Body-building action standard auxiliary system based on machine vision

Abstract

The invention discloses a fitness action standard auxiliary system based on machine vision, and particularly relates to the technical field of machine vision, which comprises a monitoring area division module, a data acquisition module, a data preprocessing module, a data analysis module, a comprehensive analysis module, a judgment feedback module and a man-machine interaction module, the monitoring area division module is used for determining a single sub-time monitoring area; the data acquisition module is used for acquiring a first key information text of fitness actions; the data preprocessing module is used for performing data preprocessing on the fitness action first key information text to obtain a fitness action first key information value; the data analysis module is used for obtaining a second key information value of the fitness action; the comprehensive analysis module is used for constructing a comprehensive fitness action analysis model to obtain a comprehensive fitness action evaluation value; the judgment feedback module is used for constructing a fitness action comprehensive evaluation warning value and constructing a feedback adjustment model; the man-machine interaction module is used for providing a visual interface and improving comprehensiveness and accuracy of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of machine vision, and more specifically, the present invention relates to a fitness movement specification assistance system based on machine vision. Background Art

[0002] With the improvement of health awareness and the continuous expansion of the sports fitness market, the sports industry has also started its digital transformation. The rapid development of machine vision technology provides technical support for the development of fitness movement specification assistance systems. With the continuous maturity of technologies such as deep learning and computer vision, the capabilities of machine vision in aspects such as image recognition, object detection, and pose estimation have been significantly improved. As an important part of intelligent fitness products, the machine vision-based assistance system can promote the industrial upgrading and innovative development of the sports fitness industry.

[0003] Traditional fitness movement specification assistance systems specifically include an action demonstration and guidance module, an action monitoring and evaluation module, a personalized training plan module, a training record and tracking module, and an auxiliary function module. The action demonstration and guidance module is used to provide detailed steps and precautions for action execution to help fitness enthusiasts gradually learn and master correct fitness actions; the action monitoring and evaluation module is used to monitor the actions of fitness enthusiasts in real time and provide immediate feedback and corrective suggestions; the personalized training plan module is used to formulate personalized training; the training record and tracking module is used to record training information and evaluate training effects and progress; the auxiliary function module is used to provide the most suitable techniques related to fitness. It has significant advantages in providing accurate action guidance, improving training effects, enhancing safety and comfort, personalized customization, and facilitating learning and mastery.

[0004] However, in actual use, there are still some drawbacks. For example, the data collected is not comprehensive enough. The parameters collected by traditional fitness movement specification assistance systems for monitoring whether fitness actions meet the standards are not comprehensive enough to accurately reflect whether the target fitness actions meet the standards, resulting in inaccurate subsequent parameter analysis; it does not make a comprehensive evaluation of fitness actions and cannot provide real-time feedback on the situation of target fitness actions, resulting in the inability to take relevant solutions in a timely manner to adjust relevant actions. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a fitness movement specification assistance system based on machine vision, through the following solutions to solve the problems raised in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solution: A fitness movement standard auxiliary system based on machine vision, including a system operation database, a system central processor, and a user information terminal, further including a monitoring area division module, a data acquisition module, a data preprocessing module, a data analysis module, a comprehensive analysis module, a judgment feedback module, and a human-computer interaction module; The system operation database includes all data information of the fitness movement standard auxiliary system and collects in real time the data information output by each module; the system central processor is used to centrally control the information text instructions output by each module; the user information terminal is an information output device for receiving the fitness movement standard auxiliary system; The monitoring area division module is used to determine a single sub-time monitoring area, divide the time of the monitored target fitness movement into single sub-time monitoring areas in the manner of equal time lengths, and sequentially mark them as 1, 2,..., i,..., n; The data acquisition module is used to acquire the first key information text of the fitness movement, including a human body posture parameter acquisition unit and a movement quality parameter acquisition unit, and transmit the acquired data to the data preprocessing module; The data preprocessing module is used to perform data preprocessing on the first key information text of the fitness movement to obtain the first key information value of the fitness movement, including a human body posture parameter preprocessing unit and a movement quality parameter preprocessing unit, and transmit the result to the data analysis module; The data analysis module is used to perform data analysis on the first key information value of the fitness movement to obtain the second key information value of the fitness movement, including a human body posture data analysis unit and a movement quality data analysis unit, and transmit the result to the comprehensive analysis module; The comprehensive analysis module is used to construct a comprehensive analysis model of the fitness movement, process and analyze the second key information value of the fitness movement through the comprehensive analysis model of the fitness movement to obtain a comprehensive evaluation value of the fitness movement, and transmit the result to the judgment feedback module; The judgment feedback module is used to construct a comprehensive evaluation warning value of the fitness movement, judge whether the target fitness movement meets the standard, and construct a feedback adjustment model, and adopt corresponding adjustment methods according to the judgment result; The human-computer interaction module is used to provide a visual interface, display in real time the parameters in the target fitness process, and transmit them to the user information terminal.

[0007] Preferably, the first key information text of the fitness movement includes human body posture parameters and movement quality parameters. The human body posture parameters include movement posture parameters and biomechanical parameters. The movement posture parameters specifically include head position, shoulder position, shoulder joint angle, elbow joint angle, knee joint angle, and hip joint angle, which are respectively marked as h 1 、h 2 、θ 1, θ 2 , θ 3 and θ 4 , the biomechanical parameters specifically include the height of the center of body gravity, the movement speed, and the movement displacement, which are respectively marked as h 3 , v, and x; the movement quality parameters include movement performance parameters and human physiological parameters. The movement performance parameters specifically include the number of movements and the movement completion degree, which are respectively marked as n 1 and η, and the human physiological parameters specifically include heart rate, blood pressure, and blood oxygen saturation, which are respectively marked as R, P, and S.

[0008] Preferably, the human posture parameter preprocessing unit is used to establish a human posture parameter preprocessing model, import the collected movement posture parameters and biomechanical parameters into the human posture parameter preprocessing model, and calculate the movement posture evaluation value and the biomechanical evaluation value. The movement posture evaluation value is specifically expressed as: , y 1 represents the movement posture evaluation value in the target fitness process, Δh represents the comprehensive height difference in the target fitness process, , h 1i represents the head position in the i-th sub-time monitoring area in the target fitness process, h 1i标 represents the standard head position in the i-th sub-time monitoring area in the target fitness process, h 2i represents the shoulder position in the i-th sub-time monitoring area in the target fitness process, h 2i标 represents the standard shoulder position in the i-th sub-time monitoring area in the target fitness process, σ 1 and σ 2 are position weight coefficients, Δθ represents the comprehensive angle difference in the target fitness process, , θ 1i represents the shoulder joint angle in the i-th sub-time monitoring area in the target fitness process, θ 2i represents the elbow joint angle in the i-th sub-time monitoring area in the target fitness process, θ 3i represents the knee joint angle in the i-th sub-time monitoring area in the target fitness process, θ 4i represents the hip joint angle in the i-th sub-time monitoring area in the target fitness process, θ 1i标 , θ 2i标 , θ 3i标 and θ 4i标 Similarly, ρ 1 , ρ 2 , ρ 3 and ρ 4 are angle weight coefficients, a 1 and a 2respectively represent the influence coefficients of the comprehensive height difference and the comprehensive angle difference on the motion posture evaluation value, and the biomechanical evaluation value is specifically expressed as: , y 2 represents the biomechanical evaluation value during the target fitness process, h represents the change range of the center of gravity during the target fitness process, , h 3i represents the center of gravity height within the i-th sub-time monitoring area during the target fitness process, h 3i标 represents the standard center of gravity height within the i-th sub-time monitoring area during the target fitness process, v represents the motion speed during the target fitness process, x represents the motion displacement during the target fitness process, c 1 is a known constant, and a3, a4, and a5 respectively represent the influence coefficients of the center of gravity change range, motion speed, and motion displacement on the biomechanical evaluation value.

[0009] Preferably, the motion quality parameter preprocessing unit is used to establish a motion quality parameter preprocessing model, import the collected motion performance parameters and human physiological parameters into the motion quality parameter preprocessing model, and calculate the motion performance evaluation value and the human physiological evaluation value. The motion performance evaluation value is specifically expressed as: , y 3 represents the motion performance evaluation value during the target fitness process, n 1 represents the number of motions during the target fitness process, η represents the motion completion degree during the target fitness process, b 1 and b 2 respectively represent the influence coefficients of the number of motions and the motion completion degree on the motion performance evaluation value, c 2 is a known constant; the human physiological evaluation value is specifically marked as: , y 4 represents the human physiological evaluation value during the target fitness process, ΔR represents the average heart rate during the target fitness process, , R i represents the heart rate within the i-th sub-time monitoring area during the target fitness process, P i represents the blood pressure within the i-th sub-time monitoring area during the target fitness process, S i represents the blood oxygen saturation within the i-th sub-time monitoring area during the target fitness process, b 3 、b 4 and b 5 respectively represent the influence coefficients of the average heart rate, blood pressure, and blood oxygen saturation on the human physiological evaluation value, c 3 and c 4 are known constants.

[0010] Preferably, the human body posture data analysis unit is used to establish a human body posture data analysis model, import the calculated motion posture evaluation value and biomechanical evaluation value into the human body posture data analysis model, and calculate the human body posture evaluation value, which is specifically expressed as: , y a represents the human body posture evaluation value during the target fitness process, y 1 represents the motion posture evaluation value during the target fitness process, y 2 represents the biomechanical evaluation value during the target fitness process, ε 1 and ε 2 respectively represent the influence coefficients of the motion posture evaluation value and the biomechanical evaluation value on the human body posture evaluation value.

[0011] Preferably, the motion quality data analysis unit is used to establish a motion quality data analysis model, import the calculated motion performance evaluation value and human body physiological evaluation value into the motion quality data analysis model, and calculate the motion quality evaluation value, which is specifically expressed as: , y q represents the motion quality evaluation value during the target fitness process, y 3 represents the motion performance evaluation value during the target fitness process, y 4 represents the human body physiological evaluation value during the target fitness process, ε 3 and ε 4 respectively represent the influence coefficients of the motion performance evaluation value and the human body physiological evaluation value on the motion quality evaluation value.

[0012] Preferably, the fitness action comprehensive analysis model is used to calculate the fitness action comprehensive evaluation value, import the calculated human body posture evaluation value and motion quality evaluation value into the fitness action comprehensive analysis model, and obtain the fitness action comprehensive evaluation value, which is specifically expressed as: , φ represents the fitness action comprehensive evaluation value during the target fitness process, y a represents the human body posture evaluation value during the target fitness process, y q represents the motion quality evaluation value during the target fitness process, y a标 represents the human body posture evaluation standard value during the target fitness process, y q标 represents the motion quality evaluation standard value during the target fitness process.

[0013] Preferably, the fitness action comprehensive evaluation warning value is marked as φ 标 , when φ > φ 标 , it means that the fitness action comprehensive evaluation value is greater than the fitness action comprehensive evaluation warning value, and the fitness action does not meet the standard; when φ < φ 标 , it means that the fitness action comprehensive evaluation value is less than the fitness action comprehensive evaluation warning value, and the fitness action meets the standard.

[0014] Preferably, to construct the feedback adjustment model, multiple groups of first key information texts of fitness movements need to be collected, which should cover all situations that occur during fitness exercises. Then, the collected first key information texts of fitness movements are processed to obtain corresponding comprehensive evaluation values of fitness movements. These comprehensive evaluation values of fitness movements are associated with corresponding solutions and stored in the system operation database.

[0015] Preferably, the visual interface should include a menu bar, a toolbar, a status bar, and a work area. The work area should display the parameters of the target monitored fitness movement in real time and synchronously send the information to the user information terminal.

[0016] The technical effects and advantages of the present invention: The present invention collects motion posture parameters, biomechanical parameters, motion performance parameters, and human physiological parameters through the human posture parameter acquisition unit and the motion quality parameter acquisition unit of the data acquisition module, and determines a single sub-time monitoring area through the monitoring area division module, comprehensively and accurately collecting the parameters for evaluating whether the fitness movement meets the standards, providing effective data support for subsequent analysis and processing; The present invention calculates the motion posture evaluation value, biomechanical evaluation value, motion performance evaluation value, human physiological evaluation value, human posture evaluation value, motion quality evaluation value, and comprehensive evaluation value of fitness movements through the data preprocessing module, data analysis module, and comprehensive analysis module, accurately reflecting whether the real-time fitness movement meets the standards and effectively improving the accuracy of the system. Brief Description of the Drawings

[0017] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0018] Figure 2 It is a schematic diagram of the method steps of the present invention. Detailed Embodiments

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0020] As shown in the attached Figure 1 A machine vision-based fitness movement specification assistance system includes a system operation database, a system central processor, and a user information terminal, and also includes a monitoring area division module, a data acquisition module, a data preprocessing module, a data analysis module, a comprehensive analysis module, a judgment feedback module, and a human-computer interaction module.

[0021] The output end of the monitoring area division module is teleconnected to the input end of the data acquisition module. The output end of the data acquisition module is teleconnected to the input end of the data preprocessing module. The output end of the data preprocessing module is teleconnected to the input end of the data analysis module. The output end of the data analysis module is teleconnected to the input end of the comprehensive analysis module. The output end of the comprehensive analysis module is teleconnected to the input end of the judgment feedback module. The output end of the judgment feedback module is teleconnected to the input end of the human-computer interaction module. The output end of the system central processor is teleconnected to the input ends of each module. The output ends of each module are teleconnected to the input end of the system operation database.

[0022] The system operation database includes all data information of the fitness movement specification assistance system and collects in real time the data information output by each module. The system central processor is used to centrally control the information text instructions output by each module. The user information terminal is an information output device for receiving the fitness movement specification assistance system. Monitoring area division module: Determine the single sub-time monitoring area, divide the time of the monitored target fitness movement into single sub-time monitoring areas in the way of equal time length, and mark them as 1, 2, …, i, …, n in sequence.

[0023] Data acquisition module: Collect the first key information text of the fitness movement, including the human body posture parameter acquisition unit and the movement quality parameter acquisition unit, and transfer the collected data to the data preprocessing module.

[0024] The first key information text of the fitness movement includes human body posture parameters and movement quality parameters. The human body posture parameters include movement posture parameters and biomechanical parameters. The movement posture parameters specifically include the head position, shoulder position, shoulder joint angle, elbow joint angle, knee joint angle, and hip joint angle, which are respectively marked as h 1 、h 2 、θ 1 、θ 2 、θ 3 and θ 4 , and the biomechanical parameters specifically include the body center of gravity height, movement speed, and movement displacement, which are respectively marked as h 3 、v and x; The movement quality parameters include movement performance parameters and human physiological parameters. The movement performance parameters specifically include the number of movements and the movement completion degree, which are respectively marked as n 1 and η, and the human physiological parameters specifically include heart rate, blood pressure, and blood oxygen saturation, which are respectively marked as R, P, and S.

[0025] It should be specifically noted in this embodiment that the collected head position and shoulder position refer to the positions within a single sub-time monitoring area during the fitness exercise process. By comparing these positions with those in the standard movements, it can be determined whether the fitness movements are standard. The number of movements may include the number of jogging laps, the number of weightlifting repetitions, the number of squats, and the number of rope skipping repetitions; the heart rate, blood pressure, and blood oxygen saturation can be collected in real time using professional equipment.

[0026] Data preprocessing module: Perform data preprocessing on the first key information text of the fitness movement to obtain the first key information value of the fitness movement, including a human body posture parameter preprocessing unit and a movement quality parameter preprocessing unit, and transfer the results to the data analysis module.

[0027] Furthermore, the human body posture parameter preprocessing unit is used to establish a human body posture parameter preprocessing model. The collected movement posture parameters and biomechanical parameters are imported into the human body posture parameter preprocessing model to calculate the movement posture evaluation value and the biomechanical evaluation value. The movement posture evaluation value is specifically expressed as: , y 1 represents the movement posture evaluation value during the target fitness process, Δh represents the comprehensive height difference during the target fitness process, , h 1i represents the head position within the i-th sub-time monitoring area during the target fitness process, h 1i标 represents the standard head position within the i-th sub-time monitoring area during the target fitness process, h 2i represents the shoulder position within the i-th sub-time monitoring area during the target fitness process, h 2i标 represents the standard shoulder position within the i-th sub-time monitoring area during the target fitness process, σ 1 and σ 2 are position weight coefficients, Δθ represents the comprehensive angle difference during the target fitness process, , θ 1i represents the shoulder joint angle within the i-th sub-time monitoring area during the target fitness process, θ 2i represents the elbow joint angle within the i-th sub-time monitoring area during the target fitness process, θ 3i represents the knee joint angle within the i-th sub-time monitoring area during the target fitness process, θ 4i represents the hip joint angle within the i-th sub-time monitoring area during the target fitness process, θ 1i标 , θ 2i标 , θ 3i标 and θ 4i标 Similarly, ρ 1 , ρ 2 , ρ 3 and ρ 4 are angle weight coefficients, a 1 and a 2respectively represent the influence coefficients of the comprehensive height difference and the comprehensive angle difference on the motion posture evaluation value, and the biomechanical evaluation value is specifically expressed as: , y 2 represents the biomechanical evaluation value during the target fitness process, h represents the change range of the center of gravity during the target fitness process, , h 3i represents the center of gravity height within the i-th sub-time monitoring area during the target fitness process, h 3i标 represents the standard center of gravity height within the i-th sub-time monitoring area during the target fitness process, v represents the motion speed during the target fitness process, x represents the motion displacement during the target fitness process, c 1 is a known constant, and a3, a4, and a5 respectively represent the influence coefficients of the center of gravity change range, motion speed, and motion displacement on the biomechanical evaluation value.

[0028] Furthermore, the motion quality parameter preprocessing unit is used to establish a motion quality parameter preprocessing model, import the collected motion performance parameters and human physiological parameters into the motion quality parameter preprocessing model, and calculate the motion performance evaluation value and the human physiological evaluation value. The motion performance evaluation value is specifically expressed as: , y 3 represents the motion performance evaluation value during the target fitness process, n 1 represents the number of motions during the target fitness process, η represents the motion completion degree during the target fitness process, b 1 and b 2 respectively represent the influence coefficients of the number of motions and the motion completion degree on the motion performance evaluation value, c 2 is a known constant; the human physiological evaluation value is specifically marked as: , y 4 represents the human physiological evaluation value during the target fitness process, ΔR represents the average heart rate during the target fitness process, , R i represents the heart rate within the i-th sub-time monitoring area during the target fitness process, P i represents the blood pressure within the i-th sub-time monitoring area during the target fitness process, S i represents the blood oxygen saturation within the i-th sub-time monitoring area during the target fitness process, b 3 , b 4 and b 5 respectively represent the influence coefficients of the average heart rate, blood pressure, and blood oxygen saturation on the human physiological evaluation value, c 3 and c 4 are known constants.

[0029] Data analysis module: It performs data analysis on the first key information value of the fitness movement to obtain the second key information value of the fitness movement, including a human body posture data analysis unit and a movement quality data analysis unit, and transfers the results to the comprehensive analysis module.

[0030] Furthermore, the human body posture data analysis unit is used to establish a human body posture data analysis model, import the calculated movement posture evaluation value and biomechanical evaluation value into the human body posture data analysis model, and calculate the human body posture evaluation value, which is specifically expressed as: , y a represents the human body posture evaluation value in the target fitness process, y 1 represents the movement posture evaluation value in the target fitness process, y 2 represents the biomechanical evaluation value in the target fitness process, ε 1 and ε 2 respectively represent the influence coefficients of the movement posture evaluation value and the biomechanical evaluation value on the human body posture evaluation value.

[0031] Furthermore, the movement quality data analysis unit is used to establish a movement quality data analysis model, import the calculated movement performance evaluation value and human body physiological evaluation value into the movement quality data analysis model, and calculate the movement quality evaluation value, which is specifically expressed as: , y q represents the movement quality evaluation value in the target fitness process, y 3 represents the movement performance evaluation value in the target fitness process, y 4 represents the human body physiological evaluation value in the target fitness process, ε 3 and ε 4 respectively represent the influence coefficients of the movement performance evaluation value and the human body physiological evaluation value on the movement quality evaluation value.

[0032] Comprehensive analysis module: It constructs a comprehensive analysis model for fitness movements, processes and analyzes the second key information value of fitness movements through the comprehensive analysis model for fitness movements to obtain the comprehensive evaluation value of fitness movements, and transfers the results to the judgment and feedback module.

[0033] Furthermore, the comprehensive analysis model for fitness movements is used to calculate the comprehensive evaluation value of fitness movements. The calculated human body posture evaluation value and movement quality evaluation value are imported into the comprehensive analysis model for fitness movements to obtain the comprehensive evaluation value of fitness movements, which is specifically expressed as: , φ represents the comprehensive evaluation value of fitness movements in the target fitness process, y a represents the human body posture evaluation value in the target fitness process, y q represents the movement quality evaluation value in the target fitness process, y a标 represents the human body posture evaluation standard value in the target fitness process, y q标Represents the standard value for evaluating the exercise quality during the target fitness process.

[0034] Judgment feedback module: Construct a comprehensive evaluation warning value for fitness actions, determine whether the target fitness actions meet the standards, and construct a feedback adjustment model to take corresponding adjustment methods according to the judgment results.

[0035] Furthermore, the comprehensive evaluation warning value for fitness actions is marked as φ 标 , when φ > φ 标 , it means that the comprehensive evaluation value of the fitness action is greater than the comprehensive evaluation warning value of the fitness action, and the fitness action does not meet the standards; when φ < φ 标 , it means that the comprehensive evaluation value of the fitness action is less than the comprehensive evaluation warning value of the fitness action, and the fitness action meets the standards.

[0036] Even further, the construction of the feedback adjustment model requires collecting multiple groups of first key information texts of fitness actions, which should include all situations that occur during fitness exercises, and processing the collected first key information texts of fitness actions to obtain the corresponding comprehensive evaluation values of fitness actions, associating the comprehensive evaluation values of fitness actions with the corresponding solutions, and storing them in the system operation database.

[0037] Human-computer interaction module: Provide a visual interface to display the parameters during the target fitness process in real time and transmit them to the user information terminal.

[0038] The visual interface should include a menu bar, a toolbar, a status bar, and a work area. The work area should display the parameters of the target monitored fitness actions in real time and synchronously send the information to the user information terminal.

[0039] It should be specifically noted in this embodiment that the menu bar of the visual interface provides user navigation and setting options, including file operations, view control, and import and export of files. The toolbar includes common tool buttons to facilitate users to interact with the interface and display the real-time parameters of the target monitored fitness actions in real time, including motion posture parameters, biomechanical parameters, exercise performance parameters, and human physiological parameters, as well as motion posture evaluation values, biomechanical evaluation values, exercise performance evaluation values, human physiological evaluation values, human posture evaluation values, exercise quality evaluation values, and comprehensive evaluation values of fitness actions.

[0040] Secondly: In the attached drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved. Other structures can refer to the general design. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other; Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A fitness action standard auxiliary system based on machine vision, including a system operation database, a system central processor and a user information terminal, characterized in that: It also includes a monitoring area division module, a data acquisition module, a data preprocessing module, a data analysis module, a comprehensive analysis module, a judgment feedback module and a human-computer interaction module; The system operation database includes all data information of the fitness action standard auxiliary system, and collects data information output by each module in real time; The system central processor is used to control the information text instructions output by each module; the user information terminal is an information output device for receiving the fitness action standard auxiliary system; The monitoring area division module is used to determine a single sub-time monitoring area, divide the time for monitoring the target fitness action into single sub-time monitoring areas according to equal time lengths, and mark them as 1, 2, ..., i, ..., n in sequence; The data acquisition module is used to collect the first key information text of the fitness action, including a human posture parameter acquisition unit and a motion quality parameter acquisition unit, and transmits the collected data to the data preprocessing module; The data preprocessing module is used to perform data preprocessing on the first key information text of the fitness action to obtain the first key information value of the fitness action, including a human posture parameter preprocessing unit and a motion quality parameter preprocessing unit, and transmit the result to the data analysis module; The data analysis module is used to perform data analysis on the first key information value of the fitness action to obtain the second key information value of the fitness action, including a human posture data analysis unit and a movement quality data analysis unit, and transmit the result to the comprehensive analysis module; The comprehensive analysis module is used to construct a comprehensive analysis model for fitness actions, process and analyze the second key information value of the fitness actions through the comprehensive analysis model for fitness actions, obtain a comprehensive evaluation value of the fitness actions, and transmit the result to the judgment feedback module; The judgment feedback module is used to construct a comprehensive assessment warning value of fitness movements, judge whether the fitness movements of the target meet the standards, and construct a feedback adjustment model to adopt a corresponding adjustment method according to the judgment result; The human-computer interaction module is used to provide a visual interface, display the parameters of the target fitness process in real time, and transmit them to the user information terminal.

2. The machine vision-based fitness action standardization auxiliary system according to claim 1, characterized in that: The first key information text of the fitness action includes human posture parameters and motion quality parameters. The human posture parameters include motion posture parameters and biomechanical parameters. The motion posture parameters specifically include head position, shoulder position, shoulder joint angle, elbow joint angle, knee joint angle and hip joint angle, which are marked as h1, h2, θ1, θ2, θ3 and θ4 respectively. The biomechanical parameters specifically include body center of gravity height, movement speed and movement displacement, which are marked as h3, v and x respectively; the motion quality parameters include motion performance parameters and human physiological parameters. The motion performance parameters specifically include the number of exercises and the degree of completion of the exercises, which are marked as n1 and η respectively. The human physiological parameters specifically include heart rate, blood pressure and blood oxygen saturation, which are marked as R, P and S respectively.

3. The machine vision-based fitness action standardization auxiliary system according to claim 1, characterized in that: The human body posture parameter preprocessing unit is used to establish a human body posture parameter preprocessing model, import the collected motion posture parameters and biomechanical parameters into the human body posture parameter preprocessing model, calculate the motion posture evaluation value and the biomechanical evaluation value, and the motion posture evaluation value is specifically expressed as: , y1 represents the motion posture evaluation value during the target fitness process, Δh represents the comprehensive height difference during the target fitness process, ,h 1i represents the head position in the i-th sub-time monitoring area during the target fitness process, h 1i标 represents the standard head position in the i-th sub-time monitoring area during the target fitness process, h 2i represents the shoulder position in the i-th sub-time monitoring area during the target fitness process, h 2i标 represents the standard shoulder position in the i-th sub-time monitoring area during the target fitness process, σ1 and σ2 are position weight coefficients, Δθ represents the comprehensive angle difference during the target fitness process, ,θ 1i represents the shoulder joint angle in the i-th sub-time monitoring area during the target fitness process, θ 2i represents the elbow joint angle in the i-th sub-time monitoring area during the target fitness process, θ 3i represents the knee joint angle in the i-th sub-time monitoring area during the target fitness process, θ 4i represents the hip joint angle in the i-th sub-time monitoring area during the target fitness process, θ 1i标 ,θ 2i标 ,θ 3i标 and θ 4i标 As above, ρ1, ρ2, ρ3 and ρ4 are angle weight coefficients, a1 and a2 represent the influence coefficients of the comprehensive height difference and the comprehensive angle difference on the motion posture evaluation value, respectively. The biomechanical evaluation value is specifically expressed as: , y2 represents the biomechanical evaluation value during the target fitness process, h represents the range of change of the center of gravity during the target fitness process, ,h 3i represents the height of the center of gravity in the i-th sub-time monitoring area during the target fitness process, h 3i标 represents the standard center of gravity height in the i-th sub-time monitoring area during the target fitness process, v represents the movement speed during the target fitness process, x represents the movement displacement during the target fitness process, c1 is a known constant, a3, a4 and a5 respectively represent the influence coefficients of the center of gravity change amplitude, movement speed and movement displacement on the biomechanical evaluation value.

4. The machine vision-based fitness action standardization auxiliary system according to claim 1, characterized in that: The motion quality parameter preprocessing unit is used to establish a motion quality parameter preprocessing model, import the collected motion performance parameters and human physiological parameters into the motion quality parameter preprocessing model, and calculate the motion performance evaluation value and the human physiological evaluation value. The motion performance evaluation value is specifically expressed as: , y3 represents the sports performance evaluation value in the target fitness process, n1 represents the number of exercises in the target fitness process, η represents the degree of completion of the exercise in the target fitness process, b1 and b2 represent the influence coefficients of the number of exercises and the degree of completion of the exercise on the sports performance evaluation value, respectively, and c2 is a known constant; the human physiological evaluation value is specifically marked as: , y4 represents the physiological evaluation value of the human body during the target fitness process, ΔR represents the average heart rate during the target fitness process, , R i represents the heart rate in the i-th sub-time monitoring area during the target fitness process, P i represents the blood pressure in the i-th sub-time monitoring area during the target fitness process, S i represents the blood oxygen saturation in the i-th sub-time monitoring area during the target fitness process, b3, b4 and b5 represent the influence coefficients of average heart rate, blood pressure and blood oxygen saturation on human physiological evaluation values, respectively, and c3 and c4 are known constants.

5. The machine vision-based fitness action standardization auxiliary system according to claim 1, characterized in that: The human body posture data analysis unit is used to establish a human body posture data analysis model, import the calculated motion posture evaluation value and biomechanical evaluation value into the human body posture data analysis model, and calculate the human body posture evaluation value, which is specifically expressed as: ,y a represents the human body posture evaluation value during the target fitness process, y1 represents the sports posture evaluation value during the target fitness process, y2 represents the biomechanical evaluation value during the target fitness process, ε1 and ε2 represent the influence coefficients of the sports posture evaluation value and the biomechanical evaluation value on the human body posture evaluation value, respectively.

6. The machine vision-based fitness action standardization auxiliary system according to claim 1, characterized in that: The motion quality data analysis unit is used to establish a motion quality data analysis model, import the calculated motion performance evaluation value and human physiological evaluation value into the motion quality data analysis model, and calculate the motion quality evaluation value, which is specifically expressed as: ,y q represents the exercise quality evaluation value during the target fitness process, y3 represents the exercise performance evaluation value during the target fitness process, y4 represents the human physiological evaluation value during the target fitness process, ε3 and ε4 represent the influence coefficients of the exercise performance evaluation value and the human physiological evaluation value on the exercise quality evaluation value, respectively.

7. The machine vision-based fitness movement standardization auxiliary system according to claim 1, characterized in that: The fitness action comprehensive analysis model is used to calculate the comprehensive evaluation value of the fitness action. The calculated human posture evaluation value and the movement quality evaluation value are introduced into the fitness action comprehensive analysis model to obtain the comprehensive evaluation value of the fitness action, which is specifically expressed as: , φ represents the comprehensive evaluation value of fitness actions during the target fitness process, y a represents the human body posture evaluation value during the target fitness process, y q Represents the exercise quality evaluation value during the target fitness process, y a标 represents the standard value of human posture evaluation during the target fitness process, y q标 Indicates the standard value for evaluating the quality of exercise during the target fitness process.

8. The machine vision-based fitness movement standardization auxiliary system according to claim 1, characterized in that: The comprehensive assessment warning value of the fitness action is marked as φ 标 , when φ>φ 标 When φ<φ, it means that the comprehensive evaluation value of the fitness action is greater than the comprehensive evaluation warning value of the fitness action, and the fitness action does not meet the standard; when φ<φ 标 , it means that the comprehensive evaluation value of the fitness action is less than the comprehensive evaluation warning value of the fitness action, and the fitness action meets the standard.

9. The machine vision-based fitness action standardization auxiliary system according to claim 1, characterized in that: The construction of the feedback adjustment model requires the collection of multiple sets of first key information texts of fitness movements, which should include all situations that occur in fitness exercises, and the collected first key information texts of fitness movements are processed to obtain corresponding comprehensive evaluation values ​​of fitness movements, and the comprehensive evaluation values ​​of fitness movements are associated with corresponding solutions and stored in the system operation database.

10. The machine vision-based fitness movement standardization auxiliary system according to claim 1, characterized in that: The visualization interface should include a menu bar, a tool bar, a status bar and a work area, wherein the work area should display the parameters of the target monitored fitness action in real time and send the information synchronously to the user information terminal.

Images