Intelligent system for acquiring and analyzing sports data

Abstract

The invention discloses an intelligent system for acquiring and analyzing sports data, and relates to the technical field of data analysis. The method comprises the following steps: acquiring heart rate data and speed data of a user, respectively constructing a heart rate data curve graph and a speed data curve graph, acquiring a climbing point and a climbing index of the user, setting analysis periods, acquiring constant indexes in different analysis periods, generating a body mass index of the user, and constructing a constant evaluation model. The method comprises the following steps: acquiring recommended exercise duration of a user by utilizing a constant evaluation model, setting a heart rate standard range, acquiring an abnormal point of the user and a speed variable amplitude of the abnormal point, constructing a variable amplitude evaluation model, acquiring a heart rate trend of the user by utilizing the variable amplitude evaluation model, generating a recommended exercise speed in real time, and feeding back the recommended exercise speed; according to the technical scheme, the aerobic exercise efficiency of the user can be improved, and the aerobic exercise quality of the user can be guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to an intelligent system for collecting and analyzing sports data. Background Art

[0002] The collection and analysis of sports data is the product of the deep integration of modern technology and sports training. By integrating advanced sensor technology, artificial intelligence algorithms and big data analysis technology, it can accurately and real-time collect various sports data of users during exercise, and conduct in-depth analysis and mining, thereby providing users with scientific training guidance and decision-making basis; In the prior art, users often carry corresponding monitoring equipment when doing aerobic exercise, but the acquired monitoring data is not well utilized and can only realize the function of data recording. If these monitoring data can be used to analyze the user's athletic ability and real-time exercise status, it will be possible to provide the user with targeted exercise suggestions, such as recommended exercise duration and recommended exercise speed. When the user exercises according to the exercise suggestions, it will be able to effectively improve their exercise efficiency and quality. To this end, the present invention provides an intelligent system for sports data collection and analysis. Summary of the invention

[0003] The object of the present invention is to provide an intelligent system for collecting and analyzing sports data.

[0004] The purpose of the present invention can be achieved through the following technical solutions: An intelligent system for sports data collection and analysis, comprising the following modules: The data acquisition module is used to obtain the user's heart rate data and speed data, and to construct corresponding heart rate data curve graphs and speed data curve graphs respectively; A data analysis module is used to obtain the user's climbing points and their corresponding climbing indexes in the heart rate data curve graph, set the analysis period, obtain the constant indexes in different analysis periods, and generate the user's physical fitness index in combination with the speed data curve graph; A data evaluation module is used to construct a constant evaluation model based on the climbing index and the constant index of different analysis cycles, and use the constant evaluation model to obtain the recommended exercise duration of the user; The data monitoring module is used to set the standard range of heart rate and obtain the abnormal points of the user and their corresponding speed variation by combining the heart rate data curve graph and the speed data curve graph; The data optimization module is used to build a variation evaluation model based on the heart rate data, body mass index, and speed variation of different users, use the variation evaluation model to obtain the user's heart rate trend, generate the corresponding recommended exercise speed in real time and provide feedback.

[0005] Furthermore, the process of obtaining the user's heart rate data and speed data and constructing corresponding heart rate data curve graphs and speed data curve graphs respectively includes: The heart rate data refers to the real-time heart rate value of the user from the beginning to the end of a single aerobic exercise, and the speed data refers to the real-time speed of the user from the beginning to the end of a single aerobic exercise; A corresponding heart rate data curve graph is constructed based on the acquired heart rate data, with the ordinate being the real-time heart rate value and the abscissa being the time node; a corresponding speed data curve graph is constructed based on the acquired speed data, with the ordinate being the real-time speed and the abscissa being the time node.

[0006] Furthermore, the process of obtaining the user's climbing point and its corresponding climbing index in the heart rate data curve graph includes: Adjacent coordinate points in the heart rate data curve graph and the speed data curve graph are connected by smooth curves. In the heart rate data curve graph, the climbing angle between the tangent of each coordinate point and the positive direction of the y-axis is obtained in sequence according to the positive direction of the x-axis. When a climbing angle equal to 90 degrees appears for the first time, the corresponding coordinate point is taken as the climbing point, and the climbing index P of the user at the climbing point is obtained. z ;

[0007] The real-time heart rate value at the climbing point is taken as the climbing heart rate value P x , the heart rate data curve graph at the initial time C s The real-time heart rate value is taken as the initial heart rate value C x , the time node of the climbing point is taken as the climbing moment P s .

[0008] Furthermore, the process of setting the analysis period, obtaining the constant index in different analysis periods, and generating the user's body mass index in combination with the speed data curve graph includes: In the heart rate data curve graph, the time nodes are divided into different analysis periods according to the order of the positive direction of the x-axis, and the divided analysis periods are numbered as i, i=1, 2, ..., n, where n is the number of analysis periods in the heart rate data curve graph; Mark the real-time heart rate values ​​corresponding to each coordinate point in a single analysis period in the order of time nodes, and record them as X j , j = 1, 2, ..., m, m is the number of all coordinate points in the analysis period, and the constant index H in the analysis period is obtained z ;

[0009] Get the constant index H of each analysis period in the same heart rate data curve graph respectivelyzi , obtain the speed mean S in each analysis period in the speed data curve graph pi , obtain the user's body mass index T in the heart rate data curve graph z ;

[0010] in, These are all preset fixed parameters.

[0011] Furthermore, a constant evaluation model is constructed according to the climbing index and the constant index of different analysis cycles. The process of using the constant evaluation model to obtain the recommended exercise duration of the user includes: Setting a constant index threshold, comparing the constant index of each analysis cycle in the same heart rate data curve graph with the constant index threshold in the order of numbers, and marking the analysis cycle whose constant index is greater than or equal to the constant index threshold as an abnormal cycle; Counting the number of abnormal cycles in the same heart rate data curve graph, marking the number corresponding to the kth abnormal cycle as the recommended cycle number, and obtaining the recommended cycle numbers in different heart rate data curve graphs respectively; Generate a constant evaluation set according to the climbing index and the recommended cycle number in different heart rate data curve graphs, and divide the constant evaluation set into a first training set and a first test set; Constructing a first convolutional neural network, using different climbing indexes in the first training set as input data of the first convolutional neural network, using the corresponding number of recommended cycles in the first training set as output data of the first convolutional neural network, and training the first convolutional neural network to obtain a first initial convolutional neural network; Using the first test set to perform model verification on the first initial convolutional neural network, outputting the first initial convolutional neural network that is less than or equal to a preset first test error threshold as a constant evaluation model; In the subsequent exercise process, the climbing index of different users is obtained in real time and input into the constant evaluation model to output the corresponding recommended cycle number. The product of the recommended cycle number and the time length of the analysis cycle is used as the recommended exercise duration of the corresponding user.

[0012] Furthermore, the process of setting a standard range of heart rate and obtaining the abnormal points of the user and the corresponding speed variation amplitude by combining the heart rate data curve graph and the speed data curve graph includes: Set the heart rate standard range, and mark each coordinate point in the heart rate data curve as an abnormally high point, an abnormally low point, or a normal point according to the heart rate standard range; The abnormal points include too high abnormal points and too low abnormal points. The real-time speed S corresponding to a single abnormal point is obtained in the speed data curve diagram. y, and the real-time speed S corresponding to the previous coordinate point q , obtain the velocity amplitude B of the abnormal point;

[0013] t is the time interval between the abnormal point and its previous coordinate point. The speed variation includes positive and negative values. When it is a positive value, it represents that the real-time heart rate value is on an upward trend. When it is a negative value, it represents that the real-time heart rate value is on a downward trend.

[0014] Furthermore, the process of constructing a speed variation evaluation model based on the heart rate data, body mass index, and speed variation of different users includes: According to the real-time heart rate values ​​and speed variation and heart rate trends of different abnormal points, combined with the body mass index of the corresponding user, a variation evaluation model is generated, and the variation evaluation set is divided into a second training set and a second test set; Constructing a second convolutional neural network, taking different real-time heart rate values, speed variation, and body mass index in the second training set as input data of the second convolutional neural network, taking the corresponding heart rate trend in the second training set as output data of the second convolutional neural network, and training the second convolutional neural network to obtain a second initial convolutional neural network; The second initial convolutional neural network is model verified using the second test set, and a second initial convolutional neural network with a value less than or equal to a preset second test error threshold is output as a variation evaluation model.

[0015] Furthermore, the process of using the amplitude evaluation model to obtain the user's heart rate trend, generating the corresponding recommended exercise speed in real time and providing feedback includes: In the subsequent exercise process, abnormal points of different users and their corresponding real-time heart rate values ​​and speed variations are obtained in real time, and combined with the body mass index of the corresponding users, they are input into the variation evaluation model to output the corresponding heart rate trends, including increasing trends and decreasing trends; The user's current speed is obtained. When it is increasing, the current speed is lowered to generate a recommended movement speed. When it is decreasing, the current speed is increased to generate a recommended movement speed. The generated recommended movement speed is fed back to the user until there are no abnormal points.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention quantifies the user's heart rate data at the start of exercise into a climbing index, quantifies the user's heart rate data during exercise into a constant index, and combines the user's speed data to obtain the user's physical fitness index, which is conducive to comprehensively reflecting the user's exercise ability. A constant evaluation model is constructed based on the climbing index and the constant index, which can generate targeted recommended exercise duration for the user, which is conducive to improving the user's efficiency in aerobic exercise; 2. The present invention monitors the abnormal heart rate of the user during exercise by setting a standard range of heart rate, obtains the corresponding speed changes, and constructs a variation evaluation model in combination with the user's physical fitness coefficient. It can judge whether the user's heart rate during exercise is abnormal and obtain the heart rate trend. According to the heart rate trend, the corresponding recommended exercise speed is generated in time and fed back to the user, which is beneficial to ensure the quality of the user's aerobic exercise. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0018] like Figure 1 As shown, an intelligent system for sports data collection and analysis includes the following modules: The data acquisition module is used to obtain the user's heart rate data and speed data, and to construct corresponding heart rate data curve graphs and speed data curve graphs respectively; A data analysis module is used to obtain the user's climbing points and their corresponding climbing indexes in the heart rate data curve graph, set the analysis period, obtain the constant indexes in different analysis periods, and generate the user's physical fitness index in combination with the speed data curve graph; A data evaluation module is used to construct a constant evaluation model based on the climbing index and the constant index of different analysis cycles, and use the constant evaluation model to obtain the recommended exercise duration of the user; The data monitoring module is used to set the standard range of heart rate and obtain the abnormal points of the user and their corresponding speed variation by combining the heart rate data curve graph and the speed data curve graph; The data optimization module is used to build a variation evaluation model based on the heart rate data, body mass index, and speed variation of different users, use the variation evaluation model to obtain the user's heart rate trend, generate the corresponding recommended exercise speed in real time and provide feedback.

[0019] It should be further explained that, in the specific implementation process, the process of obtaining the user's heart rate data and speed data and constructing the corresponding heart rate data curve graph and speed data curve graph respectively includes: The heart rate data refers to the real-time heart rate value of the user from the beginning to the end of a single aerobic exercise, and the speed data refers to the real-time speed of the user from the beginning to the end of a single aerobic exercise. For different aerobic exercises, the speed data refers to different objects; For aerobic exercise with fixed equipment, such as treadmills and elliptical machines, the speed data refers to the real-time speed displayed on the fixed equipment; for aerobic exercise without fixed equipment, such as jogging, brisk walking, and climbing stairs, the speed data refers to the real-time speed monitored by the user's monitoring device; A corresponding heart rate data graph is constructed based on the acquired heart rate data, wherein the ordinate of the heart rate data graph is the real-time heart rate value and the abscissa is the corresponding time node; a corresponding speed data graph is constructed based on the acquired speed data, wherein the ordinate of the speed data graph is the real-time speed and the abscissa is the corresponding time node.

[0020] It should be further explained that, in a specific implementation process, the process of obtaining the user's climbing point and its corresponding climbing index in the heart rate data curve graph includes: Adjacent coordinate points in the heart rate data curve graph and the speed data curve graph are connected by smooth curves. In the heart rate data curve graph, the angles (acute angles in this case) between the tangent of each coordinate point and the positive direction of the y-axis are obtained in sequence according to the positive direction of the x-axis, and recorded as the climbing angle, wherein each coordinate point corresponds to a real-time heart rate value and a time node; After the aerobic exercise begins, the user's real-time heart rate value will gradually increase until it reaches a stable floating range. During this process, the climbing angle of each coordinate point will also increase until the real-time heart rate value begins to float. During this process, a coordinate point from high to low will appear for the first time, and its climbing angle is equal to 90 degrees. This coordinate point is used as the climbing point; The real-time heart rate value corresponding to the climbing point is used as the climbing heart rate value, the real-time heart rate value at the initial moment of the heart rate data curve is used as the initial heart rate value, and the time node corresponding to the climbing point is used as the climbing moment to obtain the user's climbing index at the climbing point, which is recorded as P z ;

[0021] Among them, P x is the climbing heart rate value, C x is the initial heart rate value, P s is the climbing time, C s is the initial moment.

[0022] It should be further explained that, in the specific implementation process, the process of setting the analysis period, obtaining the constant index in different analysis periods, and generating the user's body mass index in combination with the speed data curve graph includes: In the heart rate data curve graph, the time nodes are divided into different analysis periods in the order of the positive direction of the x-axis. The time length of each analysis period is the same. The divided analysis periods are numbered and recorded as i, i=1, 2, ..., n, where n is the number of analysis periods in the heart rate data curve graph; Taking any analysis period as an example, the real-time heart rate values ​​corresponding to each coordinate point in the analysis period are marked in the order of time nodes, and recorded as X j, j = 1, 2, ..., m, m is the number of coordinate points in the analysis period, and the constant index in the analysis period is obtained, which is recorded as H z ;

[0023] The same method is used to obtain the constant index of each analysis period in the heart rate data curve, which is recorded as H zi , obtain the speed average value in each analysis period in the speed data curve graph, the speed average value is the average value of the real-time speeds contained in the speed data curve graph in each analysis period, denoted as S pi , obtain the user's body mass index in the heart rate data curve, denoted as T z ;

[0024] in, These are all preset fixed parameters.

[0025] It should be further explained that, in the specific implementation process, a constant evaluation model is constructed according to the climbing index and the constant index of different analysis cycles. The process of using the constant evaluation model to obtain the recommended exercise duration of the user includes: Setting a constant index threshold, comparing the constant index of each analysis period in the same heart rate data curve graph with the set constant index threshold in the order of numbers, and marking the analysis period whose constant index is greater than or equal to the constant index threshold as an abnormal period; Count the number of abnormal cycles in the same heart rate data curve graph, mark the number corresponding to the kth abnormal cycle as the recommended cycle number, and use the same method to obtain the recommended cycle numbers in different heart rate data curve graphs; Generate a corresponding constant evaluation set according to the climbing index and the recommended cycle number in different heart rate data curve graphs, and divide the obtained constant evaluation set into a first training set and a first test set; Constructing a first convolutional neural network, using different climbing indexes in the first training set as input data of the first convolutional neural network, using the corresponding number of recommended cycles in the first training set as output data of the first convolutional neural network, and training the first convolutional neural network to obtain a first initial convolutional neural network; Using the first test set to perform model verification on the first initial convolutional neural network, outputting a first initial convolutional neural network that is less than or equal to a preset first test error threshold as a corresponding constant evaluation model; In the subsequent exercise process, the climbing index of different users is obtained in real time and input into the constructed constant evaluation model. The constant evaluation model is used to output the corresponding recommended cycle number, and the product of the recommended cycle number and the time length of the analysis cycle is used as the recommended exercise duration for the corresponding user.

[0026] It should be further explained that, in the specific implementation process, the process of setting the heart rate standard range and obtaining the user's abnormal points and their corresponding speed variation ranges by combining the heart rate data curve graph and the speed data curve graph includes: Set corresponding heart rate standard ranges for different users' real-time heart rate values. Within the heart rate standard range, it can ensure that the user's aerobic exercise is always in an effective range, so that the heart rate is not too high to cause physical discomfort, or the heart rate is too low to cause poor aerobic efficiency; In the heart rate data curve, all coordinate points whose real-time heart rate values ​​are higher than (not including or equal to) the heart rate standard range are regarded as too high abnormal points, all coordinate points whose real-time heart rate values ​​are lower than (not including or equal to) the heart rate standard range are regarded as too low abnormal points, and all coordinate points whose real-time heart rate values ​​are within (including or equal to) the heart rate standard range are regarded as normal points; Take any abnormal point as an example, including the abnormal point that is too high and the abnormal point that is too low, and obtain the real-time speed S corresponding to the abnormal point in the speed data curve graph. y , and the real-time speed S corresponding to the previous coordinate point q , obtain the velocity amplitude of the abnormal point, recorded as B;

[0027] Among them, t is the time interval between the abnormal point and its previous coordinate point, the speed variation includes positive and negative values. When it is a positive value, it corresponds to an abnormal point that is too high, indicating that the real-time heart rate value has an increasing trend. When it is a negative value, it corresponds to an abnormal point that is too low, indicating that the real-time heart rate value has a decreasing trend. The heart rate trend includes an increasing trend and a decreasing trend.

[0028] It should be further explained that, in the specific implementation process, the process of constructing the speed amplitude evaluation model according to the heart rate data, body mass index, and speed amplitude of different users includes: According to the real-time heart rate values ​​and speed variation amplitudes of different abnormal points and their corresponding heart rate trends, combined with the body mass index of the corresponding user, a variation amplitude evaluation model is generated, and the obtained variation amplitude evaluation set is divided into a second training set and a second test set; Constructing a second convolutional neural network, taking different real-time heart rate values, speed variation, and body mass index in the second training set as input data of the second convolutional neural network, taking the corresponding heart rate trend in the second training set as output data of the second convolutional neural network, and training the second convolutional neural network to obtain a second initial convolutional neural network; The second initial convolutional neural network is model verified using the second test set, and a second initial convolutional neural network with a value less than or equal to a preset second test error threshold is output as the corresponding amplitude variation evaluation model.

[0029] It should be further explained that, in the specific implementation process, the process of using the amplitude evaluation model to obtain the user's heart rate trend, generating the corresponding recommended exercise speed in real time and providing feedback includes: In the subsequent exercise process, abnormal points of different users and their corresponding real-time heart rate values ​​and speed variations are obtained in real time, combined with the body mass index of the corresponding users, and input into the constructed variation evaluation model, and the variation evaluation model is used to output the corresponding heart rate trend, including the upward trend and the downward trend; When it is an increasing trend, the user's current speed is obtained, and the current speed is lowered to generate a corresponding recommended exercise speed. When it is a decreasing trend, the user's current speed is obtained, and the current speed is increased to generate a corresponding recommended exercise speed. The generated recommended exercise speed is fed back to the user, and the user is advised to perform aerobic exercise at the recommended exercise speed until there are no abnormal points.

[0030] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent system for sports data collection and analysis, characterized in that: Includes the following modules: The data acquisition module is used to obtain the user's heart rate data and speed data, and to construct corresponding heart rate data curve graphs and speed data curve graphs respectively; A data analysis module is used to obtain the user's climbing points and their corresponding climbing indexes in the heart rate data curve graph, set the analysis period, obtain the constant indexes in different analysis periods, and generate the user's physical fitness index in combination with the speed data curve graph; A data evaluation module is used to construct a constant evaluation model based on the climbing index and the constant index of different analysis cycles, and use the constant evaluation model to obtain the recommended exercise duration of the user; The data monitoring module is used to set the standard range of heart rate and obtain the abnormal points of the user and their corresponding speed variation by combining the heart rate data curve graph and the speed data curve graph; The data optimization module is used to build a variation evaluation model based on the heart rate data, body mass index, and speed variation of different users, use the variation evaluation model to obtain the user's heart rate trend, generate the corresponding recommended exercise speed in real time and provide feedback.

2. An intelligent system for sports data collection and analysis according to claim 1, characterized in that: The process of obtaining heart rate data and speed data and constructing a heart rate data curve graph and a speed data curve graph includes: The heart rate data refers to the real-time heart rate value of the user from the beginning to the end of a single aerobic exercise, and the speed data refers to the real-time speed of the user from the beginning to the end of a single aerobic exercise; A corresponding heart rate data curve graph is constructed based on the acquired heart rate data, with the ordinate being the real-time heart rate value and the abscissa being the time node; a corresponding speed data curve graph is constructed based on the acquired speed data, with the ordinate being the real-time speed and the abscissa being the time node.

3. The intelligent system for sports data collection and analysis according to claim 2, characterized in that: The process of obtaining the user's climbing points and climbing index includes: Adjacent coordinate points in the heart rate data curve graph and the speed data curve graph are connected by smooth curves. In the heart rate data curve graph, the climbing angle between the tangent of each coordinate point and the positive direction of the y-axis is obtained in sequence according to the positive direction of the x-axis. When a climbing angle equal to 90 degrees appears for the first time, the corresponding coordinate point is taken as the climbing point, and the climbing index P of the user at the climbing point is obtained. z ; The real-time heart rate value at the climbing point is taken as the climbing heart rate value P x , the heart rate data curve graph is at the initial time C s The real-time heart rate value is taken as the initial heart rate value C x , the time node of the climbing point is taken as the climbing moment P s .

4. The intelligent system for sports data collection and analysis according to claim 3, characterized in that: The process of setting the analysis period, obtaining the constant index in different analysis periods, and generating the user's body mass index includes: In the heart rate data curve graph, the time nodes are divided into different analysis periods according to the order of the positive direction of the x-axis, and the divided analysis periods are numbered as i, i=1, 2, ..., n, where n is the number of analysis periods in the heart rate data curve graph; Mark the real-time heart rate values ​​corresponding to each coordinate point in a single analysis period in the order of time nodes, and record them as X j , j = 1, 2, ..., m, m is the number of all coordinate points in the analysis period, and the constant index H in the analysis period is obtained z ; Get the constant index H of each analysis period in the same heart rate data curve graph respectively zi , obtain the speed mean S in each analysis period in the speed data curve graph pi , obtain the user's body mass index T in the heart rate data curve graph z ; in, These are preset fixed parameters.

5. The intelligent system for sports data collection and analysis according to claim 4, characterized in that: The process of building a constant evaluation model and obtaining the user's recommended exercise duration includes: Setting a constant index threshold, comparing the constant index of each analysis cycle in the same heart rate data curve graph with the constant index threshold in the order of numbers, and marking the analysis cycle whose constant index is greater than or equal to the constant index threshold as an abnormal cycle; Counting the number of abnormal cycles in the same heart rate data curve graph, marking the number corresponding to the kth abnormal cycle as the recommended cycle number, and obtaining the recommended cycle numbers in different heart rate data curve graphs respectively; Generate a constant evaluation set according to the climbing index and the recommended cycle number in different heart rate data curve graphs, and divide the constant evaluation set into a first training set and a first test set; Constructing a first convolutional neural network, using different climbing indexes in the first training set as input data of the first convolutional neural network, using the corresponding number of recommended cycles in the first training set as output data of the first convolutional neural network, and training the first convolutional neural network to obtain a first initial convolutional neural network; Using the first test set to perform model verification on the first initial convolutional neural network, outputting the first initial convolutional neural network that is less than or equal to a preset first test error threshold as a constant evaluation model; In the subsequent exercise process, the climbing index of different users is obtained in real time and input into the constant evaluation model to output the corresponding recommended cycle number. The product of the recommended cycle number and the time length of the analysis cycle is used as the recommended exercise duration of the corresponding user.

6. The intelligent system for sports data collection and analysis according to claim 5, characterized in that: The process of setting the standard heart rate range and obtaining the user's abnormal points and speed variation includes: Set the heart rate standard range, and mark each coordinate point in the heart rate data curve as an abnormally high point, an abnormally low point, or a normal point according to the heart rate standard range; The abnormal points include too high abnormal points and too low abnormal points. The real-time speed S corresponding to a single abnormal point is obtained in the speed data curve diagram. y , and the real-time speed S corresponding to the previous coordinate point q , obtain the velocity amplitude B of the abnormal point; t is the time interval between the abnormal point and its previous coordinate point. The speed variation includes positive and negative values. When it is a positive value, it represents that the real-time heart rate value is on an upward trend. When it is a negative value, it represents that the real-time heart rate value is on a downward trend.

7. An intelligent system for sports data collection and analysis according to claim 6, characterized in that: The process of building a variable amplitude assessment model includes: According to the real-time heart rate values ​​and speed variation and heart rate trends of different abnormal points, combined with the body mass index of the corresponding user, a variation evaluation model is generated, and the variation evaluation set is divided into a second training set and a second test set; Constructing a second convolutional neural network, taking different real-time heart rate values, speed variation, and body mass index in the second training set as input data of the second convolutional neural network, taking the corresponding heart rate trend in the second training set as output data of the second convolutional neural network, and training the second convolutional neural network to obtain a second initial convolutional neural network; The second initial convolutional neural network is model verified using the second test set, and a second initial convolutional neural network with a value less than or equal to a preset second test error threshold is output as a variation evaluation model.

8. The intelligent system for sports data collection and analysis according to claim 7, characterized in that: The process of obtaining the user's heart rate trend, generating a recommended exercise speed in real time, and providing feedback includes: In the subsequent exercise process, abnormal points of different users and their corresponding real-time heart rate values ​​and speed variations are obtained in real time, and combined with the body mass index of the corresponding users, they are input into the variation evaluation model to output the corresponding heart rate trends, including increasing trends and decreasing trends; The user's current speed is obtained. When it is increasing, the current speed is lowered to generate a recommended movement speed. When it is decreasing, the current speed is increased to generate a recommended movement speed. The generated recommended movement speed is fed back to the user until there are no abnormal points.

Images