Method for synchronizing rowing posture and virtual rowing posture

Abstract

The invention provides a method for synchronizing a rowing posture and a virtual rowing posture. The method comprises the following steps: S1, acquiring rowing action video data by using data acquisition equipment; s2, identifying the rowing action change and paddle frequency of the user in the rowing action video; s3, the rowing power is calculated according to the rowing action video, and the virtual rowing action is synchronized according to the rowing power and the rowing action of the user; s4, the data acquisition equipment analyzes the synchronized virtual rowing actions and gives corresponding prompts and modification suggestions; by synchronously constructing the virtual rowing posture of the rowing posture of the trainer, the rowing posture of the trainer is displayed in real time, modification suggestions are given, and the trainer is helped to scientifically carry out rowing training.

Description

Technical Field

[0001] The present invention relates to the field of computers, and in particular to a method for synchronizing a rowing posture with a virtual rowing posture. Background Art

[0002] In recent years, with the implementation and continuous advancement of the National Fitness Program, more and more people have begun to understand sports and participate in sports and fitness, and the phenomenon of mass fitness has gradually become popular. Indoor rowing is deeply loved by fitness enthusiasts at home and abroad. However, since people are not equipped with private coaches when doing indoor rowing training, many beginners suffer sports injuries to the elbow joints, intervertebral discs, and shoulder joints due to incorrect rowing postures, which reduces their enthusiasm for exercise. The standard rowing posture can ensure the exercise effect without damaging the rower himself. In actual training, many trainees cannot master the essentials of rowing, or the force sequence is wrong, or the movement amplitude is not in place, or the force is uneven, which greatly reduces the effect of rowing training. The method described in the present invention constructs a synchronous virtual rowing posture, and recognizes the changes in joint angles during rowing, gives corresponding prompts and modification suggestions, realizes the online training guidance function, and improves the training effect.

[0003] Existing purely mechanical rowing machines do not have a guided rowing function, and rowing machines with a human-computer interaction interface mostly only display the standard rowing posture to the trainee through video playback. The present invention, based on the original technology, constructs a synchronized virtual rowing posture to determine the standard degree of the trainee's movements, and then gives modification suggestions to guide the trainee to carry out flexibility and coordination training scientifically. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a method for synchronizing a rowing posture with a virtual rowing posture.

[0005] To achieve the above object, the present invention adopts the following technical solutions: S1: Use data acquisition equipment to collect rowing action video data; The data acquisition device includes a camera module and a user rowing action storage module; The cameras include a first camera, a second camera, and a third camera; The first camera is aimed at the head and shoulders of the trainee to collect rowing action video data of the head and shoulders of the trainee to obtain first rowing action video data; The second camera is aimed at the trainee's legs to collect rowing action video data of the trainee's legs to obtain second rowing action video data; The third camera is aimed at the trainee's waist to collect rowing action video data of the trainee's waist to obtain third rowing action video data; The user rowing action storage module is used to store the rowing action video data, and the rowing action video data includes first rowing action video data, second rowing action video data and third rowing action video data; S2: Identify the rowing action changes and paddling frequency of the user in the rowing action video; The user's rowing motion includes elbow flexion, shoulder movement, spinal flexion, and hip translation; The user's rowing motion includes pulling and retracting the oar; The paddling action is manifested as elbow flexion, shoulder joint backward movement, spine backward flexion, or hip backward movement; The paddling action is realized by straightening the elbow joint, moving the shoulder joint forward, bending the spine forward, or moving the hip forward; The timing starts from the time when the user performs the rowing action, and the time is recorded as T. When the data acquisition device recognizes that the user starts the rowing action, the starting time is marked as T0. When the amplitude of the user's rowing action reaches the maximum, the corresponding time is marked as T1. When the amplitude of the user's rowing action reaches the minimum, the corresponding time is marked as T2. The specific steps include: S21: Mark the change in the bending angle of the user's elbow joint, marked as X; The elbow bending angle when the user starts to paddle is marked as X0, the angle when the elbow bending angle is the smallest is marked as X1, and the corresponding time is T1; When the user's elbow joint is bent at the maximum angle, it is marked as X2 and the corresponding time is T2.

[0006] S22: Mark the user's shoulder joint motion trajectory, marked as M; The shoulder joint position when the user starts to paddle is marked as M0, and the corresponding time is T0; when the user's shoulder joint position reaches the farthest point, the shoulder joint position is marked as M1, and the corresponding time is T1; when the user's shoulder joint position is closest, the shoulder joint position is marked as M2, and the corresponding time is T2.

[0007] S23: marking the user's spine bending motion trajectory, marked as Y; The spinal bending angle when the user starts to paddle is marked as Y0, and the corresponding time is T0; when the spinal bending angle of the user reaches the maximum, the spinal bending angle is marked as Y1, and the corresponding time is T1; when the spinal bending angle of the user is the minimum, the spinal bending angle is marked as Y2, and the corresponding time is T2.

[0008] S24: Mark the user's hip position, marked as Z; The hip position when the user starts to paddle is marked as Z0, and the corresponding time is T0; when the hip position of the user reaches the farthest point, the hip position is marked as Z1, and the corresponding time is T1; when the hip position of the user is closest, the hip position is marked as Z2, and the corresponding time is T2.

[0009] S25: restore the user's rowing action; The data acquisition device restores the user's paddle pulling / retracting action according to the user's elbow bending angle change X, the user's shoulder joint movement trajectory M, the user's spine bending movement trajectory Y, and the user's hip position Z. At the same time, the complete paddle pulling / retracting action is represented by a normalized action, that is, the entire paddle pulling / retracting action process is represented by a unified variable value, marked as L.

[0010] S26: Calculating the paddling frequency of the user; The paddling frequency is recorded as S, which is the number of times the user pulls / retracts the paddler in one minute. S=60 / (T2-T0).

[0011] S3: Calculate rowing power according to the rowing action video, and synchronize virtual rowing action according to the rowing power and the user's rowing action; It includes the following sub-steps: S31: Calculating rowing power P according to the rowing action of the user; Based on the established mathematical model, the propeller frequency S is converted into a success rate P; The mathematical model is a polynomial prediction model, which represents the relationship between the user parameter and the pulping frequency S and the power P; The user parameters are user height H and user weight W; The mathematical model is expressed as P=a*H 2 +b*H+c*W 2 +d*W+e*H*W+f*S 2 +g*S+h*S*H*W; Where P is power, H is the user's height, W is the user's weight, and S is the paddling frequency; By obtaining the rowing user's height H and weight W, the output power and paddling frequency data in the competition, and training the model based on this data, a set of parameters {a, b, c, de, f, g, h} can be obtained, and thus the above mathematical model can be obtained.

[0012] S32: Synchronously displaying a virtual rowing action on a data acquisition device based on the power P, the rowing frequency S and the rowing action of the user; The specific steps are as follows: S321: converting the power into a rowing action speed of the user; The value of V can be calculated by the formula P=F*V, where P is power, F is the force exerted by the user on the paddle when rowing, and V is the real-time rowing speed of the user.

[0013] S322: using the rowing action of the user and the oar frequency S to synchronously display a virtual rowing action; A rowing synchronization prediction and compensation algorithm is adopted to calculate the virtual rowing motion of the virtual rowing character in the data acquisition device using the rowing motion of the user and the oar frequency S, and the virtual rowing motion is used to update the rowing motion of the virtual rowing character, thereby achieving synchronization between the rowing motion of the user and the virtual rowing character; In order to enhance the accuracy of prediction compensation, the rowing synchronization prediction compensation algorithm is adopted; The specific contents of the rowing synchronization prediction compensation algorithm are as follows: Get the user's actual rowing action at time T+1; The rowing action of the user obtained at a certain time T is recorded as Ot=(S0, V0, L0), where S0 is the rowing frequency at time T, L0 is the normalized action of the complete rowing action at time T, and V0 is the speed of the rowing action at time T; When receiving the rowing action of the user at time T+1, it is recorded as Ot+1=(S1, V1, L1), where S1 is the rowing frequency at time T+1, L1 is the normalized action of the complete rowing action at time T+1, and V1 is the rowing action speed at time T+1; Get the virtual rowing action at time T; The virtual rowing action of the virtual character at time T is recorded as O't=(S0', V0', L0'), where S0' is the rowing frequency at time T in the data acquisition device, L0' is the normalized action of the complete rowing action at time T in the data acquisition device, and V0' is the speed of the rowing action at time T; According to the rowing action at time T+1 and the virtual rowing action at time T, a virtual rowing action at time T+1 is obtained; Using the virtual rowing action O't at time T in the virtual motion state and the user's rowing action Ot+1, the virtual rowing action of the virtual rowing character at time T+1 (recorded as O't+1) in the data acquisition device is simulated; The specific formula is as follows: O't+1=α*Ot+1+(1-α)O't; Where α represents the estimated confidence, which can be an empirical constant or a debugging value during experimental testing; It should be noted that the virtual rowing action O't is obtained by using the actual rowing action Ot at time t and the virtual rowing action O't-1 at time t-1, that is, O't=α*Ot+(1-α)O't-1; Thus, the virtual rowing action O't-1 is O't-1=α*Ot-1+(1-α)O't-2, and so on; As can be seen from the above, the virtual rowing action at the current moment is simulated by the virtual rowing action at the previous moment and the user's rowing action at the current moment; The initial values ​​at the beginning of the user's actual rowing action (i.e., when the user is not rowing) are set to an actual rowing speed of 0, an acceleration of 0, a stroke frequency of 0, and a rowing action in a stationary state. At this time, the initial virtual rowing action is also stationary; As the user starts rowing, the virtual rowing action at the previous moment and the user's rowing action at the current moment are used to continuously simulate the virtual rowing action at the current moment.

[0014] S323: Generate a synchronized virtual rowing action; Applying the virtual rowing action O't+1 to the virtual rowing character in the data acquisition device, that is, using the virtual rowing action O't+1 to update the rowing action at time t+1 in the data acquisition device; Since the virtual rowing action is obtained by continuously iterating the user's rowing action, the synchronization of the virtual rowing action and the user's rowing can be achieved; S4: The data acquisition device analyzes the synchronized virtual rowing action and gives corresponding prompts and modification suggestions; The data acquisition device also includes a standard action storage module and a comparison analysis module; The standard action storage module is used to store multiple groups of standard rowing action videos in a rowing action cycle; The comparison and analysis module uses a mathematical model to compare and analyze the user's rowing action with a standard rowing action video; S41: the data acquisition device compares and analyzes the synchronized virtual rowing action according to the standard rowing posture model; The comparison and analysis module compares the synchronized virtual rowing action with the standard rowing action model, and analyzes the difference between the virtual rowing action and the standard rowing action.

[0015] S42: The data acquisition device gives corresponding prompts and modification suggestions according to the comparison and analysis results; The data acquisition device gives modification suggestions on the speed and body movements of the user's rowing movements based on the analysis results obtained by the comparison and analysis module.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: during the rowing process, the present invention synchronously constructs a virtual rowing posture for the rowing posture of the trainee, displays the action image of the rowing process in real time through the human-computer interaction interface, and gives corresponding prompts and modification suggestions based on the change recognition of the rowing action during the rowing process, so as to guide the trainee to carry out flexibility and coordination training scientifically. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of a method for synchronizing a rowing posture with a virtual rowing posture according to Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0018] In order to provide a further understanding of the purpose, structure, features, and functions of the present invention, the following detailed description is given in conjunction with the embodiments.

[0019] S1: The user turns on the data acquisition device and uses the camera to record a video of his or her rowing action; The data acquisition device includes a camera module and a user rowing action storage module; The camera includes a first camera, a second camera, and a third camera The first camera is aimed at the head and shoulders of the trainee to collect rowing action video data of the head and shoulders of the trainee to obtain first rowing action video data; The second camera is aimed at the trainee's legs to collect rowing action video data of the trainee's legs to obtain second rowing action video data; The third camera is aimed at the trainee's waist to collect rowing action video data of the trainee's waist to obtain third rowing action video data; The user rowing action storage module is used to store the rowing action video data, and the rowing action video data includes first rowing action video data, second rowing action video data and third rowing action video data; S2: The data acquisition device identifies the rowing action changes and paddling frequency of the user in the rowing action video; The user's rowing motion includes elbow flexion, shoulder movement, spinal flexion, and hip translation; The user's rowing motion includes pulling and retracting the oar; The paddling action is manifested as elbow flexion, shoulder joint backward movement, spine backward flexion, or hip backward movement; The paddling action is realized by straightening the elbow joint, moving the shoulder joint forward, bending the spine forward, or moving the hip forward; The timing starts from the time when the user performs the rowing action, and the time is recorded as T. When the data acquisition device recognizes that the user starts the rowing action, the starting time is marked as T0. When the amplitude of the user's rowing action reaches the maximum, the corresponding time is marked as T1. When the amplitude of the user's rowing action reaches the minimum, the corresponding time is marked as T2. The specific steps include: S21: The data acquisition device marks the change in the bending angle of the user's elbow joint, which is marked as X; The elbow bending angle when the user starts to paddle is marked as X0, the angle when the elbow bending angle is the smallest is marked as X1, and the corresponding time is T1; When the user's elbow joint is bent at the maximum angle, it is marked as X2 and the corresponding time is T2.

[0020] S22: The data acquisition device marks the user's shoulder joint movement trajectory, marked as M; The shoulder joint position when the user starts to paddle is marked as M0, and the corresponding time is T0; when the user's shoulder joint position reaches the farthest point, the shoulder joint position is marked as M1, and the corresponding time is T1; when the user's shoulder joint position is closest, the shoulder joint position is marked as M2, and the corresponding time is T2.

[0021] S23: The data acquisition device marks the user's spine bending motion trajectory, which is marked as Y; The spinal bending angle when the user starts to paddle is marked as Y0, and the corresponding time is T0; when the spinal bending angle of the user reaches the maximum, the spinal bending angle is marked as Y1, and the corresponding time is T1; when the spinal bending angle of the user is the minimum, the spinal bending angle is marked as Y2, and the corresponding time is T2.

[0022] S24: The data acquisition device marks the position of the user's hips, marked as Z; The hip position when the user starts to paddle is marked as Z0, and the corresponding time is T0; when the hip position of the user reaches the farthest point, the hip position is marked as Z1, and the corresponding time is T1; when the hip position of the user is closest, the hip position is marked as Z2, and the corresponding time is T2.

[0023] S25: The data acquisition device restores the user's rowing action; The data acquisition device restores the user's paddle pulling / retracting action according to the user's elbow bending angle change X, the user's shoulder joint movement trajectory M, the user's spine bending movement trajectory Y, and the user's hip position Z. At the same time, the complete paddle pulling / retracting action is represented by a normalized action, that is, the entire paddle pulling / retracting action process is represented by a unified variable value, marked as L.

[0024] When the user performs rowing exercise, the key parts of the user's rowing action are recorded to accurately restore the user's pulling / retracting action of the oar. At the same time, a unified variable value L is used to represent the pulling / retracting action of the oar, which facilitates the calculation of the pulling / retracting action of the oar into the rowing synchronization prediction compensation algorithm to ensure the accuracy of the virtual rowing action.

[0025] S26: The data acquisition device calculates the paddling frequency of the user; The paddling frequency is recorded as S, which is the number of times the user pulls / retracts the paddler in one minute. S=60 / (T2-T0).

[0026] S3: The data acquisition device calculates the rowing power according to the rowing action video, and synchronizes the virtual rowing action according to the rowing power and the user's rowing action; It includes the following sub-steps: S31: Calculating rowing power P according to the rowing action of the user; Based on the established mathematical model, the propeller frequency S is converted into a success rate P; The mathematical model is a polynomial prediction model, which represents the relationship between the user parameter and the pulping frequency S and the power P; The user parameters are user height H and user weight W; The mathematical model is expressed as P=a*H 2 +b*H+c*W 2 +d*W+e*H*W+f*S 2 +g*S+h*S*H*W; Where P is power, H is the user's height, W is the user's weight, and S is the paddling frequency; By obtaining the rowing user's height H and weight W, the output power P and the rowing frequency S data in the competition, a set of parameters {a, b, c, de, f, g, h} can be obtained based on this data for model training, so that the above mathematical model can be obtained; By taking the user's body parameters into consideration through the above mathematical model, the power P of the user when rowing can be accurately calculated for different users.

[0027] S32: Synchronously displaying a virtual rowing action on a data acquisition device based on the rowing frequency S and the rowing action of the user; The specific steps are as follows: S321: converting the power into a rowing action speed of the user; The value of V can be calculated by the formula P=F*V, where P is power, F is the force exerted by the user on the paddle when rowing, and V is the real-time rowing speed of the user; S322: Displaying a virtual rowing action using the rowing action and the paddling frequency S of the user; A rowing synchronization prediction and compensation algorithm is adopted, and the user's rowing action and the oar frequency S are used to calculate the virtual rowing action of the virtual rowing character in the data acquisition device, and the virtual rowing action is used to update the rowing action of the virtual rowing character, thereby achieving synchronization between the user's rowing action and the virtual rowing action of the virtual rowing character.

[0028] In order to enhance the accuracy of prediction compensation, the rowing synchronous prediction compensation algorithm is adopted.

[0029] The specific contents of the rowing synchronization prediction compensation algorithm are as follows: Get the user's actual rowing action at time T+1.

[0030] The rowing action of the user obtained at a certain time T is recorded as Ot=(S0, V0, L0), where S0 is the rowing frequency at time T, L0 is the normalized action of the complete rowing action at time T, and V0 is the speed of the rowing action at time T; When receiving the rowing action of the user at time T+1, it is recorded as Ot+1=(S1, V1, L1), where S1 is the rowing frequency at time T+1, L1 is the normalized action of the complete rowing action at time T+1, and V1 is the rowing action speed at time T+1; Obtaining the virtual rowing action at time T in the data acquisition device; The virtual rowing action of the virtual character at time T in the data acquisition device is recorded as O't=(S0', V0', L0'), where S0' is the rowing frequency at time T in the data acquisition device, L0' is the normalized action of the complete rowing action at time T in the data acquisition device, and V0' is the speed of the rowing action at time T; According to the rowing action at time T+1 and the virtual rowing action at time T, a virtual rowing action at time T+1 is obtained; Using the virtual rowing action O't at time T in the virtual motion state and the user's rowing action Ot+1, the virtual rowing action of the virtual rowing character at time T+1 (recorded as O't+1) in the data acquisition device is simulated; The specific formula used is: O't+1=α*Ot+1+(1-α)O't.

[0031] Where α represents the estimated confidence, which can be an empirical constant or a debugging value during experimental testing; It should be noted that the virtual rowing action O't is obtained by using the actual rowing action Ot at time t and the virtual rowing action O't-1 at time t-1, that is, O't=α*Ot+(1-α)O't-1; Thus, the virtual rowing action O't-1 is O't-1=α*Ot-1+(1-α)O't-2, and so on.

[0032] As can be seen from the above, the virtual rowing action at the current moment is simulated by the virtual rowing action at the previous moment and the user's rowing action at the current moment; The initial values ​​at the beginning of the user's actual rowing action (i.e., when the user is not rowing) are set to an actual rowing speed of 0, an acceleration of 0, a stroke frequency of 0, and a rowing action in a stationary state. At this time, the initial virtual rowing action is also stationary; As the user starts rowing, the virtual rowing action at the previous moment and the user's rowing action at the current moment are used to continuously simulate the virtual rowing action at the current moment.

[0033] S323: Generate a synchronized virtual rowing action; Applying the virtual rowing action O't+1 to the virtual rowing character in the data acquisition device, that is, using the virtual rowing action O't+1 to update the rowing action at time t+1 in the data acquisition device; Since the virtual rowing action is obtained by continuously iterating the user's rowing action, the synchronization of the virtual rowing action and the user's rowing can be achieved Using the virtual rowing action O't+1 calculated at this time can maximize the error smoothing of the collected data, reduce the synchronization difference, and improve the experience synchronization.

[0034] S4: The data acquisition device analyzes the synchronized virtual rowing action and gives corresponding prompts and modification suggestions; The data acquisition device also includes a standard action storage module and a comparison analysis module; The standard action storage module is used to store multiple groups of standard rowing action videos in a rowing action cycle; The comparison and analysis module uses a mathematical model to compare and analyze the user's rowing action with a standard rowing action video; S41: the data acquisition device compares and analyzes the synchronized virtual rowing action according to the standard rowing posture model; The comparison and analysis module compares the synchronized virtual rowing action with the standard rowing action model, and analyzes the difference between the virtual rowing action and the standard rowing action.

[0035] S42: The data acquisition device gives corresponding prompts and modification suggestions according to the comparison and analysis results; The data acquisition device gives modification suggestions on the speed and body movements of the user's rowing movements based on the analysis results obtained by the comparison and analysis module.

[0036] The present invention has been described by the above-mentioned relevant embodiments, however, the above-mentioned embodiments are only examples for implementing the present invention. It must be pointed out that the disclosed embodiments do not limit the scope of the present invention. On the contrary, changes and modifications made without departing from the spirit and scope of the present invention are all within the scope of patent protection of the present invention.

Claims

1. A method for synchronizing a rowing posture with a virtual rowing posture, characterized in that: The following steps are involved: S1: Use data acquisition equipment to collect rowing action video data; The data acquisition device includes a camera module and a user rowing action storage module; The camera includes a first camera, a second camera, and a third camera The first camera is aimed at the head and shoulders of the trainee to collect rowing action video data of the head and shoulders of the trainee to obtain first rowing action video data; The second camera is aimed at the trainee's legs to collect rowing action video data of the trainee's legs to obtain second rowing action video data; The third camera is aimed at the trainee's waist to collect rowing action video data of the trainee's waist to obtain third rowing action video data; The user rowing action storage module is used to store the rowing action video data, and the rowing action video data includes first rowing action video data, second rowing action video data and third rowing action video data; S2: Identify the rowing action changes and paddling frequency of the user in the rowing action video; The timing starts from the time when the user performs the rowing action, and the time is recorded as T. When the data acquisition module recognizes that the user starts the rowing action, the starting time is marked as T0. When the amplitude of the user's rowing action reaches the maximum, the corresponding time is marked as T1. When the amplitude of the user's rowing action reaches the minimum, the corresponding time is marked as T2. The specific steps include: S21: The data acquisition device marks the change in the bending angle of the user's elbow joint, which is marked as X; The elbow bending angle when the user starts to paddle is marked as X0, the angle when the elbow bending angle is the smallest is marked as X1, and the corresponding time is T1; When the user's elbow flexion angle is at its maximum, it is marked as X2, and the corresponding time is T2; S22: The data acquisition device marks the user's shoulder joint movement trajectory, marked as M; The shoulder joint position when the user starts to paddle is marked as M0, and the corresponding time is T0; when the user's shoulder joint position reaches the farthest point, the shoulder joint position is marked as M1, and the corresponding time is T1; when the user's shoulder joint position is closest, the shoulder joint position is marked as M2, and the corresponding time is T2; S23: The data acquisition device marks the user's spine bending motion trajectory, which is marked as Y; The spinal bending angle when the user starts to paddle is marked as Y0, and the corresponding time is T0; when the spinal bending angle of the user reaches the maximum, the spinal bending angle is marked as Y1, and the corresponding time is T1; when the spinal bending angle of the user is the minimum, the spinal bending angle is marked as Y2, and the corresponding time is T2; S24: The data acquisition device marks the position of the user's hips, marked as Z; The hip position when the user starts to paddle is marked as Z0, and the corresponding time is T0; when the hip position of the user reaches the farthest point, the hip position is marked as Z1, and the corresponding time is T1; when the hip position of the user is closest, the hip position is marked as Z2, and the corresponding time is T2; S25: The data acquisition device restores the user's rowing action; The data acquisition device restores the user's paddle pulling / retracting action according to the user's elbow bending angle change X, the user's shoulder joint motion trajectory M, the user's spine bending motion trajectory Y, and the user's hip position Z. At the same time, the complete paddle pulling / retracting action is represented by a normalized action, that is, the entire paddle pulling / retracting action process is represented by a unified variable value, marked as L; S26: Calculating the paddling frequency of the user; The paddling frequency is recorded as S, which is the number of times the user pulls / retracts the paddler in one minute. S=60 / (T2-T0); S3: Calculate rowing power according to the rowing action video, and synchronize virtual rowing action according to the rowing power and the user's rowing action; It includes the following sub-steps: S31: Calculating rowing power P according to the rowing action of the user; Based on the established mathematical model, the propeller frequency S is converted into a success rate P; The mathematical model is a polynomial prediction model, which represents the relationship between the user parameter and the pulping frequency S and the power P; The user parameters are user height H and user weight W; The mathematical model is expressed as P=a*H 2 +b*H+c*W 2 +d*W+e*H*W+f* S 2 +g*S+h*S*H*W; Where P is power, H is the user's height, W is the user's weight, and S is the paddling frequency; By obtaining the rowing user's height H and weight W, the output power and rowing frequency data in the competition, and training the model based on this data, a set of parameters {a, b, c, de, f, g, h} can be obtained, and thus the above mathematical model can be obtained; S32: Synchronously displaying a virtual rowing action on a data acquisition device based on the rowing frequency S and the rowing action of the user; The specific steps are as follows: S321: converting the power into a rowing action speed of the user; The value of V can be calculated by the formula P=F*V, where P is power, F is the force exerted by the user on the paddle when rowing, and V is the real-time rowing speed of the user; S322: using the rowing action of the user and the oar frequency S to synchronously display a virtual rowing action; A rowing synchronization prediction and compensation algorithm is adopted to calculate the virtual rowing motion of the virtual rowing character in the data acquisition device using the rowing motion of the user and the oar frequency S, and the virtual rowing motion is used to update the rowing motion of the virtual rowing character, thereby achieving synchronization between the rowing motion of the user and the virtual rowing character; In order to enhance the accuracy of prediction compensation, the rowing synchronization prediction compensation algorithm is adopted; S323: Generate a synchronized virtual rowing action; The rowing action at time t+1 in the data acquisition device is updated to the virtual rowing action at time t+1, so as to synchronize the rowing action in the data acquisition device with the actual rowing action; Applying the virtual rowing action O't+1 to the virtual rowing character in the data acquisition device, that is, using the virtual rowing action O't+1 to update the rowing action at time t+1 in the data acquisition device; Since the virtual rowing action is obtained by continuously iterating the user's rowing action, the synchronization of the virtual rowing action and the user's rowing can be achieved; S4: The data acquisition device analyzes the synchronized virtual rowing action and gives corresponding prompts and modification suggestions; The data acquisition device also includes a standard action storage module and a comparison analysis module; The standard action storage module is used to store multiple groups of standard rowing action videos in a rowing action cycle; The comparison and analysis module uses a mathematical model to compare and analyze the user's rowing action with a standard rowing action video; S41: the data acquisition device compares and analyzes the synchronized virtual rowing action according to the standard rowing posture model; The comparison and analysis module compares the synchronized virtual rowing action with the standard rowing action model to analyze the difference between the virtual rowing action and the standard rowing action; S42: The data acquisition device gives corresponding prompts and modification suggestions according to the comparison and analysis results; The data acquisition device gives modification suggestions on the speed and body movements of the user's rowing movements based on the analysis results obtained by the comparison and analysis module.

2. The method for synchronizing a rowing action with a virtual rowing action according to claim 1, characterized in that: In step S31, the mathematical model is expressed as P=a*H 2 +b*H+c*W 2 +d*W+e*H*W+f*S 2 +g*S+h*S*H*W; Where P is power, H is the user's height, W is the user's weight, and S is the paddling frequency; By obtaining the rowing user's height H and weight W, the output power and paddling frequency data in the competition, and training the model based on this data, a set of parameters {a, b, c, de, f, g, h} can be obtained, and thus the above mathematical model can be obtained.

3. The method for synchronizing a rowing action with a virtual rowing action according to claim 1, characterized in that: In step S321, the rowing synchronization prediction compensation algorithm is specifically as follows: Get the user's actual rowing action at time T+1; The rowing action of the user obtained at a certain time T is recorded as Ot=(S0, V0, L0), where S0 is the rowing frequency at time T, L0 is the normalized action of the complete rowing action at time T, and V0 is the speed of the rowing action at time T; When receiving the rowing action of the user at time T+1, it is recorded as Ot+1=(S1, V1, L1), where S1 is the rowing frequency at time T+1, L1 is the normalized action of the complete rowing action at time T+1, and V1 is the rowing action speed at time T+1; Obtaining the virtual rowing action at time T in the data acquisition device; The virtual rowing action of the virtual character at time T in the data acquisition device is recorded as O't=(S0', V0', L0'), where S0' is the rowing frequency at time T in the data acquisition device, L0' is the normalized action of the complete rowing action at time T in the data acquisition device, and V0' is the speed of the rowing action at time T; According to the rowing action at time T+1 and the virtual rowing action at time T, a virtual rowing action at time T+1 is obtained; Using the virtual rowing action O't at time T in the virtual motion state and the user's rowing action Ot+1, the virtual rowing action of the virtual rowing character at time T+1 (recorded as O't+1) in the data acquisition device is simulated; The specific formula is as follows: O't+1=α*Ot+1+(1-α)O't; Where α represents the estimated confidence, which can be an empirical constant or a debugging value during experimental testing; It should be noted that the virtual rowing action O't is obtained by using the actual rowing action Ot at time t and the virtual rowing action O't-1 at time t-1, that is, O't=α*Ot+(1-α)O't-1; Thus, the virtual rowing action O't-1 is O't-1=α*Ot-1+(1-α)O't-2, and so on; The initial values ​​at the beginning of the user's actual rowing action (i.e., when the user is not rowing) are set to the actual rowing speed of 0, the acceleration of 0, the rowing frequency S of 0, and the rowing action is in a static state. At this time, the initial virtual rowing action is also static; As the user starts rowing, the virtual rowing action at the previous moment and the user's rowing action at the current moment are used to continuously simulate the virtual rowing action at the current moment.

Images