to whether the action is standard or not;
reminding the user based on the action category to which the action belongs;
the step of determining the action category to which the action belongs according to the physiological index data and the motion data comprises:
processing the physiological index data and the motion data by utilizing at least one action category classification model, and determining a possible action category to which the action belongs under each action category classification model; the action category classification model is used for representing the incidence relation between physiological index data and motion data and action categories;
counting the possible action categories to which the actions belong, and selecting the possible action category with the most counting times as the action category to which the actions belong;
before the step of obtaining the physiological index data and the motion data corresponding to an action of the user during fitness, the method further comprises the following steps:
determining at least one action category training data set and a basic classification model of the user, wherein each action category training data set comprises a plurality of groups of data divided into two action categories, each group of data is physiological index data and motion data of the user under one action category, and the basic classification model is used for reflecting the incidence relation among the physiological index data, the motion data and the action categories;
and training the basic classification model by using the at least one action class training data set to obtain the at least one action class classification model.
2. The fitness monitoring method of claim 1, wherein the step of alerting the user based on the category of action to which the action belongs comprises:
determining a reminding mode corresponding to the action type based on the action type to which the action belongs;
and reminding the user according to the determined reminding mode.
3. The fitness monitoring method of claim 1, wherein the action belongs to the action category that is action amplitude standard, action amplitude not meeting standard, or action amplitude exceeding standard.
4. A fitness monitoring method according to any of claims 1-3, wherein the physiological metric data comprises at least one of: pulse frequency, heart rate variability value, body temperature value and myoelectric value.
5. The fitness monitoring method of any of claims 1-3, wherein the movement data comprises at least one of: motion displacement, motion angle, and motion arc.
6. A fitness monitoring device, comprising:
the acquisition module is used for acquiring physiological index data and motion data corresponding to an action of a user during fitness;
the determining module is used for determining an action type to which the action belongs according to the physiological index data and the motion data, and the action type is related to whether the action is standard or not;
the reminding module is used for reminding the user based on the action category to which the action belongs;
the determining module comprises:
the processing unit is used for processing the physiological index data and the motion data by utilizing at least one action category classification model and determining a possible action category to which the action belongs under each action category classification model; the action category classification model is used for representing the incidence relation between physiological index data and motion data and action categories;
the statistical unit is used for counting the possible action categories to which the actions belong, and selecting the possible action category with the most statistical times as the action category to which the actions belong;
the fitness monitoring device further comprises:
the determination module is used for determining at least one action category training data set and a basic classification model of the user, each action category training data set comprises a plurality of groups of data divided into two action categories, each group of data is physiological index data and motion data of the user of one action category, and the basic classification model is used for reflecting the incidence relation among the physiological index data, the motion data and the action categories;
and the training module is used for training the basic classification model by utilizing the at least one action category training data set to obtain the at least one action category classification model.
7. The fitness monitoring device of claim 6, wherein the reminder module comprises:
the determining unit is used for determining a reminding mode corresponding to the action type based on the action type to which the action belongs;
and the reminding unit is used for reminding the user according to the determined reminding mode.
8. The fitness monitoring device of claim 6, wherein the action belongs to the action category that is action amplitude criteria, action amplitude not meeting criteria, or action amplitude exceeding criteria.
9. The fitness monitoring device of any of claims 6-8, wherein the physiological metric data comprises at least one of: pulse frequency, heart rate variability value, body temperature value and myoelectric value.
10. The fitness monitoring device of any of claims 6-8, wherein the movement data comprises at least one of: motion displacement, motion angle, and motion arc.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611216325.9A CN108242260B (en) | 2016-12-26 | 2016-12-26 | Fitness monitoring method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611216325.9A CN108242260B (en) | 2016-12-26 | 2016-12-26 | Fitness monitoring method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108242260A CN108242260A (en) | 2018-07-03 |
| CN108242260B true CN108242260B (en) | 2020-12-08 |
Family
ID=62704767
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611216325.9A Active CN108242260B (en) | 2016-12-26 | 2016-12-26 | Fitness monitoring method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108242260B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109102859A (en) * | 2018-07-24 | 2018-12-28 | 北京大学第医院 | A kind of motion control method and system |
| CN109544188B (en) * | 2018-11-27 | 2020-09-15 | 湖南共睹互联网科技有限责任公司 | User behavior monitoring method based on guarantee transaction, terminal and readable storage medium |
| CN111028911A (en) * | 2019-12-04 | 2020-04-17 | 广州华立科技职业学院 | Motion data analysis method and system based on big data |
| CN113457105B (en) * | 2020-03-30 | 2022-09-13 | 乔山健身器材（上海）有限公司 | Intelligent mirror with body-building menu |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101785927A (en) * | 2010-03-08 | 2010-07-28 | 游图明 | Interactive body building system |
| TW201111014A (en) * | 2009-09-25 | 2011-04-01 | Footwear & Recreation Technology Res Inst | Human-body muscle training feedback system and method |
| CN103810504A (en) * | 2014-01-14 | 2014-05-21 | 三星电子（中国）研发中心 | Image processing method and device |
| CN104888444A (en) * | 2015-05-20 | 2015-09-09 | 深圳大学 | Intelligent gloves, method and system for calorie consumption and hand posture recognition |
| CN105233485A (en) * | 2015-10-30 | 2016-01-13 | 山西睿智健科技有限公司 | A bodybuilding-assisting motion mode correction system and method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7970176B2 (en) * | 2007-10-02 | 2011-06-28 | Omek Interactive, Inc. | Method and system for gesture classification |
-
2016
- 2016-12-26 CN CN201611216325.9A patent/CN108242260B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW201111014A (en) * | 2009-09-25 | 2011-04-01 | Footwear & Recreation Technology Res Inst | Human-body muscle training feedback system and method |
| CN101785927A (en) * | 2010-03-08 | 2010-07-28 | 游图明 | Interactive body building system |
| CN103810504A (en) * | 2014-01-14 | 2014-05-21 | 三星电子（中国）研发中心 | Image processing method and device |
| CN104888444A (en) * | 2015-05-20 | 2015-09-09 | 深圳大学 | Intelligent gloves, method and system for calorie consumption and hand posture recognition |
| CN105233485A (en) * | 2015-10-30 | 2016-01-13 | 山西睿智健科技有限公司 | A bodybuilding-assisting motion mode correction system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108242260A (en) | 2018-07-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108242260B (en) | Fitness monitoring method and device | |
| Quaid et al. | Wearable sensors based human behavioral pattern recognition using statistical features and reweighted genetic algorithm | |
| CN106984027B (en) | A kind of movement comparative analysis method, apparatus and a kind of display | |
| US10970661B2 (en) | System and method for monitoring motion and orientation patterns associated to physical activities of users | |
| WO2018127506A1 (en) | Apparatus and method for triggering a fall risk alert to a person | |
| CN109363659B (en) | Heart rate monitoring method and device based on deep learning and storage medium | |
| US20190143174A1 (en) | Motion training guide system based on wearable sensor and method thereof | |
| CN108888277B (en) | Psychological test method, psychological test system and terminal equipment | |
| CN102368297A (en) | Equipment, system and method for recognizing actions of detected object | |
| JP2018029706A (en) | Terminal device, evaluation system, and program | |
| CN109595757B (en) | Control method and device of air conditioner and air conditioner with control device | |
| WO2021142532A1 (en) | Activity recognition with deep embeddings | |
| CN107247974B (en) | Body-building exercise identification method and system based on multi-source data fusion | |
| CN115661930A (en) | Action scoring method and device, action scoring equipment and storage medium | |
| TW201947358A (en) | Adaptive sport posture sensing system and method | |
| CN111814552A (en) | Method, device, system, equipment and medium for recognizing kicking behavior | |
| EP3664101A1 (en) | A computer-implemented method and an apparatus for use in detecting malingering by a first subject in one or more physical and/or mental function tests | |
| CN115100560A (en) | Method, device and equipment for monitoring bad state of user and computer storage medium | |
| EP3425637A2 (en) | Electronic system and method for classifying a physiological state | |
| US11361674B2 (en) | Encouraging speech system, encouraging speech method, and computer readable medium | |
| US11786156B2 (en) | Method and apparatus for use in detecting malingering by a first subject in tests of physical and/or mental function of the first subject | |
| Sosopoulos et al. | IoT smart athletics: Boxing glove sensors implementing machine learning for an integrated training solution | |
| CN112580514A (en) | Baby crawling state identification method and system | |
| CN112396114A (en) | Evaluation system, evaluation method and related product | |
| CN116304595B (en) | Intelligent motion analysis system and method based on shared cloud platform |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |