Special place video fusion supervision method based on digital twinning

Abstract

The invention relates to the technical field of digital twinning, and particularly discloses a special place video fusion supervision method based on digital twinning, and the method comprises the steps: building a digital twinning 3D scene through wearing a human body detection device for a student in a training room and combining with a shooting image of a streaming media pull-stream camera, through combining body parameters of different students in the motion process, body states of the different students in the motion process are analyzed, and then motion risk coefficients of the different students in the motion process are further analyzed in combination with motion state data of the students in camera pictures of multiple groups of streaming media pull stream cameras. The exercise state and the exercise risk of the trainee are analyzed based on diversified data, the accuracy of the physical state analysis result of the trainee can be improved, the situation that the coach makes a wrong exercise plan for the trainee due to analysis result errors caused by single data is avoided, and therefore the situation that the trainee is injured is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twins, and in particular to a video fusion supervision method for special places based on digital twins. Background Art

[0002] Digital twins make full use of physical models, sensor updates, operation history and other data, integrate multi-disciplinary, multi-physical quantity, multi-scale, and multi-probability simulation processes, complete mapping in virtual space, and thus reflect the entire life cycle of the corresponding physical equipment.

[0003] Through digital twin technology, the activities of people in a space can be mapped in a 3D scene. Therefore, based on this technology, the human body model of students taking classes in the sports training room can be established through digital twin technology, and the students can be equipped with human body detection equipment to monitor their physical condition and analyze their training status, thereby assisting coaches in understanding the status data of different students and facilitating adjustments to the students' exercise plans.

[0004] In the prior art, the adjustment of the exercise plan for students is generally based on having the students wear human body detection equipment and analyzing the students' physical condition by monitoring it. However, the commonly used detection equipment has low sensitivity, and each person's tolerance is different. Relying solely on physical condition data for analysis may lead to errors in the analysis results, thereby causing the coach to formulate the wrong exercise plan for the students, which in turn may cause the students to get injured. Summary of the invention

[0005] The purpose of this invention is to provide a special place video fusion supervision method based on digital twins to solve the following technical problems:

[0006] How to improve the accuracy of analysis results of students' physical condition.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A video fusion supervision method for special places based on digital twins, the method comprising:

[0009] S1: A digital twin 3D scene is established by having trainees wear human body detection equipment in the training room and combining it with the video footage from the streaming media camera;

[0010] S2: Collect the physical parameters of different trainees in the training room during exercise through human body detection equipment;

[0011] S3: Analyze the physical conditions of different students during exercise based on their physical parameters during exercise, and decide whether to issue an early warning based on the analysis results;

[0012] S4: combining the exercise status data and physical status data of the trainees in the camera images of multiple streaming media cameras, analyzing the exercise risk coefficients of different trainees during exercise, and deciding whether to issue an early warning based on the analysis results;

[0013] S5: Based on the physical condition data and sports risk coefficient of different students during the exercise process, the physical condition of different students after completing a set of exercises is analyzed, and a decision is made whether to issue a warning based on the analysis results.

[0014] Furthermore, the human body detection device includes a body state monitoring bracelet, which is used to collect body data of the human body during exercise, including blood pressure, blood oxygen concentration, heart rate and respiratory rate;

[0015] The trainee's motion state data in the video footage of the streaming media camera includes data on changes in relative positions of key points of the human body during exercise, rest data, and body balance data, wherein the key points of the human body are the limbs.

[0016] Furthermore, the process of analyzing the physical states of different trainees during exercise in S3 includes:

[0017] By formula Calculate the influence coefficient r of the physical condition of the a-th student at the i-th time point ai ;

[0018] Among them, a is any student, i is any time point after the start of exercise, xl ai is the heart rate of the a-th student at the i-th time point, xl y is the preset heart rate, hx ai is the respiratory rate of the ath student at the ith time point, hx y is the preset respiratory rate, xy ai is the blood oxygen saturation of the ath student at the i-th time point, xy y is the preset blood oxygen saturation, ρ is the error influence coefficient, which is set according to empirical fitting, and f t (x) is the defined function. If x>1, then f t (x) = x, if x ≤ 1, then f t (x) = 1, xyz ai is the blood pressure value of the a-th student at the i-th time point, xyz y is the preset blood pressure value, xyz b for xyz ai The standard value of .

[0019] Furthermore, the process of analyzing the physical states of different trainees during exercise in S3 also includes:

[0020] By calculating the influence coefficient r of the physical condition of the a-th student at the i-th time point ai The threshold interval of the influence coefficient of the preset physical condition [r 1 , r 2 ] for comparison;

[0021] If r ai <r 1 , it is judged that the current exercise load of the a-th student is low and the exercise effect is poor;

[0022] If r ai ∈[r 1 , r 2 ], judging that the current exercise load of the a-th student is medium to high, and the exercise effect is good;

[0023] If r ai >r 2 , it is judged that the current exercise load of the ath student is too heavy and he cannot continue to exercise, and an early warning is issued.

[0024] Furthermore, the process of analyzing the sports risk coefficients of different trainees during exercise in S4 includes:

[0025] By formula Calculate the sports risk coefficient p of the a-th student when doing the m-th action am ;

[0026] Among them, m is any action process, py am is the position offset distance of all key points of the a-th student during the m-th action, py y is the preset offset distance, py b for py am The standard value, hd am is the number of shakes of the a-th student during the m-th action, hd b For hd am The standard value of xi am is the rest time of the a-th student after doing the m-th action, xi y is the preset rest time, r ami is the influence coefficient of the body condition of the a-th student at the i-th time point during the m-th action.

[0027] Furthermore, the process of analyzing the sports risk coefficients of different trainees during exercise in S4 also includes:

[0028] By calculating the sports risk coefficient p of the a-th student when doing the m-th action am Compared with the preset motion risk coefficient threshold p 01 Make a comparison;

[0029] If p am ≥p 01 , judge that the completion degree of the a-th student is low when doing the action, and there is a movement risk, and issue a warning in time;

[0030] If p am <p 01 , it is judged that the completion degree of the ath student is high when performing the action and there is no movement risk, so there is no need to issue a warning.

[0031] Furthermore, the process of analyzing the physical conditions of different trainees after completing a set of exercises in S5 includes:

[0032] The physical condition influence coefficient r of the a-th student at the i-th time point obtained by real-time calculation ai , establish the change curve of the influence coefficient of the a-th student's physical condition β a (t);

[0033] And through the formula Calculate the state influence coefficient δ of the ath student completing a set of movements a ;

[0034] Where n is the total number of actions in a set of movements, For all p am The average value of t1 is the start time point of a group of movements, and t2 is the end time point of a group of movements.

[0035] Furthermore, the process of analyzing the physical conditions of different trainees after completing a set of exercises in S5 also includes:

[0036] By calculating the state influence coefficient δ of the a-th student completing a set of movements a And the preset state influence coefficient δ 01 Thresholds are compared;

[0037] If δ a ≤δ 01 , judging that the student is in good physical condition and has high tolerance after completing a set of exercises, and can continue to exercise;

[0038] If δ a ≥δ 01 , judge that the student’s physical condition is poor and tolerance is low after completing a set of exercises, and that he cannot continue exercising, and issue a warning.

[0039] Beneficial effects of the present invention:

[0040] (1) The present invention analyzes the physical conditions of different students during exercise by combining the physical parameters of different students during exercise, and then further analyzes the exercise risk coefficients of different students during exercise by combining the exercise status data of the students in the camera images of multiple groups of streaming media cameras. The exercise status and exercise risk of the students are analyzed based on diversified data, which can improve the accuracy of the analysis results of the students' physical conditions, avoid errors in the analysis results due to single data, and further avoid the situation where the coach makes wrong exercise plans for the students, thereby avoiding injuries to the students.

[0041] (2) The present invention calculates the physical condition influence coefficient r of the a-th student at the i-th time point ai The threshold interval of the influence coefficient of the preset physical condition [r 1 , r 2 ] for comparison. Through this comparison method, it is possible to make an accurate judgment on the exercise load of different students at different time points, and further conduct a preliminary analysis of the effect of their exercise training based on the judgment results of the exercise load. The analysis results can be used as diversified data to provide multi-dimensional data support for the subsequent analysis of the students' physical condition data, thereby improving the accuracy of the analysis results of the students' physical condition.

[0042] (3) The present invention calculates the sports risk coefficient p of the a-th student when performing the m-th action. am Compared with the preset motion risk coefficient threshold p 01 Through this comparison method, an accurate judgment can be made on the degree of completeness of the a-th student in doing the m-th action and whether there is any sports risk. Moreover, since the data is corrected based on the influence coefficient of the a-th student's physical condition at the ith time point during the m-th action, the accuracy of the judgment result can be further improved, thereby providing accurate data support for the subsequent analysis of the physical conditions of different students to ensure the accuracy of the analysis results.

[0043] (4) The present invention uses the state influence coefficient δ of the a-th student completing a set of movements a And the preset state influence coefficient δ 01 Through this comparison method, combined with diversified data support, it is possible to make an accurate judgment on the student's physical condition after completing a set of exercises, thereby improving the accuracy of the analysis results of the student's physical condition and avoiding relying on a single data for analysis, which may lead to errors in the analysis results, and then cause the coach to formulate wrong exercise plans for the students, resulting in injuries to the students. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The present invention will be further described below in conjunction with the accompanying drawings.

[0045] Figure 1 It is a flow chart of a video fusion supervision method for special places based on digital twins in the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] See also Figure 1 As shown, in one embodiment, the present application provides a special place video fusion supervision method based on digital twins, the method comprising:

[0048] S1: A digital twin 3D scene is established by having trainees wear human body detection equipment in the training room and combining it with the video footage from the streaming media camera;

[0049] S2: Collect the physical parameters of different trainees in the training room during exercise through human body detection equipment;

[0050] S3: Analyze the physical conditions of different students during exercise based on their physical parameters during exercise, and decide whether to issue an early warning based on the analysis results;

[0051] S4: combining the exercise status data and physical status data of the trainees in the camera images of multiple streaming media cameras, analyzing the exercise risk coefficients of different trainees during exercise, and deciding whether to issue an early warning based on the analysis results;

[0052] S5: Based on the physical condition data and sports risk coefficient of different students during exercise, analyze the physical condition of different students after completing a set of exercises, and decide whether to issue a warning based on the analysis results;

[0053] Through the above technical scheme, this embodiment provides a video fusion supervision method for special places based on digital twins. First, by having the trainees in the training room wear human body detection equipment and combining the camera footage of the streaming media camera, a digital twin 3D scene is established. Then, the body parameters of different trainees in the training room during exercise are collected through the human body detection equipment. In combination with the body parameters of different trainees during exercise, the physical state of different trainees during exercise is analyzed, and a decision is made based on the analysis results whether to issue an early warning. Then, the motion state data and body state data of the trainees in the camera footage of multiple groups of streaming media cameras are combined to analyze the motion risk coefficients of different trainees during exercise, and a decision is made based on the analysis results whether to issue an early warning. Finally, based on the body state data and motion risk coefficients of different trainees during exercise, the body state of different trainees after a group of exercises is analyzed, and a decision is made based on the analysis results whether to issue an early warning.

[0054] Through such a setting, the physical conditions of different students during exercise are analyzed by combining the physical parameters of different students during exercise, and then the exercise status data of the students in the camera images of multiple sets of streaming media cameras are further combined to analyze the exercise risk coefficients of different students during exercise. By analyzing the students' exercise status and exercise risk based on diversified data, the accuracy of the students' physical condition analysis results can be improved, avoiding errors in the analysis results due to single data, and then causing the coach to make wrong exercise plans for students, thereby avoiding injuries to students.

[0055] The human body detection device includes a body state monitoring bracelet, which is used to collect body data of the human body during exercise, including blood pressure, blood oxygen concentration, heart rate and respiratory rate;

[0056] The exercise state data of the trainee in the video footage of the streaming media camera includes the change data of the relative positions of the key points of the human body during the exercise, the rest data and the body balance data, wherein the key points of the human body are the limbs;

[0057] Through the above technical scheme, this embodiment provides the body data of the human body during exercise collected by the body status monitoring bracelet, and the exercise status data of the trainee in the camera picture of the streaming media camera. By obtaining two sets of diversified data in different dimensions, accurate data support can be provided for the subsequent analysis of the trainee's physical condition, thereby improving the accuracy of the analysis results. In addition, multi-dimensional data can improve the sensitivity of the analysis data and further improve the accuracy of the analysis results.

[0058] The process of analyzing the physical states of different trainees during exercise in S3 includes:

[0059] By formula Calculate the influence coefficient r of the physical condition of the a-th student at the i-th time point ai ;

[0060] Among them, a is any student, i is any time point after the start of exercise, xl ai is the heart rate of the a-th student at the i-th time point, xl y is the preset heart rate, hx ai is the respiratory rate of the ath student at the ith time point, hx y is the preset respiratory rate, xy ai is the blood oxygen saturation of the ath student at the i-th time point, xy y is the preset blood oxygen saturation, ρ is the error influence coefficient, which is set according to empirical fitting, and f t (x) is the defined function. If x>1, then f t (x) = x, if x ≤ 1, then f t (x) = 1, xyz ai is the blood pressure value of the a-th student at the i-th time point, xyz y is the preset blood pressure value, xyz b for xyz ai The standard value of the above standard value can be selected and set according to the allowable error in the empirical data;

[0061] Through the above technical solution, this embodiment provides the physical condition influence coefficient r of the a-th student at the i-th time point ai , can be obtained by formula It is calculated that, obviously, the higher the heart rate, blood pressure and respiratory rate of the a-th student at the i-th time point, and the lower the blood oxygen saturation, the worse the physical condition of the a-th student at the i-th time point, indicating that the effect of his exercise is gradually increasing; conversely, the lower the heart rate, blood pressure and respiratory rate of the a-th student at the i-th time point, and the higher the blood oxygen saturation, the better the physical condition of the a-th student at the i-th time point, indicating that the effect of his exercise has not increased significantly. Through this calculation method, accurate data can be provided for the subsequent judgment of the exercise effect of the a-th student at the i-th time point, thereby improving the accuracy of the judgment result.

[0062] The process of analyzing the physical conditions of different trainees during exercise in S3 also includes:

[0063] By calculating the influence coefficient r of the physical condition of the a-th student at the i-th time point ai The threshold interval of the influence coefficient of the preset physical condition [r 1 , r 2] for comparison;

[0064] If r ai <r 1 , it is judged that the current exercise load of the a-th student is low and the exercise effect is poor;

[0065] If r ai ∈[r 1 , r 2 ], judging that the current exercise load of the a-th student is medium to high, and the exercise effect is good;

[0066] If r ai >r 2 , judge that the current exercise load of the ath student is too heavy and cannot continue to exercise, and issue a warning;

[0067] Through the above technical solution, this embodiment calculates the physical condition influence coefficient r of the a-th student at the i-th time point ai The threshold interval of the influence coefficient of the preset physical condition [r 1 , r 2 ] for comparison. Through this comparison method, it is possible to make an accurate judgment on the exercise load of different students at different time points, and further conduct a preliminary analysis of the effect of their exercise training based on the judgment results of the exercise load. The analysis results can be used as diversified data to provide multi-dimensional data support for the subsequent analysis of the students' physical condition data, thereby improving the accuracy of the analysis results of the students' physical condition.

[0068] The process of analyzing the sports risk coefficients of different trainees during exercise in S4 includes:

[0069] By formula Calculate the sports risk coefficient p of the a-th student when doing the m-th action am ;

[0070] Among them, m is any action process, py am is the position offset distance of all key points of the a-th student during the m-th action, py y is the preset offset distance, py b for py am The standard value can be selected and set according to the allowable error in the empirical data, hd am is the number of shakes of the a-th student during the m-th action, hd b For hd am The standard value can be selected and set according to the allowable error in the empirical data, xi am is the rest time of the a-th student after doing the m-th action, xi y is the preset rest time, rami is the influence coefficient of the body condition of the a-th student at the i-th time point during the m-th action;

[0071] Through the above technical solution, this embodiment provides the sports risk coefficient p of the a-th student when doing the m-th action am , can be obtained by formula It is obvious that when the position offset distance, shaking times and rest time of all key points of the a-th student in the process of doing the m-th action are more, the sports risk coefficient p of the a-th student in doing the m-th action is higher. am The larger the value, the worse the condition of the a-th student is when doing the m-th action. On the contrary, the smaller the position offset distance, shaking times and rest time of all key points of the a-th student when doing the m-th action, the lower the sports risk coefficient p of the a-th student when doing the m-th action. am The smaller it is, it means that the a-th student is in good condition when doing the m-th action. Through this calculation method, accurate data can be provided for the subsequent judgment of whether there is sports risk for the a-th student, thereby improving the accuracy of the judgment result.

[0072] The process of analyzing the sports risk coefficients of different trainees during the sports process in S4 also includes:

[0073] By calculating the sports risk coefficient p of the a-th student when doing the m-th action am Compared with the preset motion risk coefficient threshold p 01 Make a comparison;

[0074] If p am ≥p 01 , judge that the completion degree of the a-th student is low when doing the action, and there is a movement risk, and issue a warning in time;

[0075] If p am <p 01 , it is judged that the completion degree of the a-th student is high when doing the action, and there is no movement risk, so there is no need to issue a warning;

[0076] Through the above technical solution, this embodiment calculates the sports risk coefficient p of the a-th student when doing the m-th action. am Compared with the preset motion risk coefficient threshold p 01Through this comparison method, an accurate judgment can be made on the degree of completeness of the a-th student in doing the m-th action and whether there is any sports risk. Moreover, since the data is corrected based on the influence coefficient of the a-th student's physical condition at the ith time point during the m-th action, the accuracy of the judgment result can be further improved, thereby providing accurate data support for the subsequent analysis of the physical conditions of different students to ensure the accuracy of the analysis results.

[0077] The process of analyzing the physical conditions of different trainees after completing a set of exercises in S5 includes:

[0078] The physical condition influence coefficient r of the a-th student at the i-th time point obtained by real-time calculation ai , establish the change curve of the influence coefficient of the a-th student's physical condition β a (t);

[0079] And through the formula Calculate the state influence coefficient δ of the ath student completing a set of movements a ;

[0080] Where n is the total number of actions in a set of movements, For all p am The average value of , t1 is the start time point of a group of movements, and t2 is the end time point of a group of movements;

[0081] Through the above technical solution, this embodiment provides the state influence coefficient δ of the a-th student completing a set of exercises a , can be obtained by formula Calculated, through this calculation method, combined with the physical condition influence coefficient r of the a-th student at the i-th time point ai The state influence coefficient δ of completing a set of exercises with the a-th student a After obtaining two sets of diversified data, the accuracy of the calculation results of the state influence coefficient of the a-th student completing a set of exercises can be improved.

[0082] The process of analyzing the physical conditions of different trainees after completing a set of exercises in S5 also includes:

[0083] By calculating the state influence coefficient δ of the a-th student completing a set of movements a And the preset state influence coefficient δ 01 Thresholds are compared;

[0084] If δ a ≤δ 01 , judging that the student is in good physical condition and has high tolerance after completing a set of exercises, and can continue to exercise;

[0085] If δa ≥δ 01 , judge that the student is in poor physical condition and has low tolerance after completing a set of exercises, and cannot continue to exercise, and issue a warning;

[0086] Through the above technical solution, this embodiment uses the state influence coefficient δ of the a-th student completing a set of exercises a And the preset state influence coefficient δ 01 Through this comparison method, combined with diversified data support, it is possible to make an accurate judgment on the student's physical condition after completing a set of exercises, thereby improving the accuracy of the analysis results of the student's physical condition and avoiding relying on a single data for analysis, which may lead to errors in the analysis results, and then cause the coach to formulate wrong exercise plans for the students, resulting in injuries to the students.

[0087] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A video fusion supervision method for special places based on digital twins, characterized in that: The method comprises: S1: A digital twin 3D scene is established by having trainees wear human body detection equipment in the training room and combining it with the video footage from the streaming media camera; S2: Collect the physical parameters of different trainees in the training room during exercise through human body detection equipment; S3: Analyze the physical conditions of different students during exercise based on their physical parameters during exercise, and decide whether to issue an early warning based on the analysis results; S4: combining the exercise status data and physical status data of the trainees in the camera images of multiple streaming media cameras, analyzing the exercise risk coefficients of different trainees during exercise, and deciding whether to issue an early warning based on the analysis results; S5: Based on the physical condition data and sports risk coefficient of different students during the exercise process, the physical condition of different students after completing a set of exercises is analyzed, and a decision is made whether to issue a warning based on the analysis results.

2. According to claim 1, a special place video fusion supervision method based on digital twins is characterized in that: The human body detection device includes a body state monitoring bracelet, which is used to collect body data of the human body during exercise, including blood pressure, blood oxygen concentration, heart rate and respiratory rate; The trainee's motion state data in the video footage of the streaming media camera includes data on changes in relative positions of key points of the human body during exercise, rest data, and body balance data, wherein the key points of the human body are the limbs.

3. According to claim 2, a special place video fusion supervision method based on digital twins is characterized in that: The process of analyzing the physical states of different trainees during exercise in S3 includes: By formula Calculate the influence coefficient r of the physical condition of the a-th student at the i-th time point ai ; Among them, a is any student, i is any time point after the start of exercise, xl ai is the heart rate of the a-th student at the i-th time point, xl y is the preset heart rate, hx ai is the respiratory rate of the ath student at the ith time point, hx y is the preset respiratory rate, xy ai is the blood oxygen saturation of the ath student at the i-th time point, xy y is the preset blood oxygen saturation, ρ is the error influence coefficient, which is set according to empirical fitting, and f t (x) is the defined function. If x>1, then f t (x) = x, if x ≤ 1, then f t (x) = 1, xyz ai is the blood pressure value of the a-th student at the i-th time point, xyz y is the preset blood pressure value, xyz b for xyz ai The standard value of .

4. According to claim 3, a special place video fusion supervision method based on digital twins is characterized in that: The process of analyzing the physical conditions of different trainees during exercise in S3 also includes: By calculating the influence coefficient r of the physical condition of the a-th student at the i-th time point ai Compare with the preset threshold interval [r1, r2] of the physical condition influence coefficient; If r ai <r1, it is judged that the current exercise load of the ath student is low and the exercise effect is poor; If r ai ∈[r1, r2], it is judged that the current exercise load of the a-th student is medium to high, and the exercise effect is good; If r ai >r2, it is judged that the current exercise load of the ath student is too heavy and he cannot continue to exercise, and a warning is issued.

5. According to claim 4, a special place video fusion supervision method based on digital twins is characterized in that: The process of analyzing the sports risk coefficients of different trainees during exercise in S4 includes: By formula Calculate the sports risk coefficient p of the a-th student when doing the m-th action am ; Among them, m is any action process, py am is the position offset distance of all key points of the a-th student during the m-th action, py y is the preset offset distance, py b for py am The standard value, hd am is the number of shakes of the a-th student during the m-th action, hd b For hd am The standard value of xi am is the rest time of the a-th student after doing the m-th action, xi y is the preset rest time, r ami is the influence coefficient of the body condition of the a-th student at the i-th time point during the m-th action.

6. According to claim 5, a special place video fusion supervision method based on digital twins is characterized in that: The process of analyzing the sports risk coefficients of different trainees during the sports process in S4 also includes: By calculating the sports risk coefficient pa of the a-th student when doing the m-th action m Compare with the preset motion risk coefficient threshold p01; If m ≥p01, judge that the completion degree of the a-th student is low when doing the action, and there is a movement risk, and issue a warning in time; If m <p01, it is judged that the completion degree of the ath student is high when doing the action, and there is no movement risk, so there is no need to issue a warning.

7. According to claim 6, a special place video fusion supervision method based on digital twins is characterized in that: The process of analyzing the physical conditions of different trainees after completing a set of exercises in S5 includes: The physical condition influence coefficient r of the a-th student at the i-th time point obtained by real-time calculation ai , establish the change curve of the influence coefficient of the physical condition of the a-th student β a (t); And through the formula Calculate the state influence coefficient δ of the ath student completing a set of movements a ; Where n is the total number of actions in a set of movements, For all p am The average value of t1 is the start time point of a group of movements, and t2 is the end time point of a group of movements.

8. According to claim 7, a special place video fusion supervision method based on digital twins is characterized in that: The process of analyzing the physical conditions of different trainees after completing a set of exercises in S5 also includes: By calculating the state influence coefficient δ of the a-th student completing a set of movements a And the preset state influence coefficient δ 01 Thresholds are compared; If δ a ≤δ 01 , judging that the student is in good physical condition and has high tolerance after completing a set of exercises, and can continue to exercise; If δ a ≥δ 01 , judge that the student’s physical condition is poor and tolerance is low after completing a set of exercises, and that he cannot continue exercising, and issue a warning.

Images