A human posture analysis system and method for sports equipment based on big data

Abstract

This invention discloses a human posture analysis system and method for sports equipment based on big data, belonging to the field of human posture analysis technology. The system includes acquiring joint angle data of the user, the speed generated by the equipment, and the force exertion data of the user while using the equipment; establishing a static posture analysis model based on the joint angle values ​​to generate a static posture score; establishing a dynamic posture analysis model based on the joint angular velocity and the equipment data to generate a dynamic posture score; establishing a comprehensive posture analysis model based on the user's force exertion data, static posture score, and dynamic posture score to generate a comprehensive posture score; and analyzing the human posture based on the comprehensive posture score. This invention solves the problems of limited analytical dimensions and insufficient quantification of human-machine collaboration efficiency in existing technologies, improves the accuracy of motion quality assessment, and provides users with guidance that balances safety and effectiveness.

Description

Technical Field

[0001] This invention belongs to the field of human posture analysis technology, and in particular relates to a human posture analysis system and method for sports equipment based on big data. Background Technology

[0002] With the deepening implementation of the national fitness strategy and the rapid development of the smart sports industry, the public's demand for scientific, precise, and personalized fitness guidance is growing. In various sports training and fitness activities, correct body posture is key to improving training effectiveness and preventing sports injuries.

[0003] Currently, most existing human posture analysis technologies rely on computer vision solutions, which use cameras to capture key points of the human body and perform two-dimensional or three-dimensional reconstruction to analyze posture. This technology has the following technical shortcomings: traditional solutions using wearable sensors often only collect single-dimensional motion parameters, such as static joint angles or simple velocity data, and fail to establish a deep correlation with the operating status of sports equipment, resulting in fragmented analysis dimensions.

[0004] More notably, existing technologies lack an evaluation system that integrates posture standardization, movement fluidity, and force economy from multiple dimensions. They fail to quantify human-machine collaboration efficiency by matching equipment operation data with human movement parameters in time and space, and do not comprehensively evaluate movement quality from a biomechanical perspective, thus failing to provide users with quantitative guidance that balances safety and effectiveness. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a human posture analysis system and method for sports equipment based on big data, thus solving the aforementioned problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for analyzing human posture in sports equipment based on big data, specifically including the following steps: Acquire joint angle data of sports equipment users, equipment speed generated by the sports equipment, and force data of users when using the sports equipment; among which, joint angle data includes joint angle values ​​and joint angular velocities; Based on the joint angle values, a static posture analysis model is established to generate a static posture score. Based on joint angular velocity and equipment data generated by sports equipment, a dynamic posture analysis model is established to generate a dynamic posture score. Based on the force data, static posture score, and dynamic posture score of users when using sports equipment, a comprehensive posture analysis model is established to generate a comprehensive posture score. Human posture is analyzed based on a comprehensive posture score.

[0007] Based on the above technical solutions, the present invention also provides the following optional technical solutions: Further technical solution: The method for generating the static attitude score specifically includes: Based on the joint angle values, a joint angle deviation score is generated; A static posture analysis model is established based on the joint angle deviation score, and a static posture score is generated.

[0008] Further technical solution: The method for generating the joint angle deviation score specifically includes: Through the formula:

[0009] Generate joint angle deviation score ; In the formula, This represents the angle value of the i-th joint. This represents the standard angle value of the i-th joint. This represents the maximum permissible deviation angle of the i-th joint, and k is the attenuation coefficient.

[0010] Further technical solution: The expression of the attitude static analysis model is specifically as follows:

[0011] In the expression, This represents the static attitude score. This represents the joint angle deviation score. This represents the weight coefficient of the i-th joint, where n is the total number of joints.

[0012] Further technical solution: The method for generating the attitude dynamic score specifically includes: Based on the joint angular velocity, generate the joint angular velocity stability coefficient; The motion rhythm matching degree is generated based on the joint angular velocity and the equipment speed generated by the sports equipment; A dynamic posture analysis model is established based on the joint angular velocity stability coefficient and the motion rhythm matching degree, and a dynamic posture score is generated.

[0013] Further technical solution: The method for generating the joint angular velocity stability coefficient specifically includes: Through the formula:

[0014] Generate joint angular velocity stability coefficients ; In the formula, This represents the standard deviation of the joint angular velocity. It represents the arithmetic mean of the joint angular velocities.

[0015] Further technical solution: The specific method for generating the motion rhythm matching degree includes: Through the formula:

[0016] Generate motion rhythm matching degree ; In the formula, This represents the joint angular velocity at the j-th sampling point. This represents the average joint angular velocity across all sampling points. This represents the device speed at the j-th sampling point. This represents the average speed of the device across all sampling points, where m represents the number of sampling points.

[0017] Further technical solution: The expression of the attitude dynamic analysis model is specifically as follows:

[0018] In the expression, This represents the joint angular velocity stability coefficient. This represents the threshold value of the joint angular velocity stability coefficient. This represents the degree of rhythm matching in motion, where h is a constant. , All are weighting coefficients, and .

[0019] Further technical solution: The method for generating the overall attitude score specifically includes: Based on the force data of users when using sports equipment, the force efficiency of users when using sports equipment is generated; the force data includes the force vector applied by the user to the equipment, the motion velocity vector of the force point on the equipment, and the motion velocity vector of the user's limbs. Based on force efficiency, static posture score, and dynamic posture score, a comprehensive posture analysis model is established to generate a comprehensive posture score. The specific methods for generating the power efficiency include: Through the formula:

[0020] Generate power efficiency ; In the formula, This represents the force vector applied to the device by the user. This represents the velocity vector of the point on the equipment subjected to force. This represents the velocity vector of the user's limbs; The expression for the attitude synthesis analysis model is as follows:

[0021] In the expression, This indicates the efficiency of force exertion. This represents the static attitude score. This represents the attitude dynamic score. , All are weighting coefficients, and .

[0022] A human posture analysis system for sports equipment based on big data is used to execute the aforementioned human posture analysis method for sports equipment based on big data; specifically, it includes: The data acquisition unit is used to acquire joint angle data of the user of the sports equipment, the speed generated by the sports equipment, and the force data of the user when using the sports equipment; among which, the joint angle data includes joint angle values ​​and joint angular velocities; The static analysis unit is used to establish a static posture analysis model based on joint angle values ​​and generate a static posture score. The dynamic analysis unit is used to establish a dynamic posture analysis model and generate a dynamic posture score based on joint angular velocity and equipment data generated by sports equipment. The comprehensive analysis unit is used to establish a comprehensive posture analysis model and generate a comprehensive posture score based on the force data, static posture score and dynamic posture score of the user when using sports equipment. The human posture analysis unit is used to analyze human posture based on a comprehensive posture score.

[0023] This invention provides a human posture analysis system and method for sports equipment based on big data, which has the following advantages compared with the prior art: This invention integrates data on static deviation of joint angles, dynamic matching degree of movement rhythm, and force efficiency to construct a multi-dimensional scoring model. This solves the problems of single analysis dimensions and insufficient quantification of human-machine collaboration efficiency in existing technologies, improves the accuracy of movement quality assessment, and provides users with guidance that balances safety and effectiveness. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating a method for analyzing human posture in sports equipment based on big data, as provided by the present invention.

[0025] Figure 2 This is a schematic diagram of the structure of a human posture analysis system for sports equipment based on big data, provided by the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0027] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0028] Please see Figure 1 The present invention provides a method for analyzing human posture in sports equipment based on big data, comprising the following steps: Step S10: Obtain the joint angle data of the user of the sports equipment, the speed generated by the sports equipment, and the force data of the user when using the sports equipment; wherein, the joint angle data includes joint angle values ​​and joint angular velocities; Step S20: Based on the joint angle values, establish a static posture analysis model and generate a static posture score; Step S30: Based on the joint angular velocity and the equipment data generated by the sports equipment, establish a dynamic posture analysis model and generate a dynamic posture score; Step S40: Based on the force data, static posture score, and dynamic posture score of the user when using the sports equipment, establish a comprehensive posture analysis model and generate a comprehensive posture score; Step S50: Analyze human posture based on the overall posture score; Among them, joint angle data refers to the actual angles of the joints collected by inertial measurement units or optical sensors. Specifically, wearable sensors can be used to measure the angle values ​​and angular velocities of each joint in real time, providing basic data for static and dynamic analysis. Equipment speed refers to the speed of the force points during the operation of sports equipment. It can be obtained through the built-in encoder or speed sensor of the equipment and is used to evaluate the matching degree of human-machine movement rhythm. Force data includes the applied force vector, the velocity of the force point on the equipment, and the velocity of the limb. Specifically, it can be collected synchronously by a six-dimensional force sensor and a motion capture system to calculate the force efficiency. The static posture analysis model generates a score by quantifying the deviation of joint angles from standard angles, thus solving the problem of standardized assessment of static posture. The attitude dynamic analysis model combines angular velocity stability with equipment speed matching to evaluate dynamic coordination; The comprehensive analysis model introduces force efficiency as a correction factor to achieve a synergistic evaluation of static, dynamic, and energy transfer efficiency.

[0029] Specifically, through simultaneous acquisition of multi-source data, joint angle values, angular velocities, equipment speed, and force data are first obtained. In the static analysis phase, the deviation between the actual angles of each joint and the standard angles is calculated. The deviation score is quantified using an exponential function and then weighted and summed to obtain a static score; for example, the degree of deviation between the actual knee flexion angle and the standard value. In the dynamic analysis phase, the coefficient of variation of joint angular velocities is calculated to assess stability. Simultaneously, the synchronicity between joint angular velocities and equipment speed is analyzed using the Pearson correlation coefficient to generate a dynamic score. In the comprehensive analysis phase, the weighted results of the static and dynamic scores are corrected using force efficiency. For example, when the force direction is highly consistent with the equipment's movement direction, the efficiency value is close to 1, thereby improving the accuracy of the comprehensive score. Finally, posture analysis results are output based on the comprehensive score to guide users in adjusting their movement patterns.

[0030] Compared to existing technologies, current solutions rely on a single data source and lack a multi-dimensional evaluation system. For example, they may only judge posture correctness based on joint angles or analyze the simple correlation between equipment speed and human movement. This invention solves the problems of single evaluation dimensions and fragmented human-machine data in traditional methods by integrating data on joint static deviations, dynamic fluctuations, human-machine coordination, and force efficiency to construct a hierarchical analysis model. For example, in rowing machine training, existing technologies cannot simultaneously assess the stability of the user's torso angle, the matching degree between rowing rhythm and equipment resistance changes, and the efficiency of core muscle activation. However, this invention, through multi-model collaborative calculation, can comprehensively reflect the quality of movement.

[0031] Through the above technical solution, this invention achieves multi-dimensional quantitative assessment of human posture during the use of sports equipment, effectively improving the comprehensiveness and scientific rigor of posture analysis. Static scoring ensures the standardization of joint angles, dynamic scoring optimizes coordination during movement, and force efficiency assessment improves the economy of movement. For example, in barbell squat training, the system can identify static deviations in knee valgus, fluctuations in hip joint angular velocity during the ascent phase, and rhythmic mismatch with the barbell trajectory. Combined with lower limb force efficiency, it comprehensively prompts the user to adjust their center of gravity distribution and force sequence, thereby reducing the risk of knee joint injury and improving training effectiveness.

[0032] Preferably, the present invention further proposes a method for generating the attitude static score, specifically including: Step S21: Generate a joint angle deviation score based on the joint angle values; Step S22: Establish a static posture analysis model based on the joint angle deviation score and generate a static posture score; Specifically, the deviation of the actual angle of a single joint from the standard angle is first calculated using a joint angle deviation scoring function. The exponential function design ensures that the score approaches 1 when the actual angle is close to the standard angle, and rapidly decays when the deviation approaches the maximum allowable value, thus intuitively reflecting the impact of deviation on posture standardization. Subsequently, a static posture analysis model is used to weight and sum the deviation scores of multiple joints. The weighting coefficients can be dynamically adjusted according to the importance of the joints in different movement scenarios. For example, the weighting coefficient of the hip joint can be set higher than that of the knee joint in a squatting motion, while the weighting coefficient of the shoulder joint can be set even higher in a rowing motion. This phased processing approach preserves the independent evaluation of individual joints while achieving a coordinated evaluation of the overall posture through weight integration.

[0033] Compared to existing technologies, traditional methods typically rely solely on thresholds to determine whether joint angles meet standards, or perform simple averaging of multiple joint angles. These methods fail to differentiate the importance of different joints and lack precise quantification of deviation levels. This invention achieves continuous quantification of deviation levels through an exponential scoring function, while a weighted model reflects the differences in the impact of different joints on the overall posture. This addresses the problems of traditional methods, such as a single evaluation dimension and a lack of scientific quantification.

[0034] Through the above technical solution, this invention can accurately quantify the degree of deviation of each joint angle from the standard posture, and achieve multi-joint collaborative evaluation through weight allocation, significantly improving the comprehensiveness and accuracy of static posture analysis. For example, in the scenario of using fitness equipment, it can identify abnormal situations where the user's knee joint angle is slightly out of tolerance but has a high weight, thereby providing more targeted posture correction suggestions.

[0035] Preferably, the present invention further proposes a method for generating the joint angle deviation score, specifically including: Through the formula:

[0036] Generate joint angle deviation score ; In the formula, This represents the angle value of the i-th joint. This represents the standard angle value of the i-th joint. This represents the maximum permissible deviation angle of the i-th joint, where k is the attenuation coefficient; in, It refers to the joint angle value obtained by actual measurement, which can be realized by collecting data using an inertial measurement unit or an optical motion capture system, and is used to reflect the user's actual posture during movement; It refers to the pre-set standard joint angle value, which can be set according to the sports biomechanics model or professional training specifications, and is used to provide a reference benchmark for ideal posture; It refers to the maximum allowable deviation of a single joint from the standard angle. The specific value can be determined through clinical trials or sports injury studies and is used to normalize the deviation tolerance differences of different joints. k is the decay coefficient, which can be set to an empirical value or dynamically adjusted according to the training objective, and is used to control the rate at which the score decreases as the deviation increases.

[0037] Specifically, the normalized relative deviation is obtained by calculating the absolute deviation between the actual angle and the standard angle and dividing it by the maximum permissible deviation angle for that joint. This relative deviation is then multiplied by a decay coefficient and input into an exponential function, causing the score to decrease non-linearly with increasing deviation. When the actual angle approaches the standard value, the score approaches 1; when the deviation approaches the maximum permissible threshold, the score rapidly decreases to near 0. This calculation method allows for lateral comparison of the deviation levels of different joints, and by adjusting the k value, it can adapt to the stringent requirements of posture control at different training stages.

[0038] Compared to existing technologies, traditional methods typically use fixed thresholds to determine whether joint angles exceed limits, failing to quantify the degree of deviation and ignoring the physiological differences between different joints. This invention solves the problem of inconsistent evaluation standards for multiple joints by introducing the maximum permissible deviation angle as a normalization benchmark; it overcomes the deficiency of insufficient sensitivity of linear scoring in the critical region by constructing a nonlinear scoring model using an exponential function; and it achieves dynamic control of the scoring curve through an adjustable decay coefficient, meeting personalized training needs.

[0039] Through the above technical solution, the present invention can accurately quantify the degree of deviation of each joint angle from the standard value, eliminate the influence of differences in the physiological characteristics of different joints on the evaluation results, and at the same time enhance the ability to identify critical deviations through a nonlinear scoring mechanism, providing objective and comparable joint angle deviation data for subsequent posture analysis.

[0040] Preferably, the present invention further proposes the following expression for the attitude static analysis model:

[0041] In the expression, This represents the static attitude score. This represents the joint angle deviation score. This represents the weight coefficient of the i-th joint, where n is the total number of joints; Weighting coefficient This refers to the difference in importance of different joints in posture assessment. It can be achieved through preset or dynamic adjustment, such as assigning different values ​​according to the type of exercise or training goal, to reflect the degree of influence of core joints on overall posture. The total number of joints n refers to the number of human joints involved in the assessment. The specific number can be adjusted according to the type of sports equipment. For example, the assessment of a treadmill involves the number of lower limb joints, while the assessment of a rowing machine can be extended to the upper limb joints. Specifically, this technical solution integrates the local deviation information of each joint into a global score using a weighted summation model. The angle deviation score for each joint is... Its weighting coefficient The results are multiplied and summed to form a comprehensive static posture score. The differentiated design of the weighting coefficients allows for assigning higher weights to key joints; for example, in squat training, the hip joint weight can be set to 0.4, while the knee joint weight is 0.3, thus better aligning with biomechanical principles. The flexibility of the total number of joints, n, allows the model to adapt to different equipment. For instance, when evaluating an elliptical machine involving 6 joints, n is set to 6, while dumbbell presses can be extended to include the shoulder and elbow joints, increasing the total number of joints n to 8. Through this model, the impact of local angular deviations on the overall score is adjusted by the weighting coefficients, avoiding the evaluation distortion problems caused by simple averaging in traditional methods.

[0042] Compared to existing technologies, current static posture scoring methods typically use single joint angle comparisons or simple arithmetic averages, failing to consider the varying impacts of different joints on posture. For example, traditional methods may only calculate the deviation of the knee joint angle from the standard value, ignoring the importance of the hip joint, or directly average all joint deviation scores, resulting in the core joint deviations being diluted by non-critical joint data. This invention, by introducing weighting coefficients and a weighted summation model, achieves differentiated evaluation of multi-joint angle deviations, making the scoring results more closely reflect actual biomechanical requirements.

[0043] Through the above technical solution, this invention solves the problems of existing static posture scoring methods being too simplistic and failing to consider the differences in influence between different joints. By integrating the angle deviation scores of each joint through a weighted summation model, and combining the weight coefficients to reflect the differences in joint importance, a comprehensive assessment of multi-joint angle deviations is achieved. This solution improves the scientific rigor and accuracy of static posture scoring. For example, in rowing training, by increasing the weight of trunk joints, the impact of abnormal lumbar posture on the overall score can be captured more sensitively, thus providing a precise basis for correcting movement posture.

[0044] Preferably, the present invention further proposes a method for generating the attitude dynamic score, specifically including: Step S31: Generate the joint angular velocity stability coefficient based on the joint angular velocity; Step S32: Generate the motion rhythm matching degree based on the joint angular velocity and the equipment speed generated by the sports equipment; Step S33: Establish a dynamic posture analysis model based on the joint angular velocity stability coefficient and the motion rhythm matching degree, and generate a dynamic posture score; The joint angular velocity stability coefficient is an indicator that quantifies the degree of angular velocity fluctuation by calculating the ratio of the standard deviation to the mean of the joint angular velocity. Specifically, it can be achieved using the coefficient of variation formula in statistics. This coefficient can reflect the dynamic stability of joint control during movement. Movement rhythm matching degree refers to the index that measures the consistency of the changing trends of joint angular velocity and equipment speed by calculating the Pearson correlation coefficient between the two. Specifically, it can be achieved by using the ratio of the product of covariance and standard deviation. This index can assess the coordination between human movement rhythm and equipment operating status. The posture dynamic analysis model is a mathematical model that weights and integrates the stability coefficient and rhythm matching degree. Specifically, it can be implemented by linear weighted summation. This model can integrate dynamic stability and human-machine collaboration efficiency to form a multi-dimensional evaluation system.

[0045] Specifically, when generating the joint angular velocity stability coefficient, continuous time-series joint angular velocity data is collected, and the ratio of its standard deviation to the mean is calculated. A smaller ratio indicates less angular velocity fluctuation and higher dynamic stability. When generating the motion rhythm matching degree, the joint angular velocity sequence and the equipment velocity sequence within the same time period are standardized, and their correlation coefficient is calculated. A coefficient closer to 1 indicates more synchronized human-machine motion rhythm. When establishing the posture dynamic analysis model, the stability coefficient and rhythm matching degree are assigned weights and then summed, thereby achieving a comprehensive evaluation of dynamic stability and human-machine coordination.

[0046] Compared to existing technologies, traditional methods score based solely on the static deviation of a single joint angle, failing to consider the dynamic fluctuations of angular velocity during movement or to correlate human motion data with equipment operating data. This invention, by introducing an angular velocity stability coefficient and motion rhythm matching degree, achieves a dual breakthrough in both dynamic stability quantification and human-machine collaborative efficiency evaluation across the data dimension.

[0047] Through the above technical solution, this invention effectively solves the problem of a single dimension in dynamic posture evaluation in existing technologies. By simultaneously quantifying the stability of joint movements and the rhythm coordination between the human body and the equipment, it provides a comprehensive evaluation index for motion posture analysis that includes dynamic control capability and human-computer interaction efficiency. In the application of strength training equipment, this solution can accurately identify movement deformation caused by unstable joint control, and simultaneously detect energy loss caused by the mismatch between the force exertion rhythm and changes in equipment resistance, thereby guiding users to optimize their movement patterns.

[0048] Preferably, the present invention further proposes a method for generating the joint angular velocity stability coefficient, specifically including: Through the formula:

[0049] Generate joint angular velocity stability coefficients ; In the formula, This represents the standard deviation of the joint angular velocity. It represents the arithmetic mean of the joint angular velocities; Among them, the joint angular velocity stability coefficient This refers to transforming the dispersion of joint angular velocity into a normalized evaluation index through statistical methods. Specifically, it can be achieved by using the ratio of the standard deviation to the mean. This ratio can eliminate the influence of differences in the range of motion among different individuals on stability assessment. Standard deviation It refers to the fluctuation range of joint angular velocity over time, which can be specifically achieved by calculating the square root of the sum of squares of the deviations of the angular velocity at each sampling point from the mean, reflecting the degree of drastic change in angular velocity; Arithmetic mean It refers to the average level of joint angular velocity over time. Specifically, it can be achieved by calculating the ratio of the algebraic sum of the angular velocities of all sampling points to the number of sampling points, thus characterizing the overall velocity characteristics of joint movement.

[0050] Specifically, after continuously collecting joint angular velocity data at multiple time points, the arithmetic mean of these data is first calculated to characterize the average angular velocity level of the joint within the motion cycle. Then, the standard deviation is calculated to quantify the dispersion of the angular velocity relative to the mean. By using the ratio of the standard deviation to the mean as a stability coefficient, evaluation bias caused by individual differences in motion amplitude can be effectively eliminated. For example, when two users perform the same movement with larger and smaller amplitudes respectively, this coefficient can eliminate the amplitude difference factor and only reflect the stability of the movement rhythm. This approach allows users with different exercise habits or fitness levels to be objectively compared under the same evaluation system, providing standardized input parameters for dynamic posture scoring.

[0051] Compared to existing technologies, traditional methods typically only calculate the absolute fluctuation value of angular velocity or a simple statistical range, without considering the impact of individual differences in motion amplitude on stability assessment. For example, when existing technologies directly use the absolute value of the standard deviation as an evaluation indicator, the standard deviation for users with larger motion amplitudes will inevitably be significantly higher than that for users with smaller amplitudes, resulting in evaluation results that cannot truly reflect motion stability. This invention, however, introduces a coefficient of variation to normalize the standard deviation and mean, enabling the evaluation indicator to accurately reflect the relative fluctuation level and avoiding misjudgments caused by differences in individual motion amplitude.

[0052] Through the above technical solution, this invention can provide a dynamic stability quantification index independent of the amplitude of motion, effectively identifying coordination defects in the rhythm of joint movements. For example, in rowing machine training, if the angular velocity stability coefficient of the user's rowing motion is too low, it indicates that there is a problem with the rhythm of the movement or uneven force application. At this time, the system can provide targeted prompts to adjust the movement frequency. This technical solution provides an objective and comparable mathematical evaluation basis for dynamic posture analysis, solving the evaluation distortion problem caused by neglecting the differences in the amplitude of motion in traditional methods.

[0053] Preferably, the present invention further proposes a method for generating the motion rhythm matching degree, specifically including: Through the formula:

[0054] Generate motion rhythm matching degree ; In the formula, This represents the joint angular velocity at the j-th sampling point. This represents the average joint angular velocity across all sampling points. This represents the device speed at the j-th sampling point. This represents the average speed of the device across all sampling points, where m represents the number of sampling points. Among them, the degree of matching of movement rhythm It refers to the synchronicity measure of the changes in joint angular velocity and equipment speed. Specifically, it can be implemented using the Pearson correlation coefficient algorithm. This algorithm eliminates the difference in dimensions by using the ratio of covariance to standard deviation, and can objectively reflect the degree of linear correlation between the two sets of data. The number of sampling points m refers to the number of equally spaced measurements within the data acquisition period. Specifically, it can be achieved using a fixed-frequency sensor, for example, sampling once every 0.1 seconds to ensure coverage of the complete motion cycle. The average speed of the equipment refers to the average speed of the sports equipment during the exercise cycle. Specifically, it can be achieved by taking the arithmetic average of the speed data measured by the built-in encoder of the equipment, which is used to eliminate the reference offset of the equipment speed.

[0055] Specifically, this technical solution involves synchronously collecting time-series data on joint angular velocity and equipment speed. After calculating the mean of the two sets of data, the deviation values ​​at each sampling point are multiplied and summed. The numerator represents the cumulative amount of the same directionality between the joint movement trend and the equipment movement trend, while the denominator is normalized by multiplying the standard deviations. When the direction of change of joint angular velocity is completely consistent with the change of equipment speed, the correlation coefficient approaches 1, indicating a high degree of coordination between human and machine movement rhythms; when the directions of change are opposite, the coefficient approaches -1, indicating a significant rhythm conflict; and when there is no significant correlation, the coefficient approaches 0. This calculation method effectively eliminates the interference caused by individual differences in movement amplitude. For example, in rowing machine training, regardless of whether the user adopts a large stroke distance and high power or a small stroke distance and fast frequency movement mode, as long as the joint flexion and extension rhythm is consistent with the trend of change of the paddle speed, a high matching score can be obtained.

[0056] Compared to existing technologies, traditional methods only detect human posture through visual analysis or a single sensor, without considering the impact of the sports equipment's operating status on the quality of movement. For example, in the scenario of using an elliptical machine, existing technologies can only detect whether the user's knee joint angle meets the standard, but cannot determine the degree of matching between the pedaling speed and the change in flywheel resistance. In contrast, this invention, by establishing a mathematical correlation model between joint angular velocity and equipment speed, achieves for the first time a quantitative assessment of the rhythm of human-machine coordinated movement.

[0057] Through the above technical solution, this invention can accurately identify the temporal correlation between human joint activity and equipment operation during exercise, solving the problem that traditional posture analysis ignores human-computer interaction coordination. For example, in intelligent exercise bike training, it can detect whether the cyclist maintains synchronization between cadence and virtual slope changes during the uphill phase, providing data support for correcting pedaling rhythm imbalance, thereby avoiding compensatory muscle damage caused by human-machine rhythm mismatch.

[0058] Preferably, the present invention further proposes the following expression for the attitude dynamic analysis model:

[0059] In the expression, This represents the joint angular velocity stability coefficient. This represents the threshold value of the joint angular velocity stability coefficient. This represents the degree of rhythm matching, where h is a constant with a value of 2. , All are weighting coefficients, and ; Among them, the joint angular velocity stability coefficient This refers to quantifying the degree of fluctuation in joint movement by the ratio of the standard deviation to the mean of the joint angular velocity. Specifically, it can be achieved by using statistical methods to calculate the dispersion of the joint angular velocity, which is used to reflect dynamic stability. Movement rhythm matching It measures the coordination between joint angular velocity and equipment speed, which can be achieved by using time series data correlation analysis to evaluate the consistency of human-machine movement rhythm; Weighting coefficient , This refers to the proportion of contribution of joint stability and rhythm matching in dynamic scoring, which can be determined by empirical values ​​or machine learning optimization, and is used to balance the analysis weights of different dimensions. The constant h refers to the dimensional range used to adjust the matching degree of the movement rhythm. Specifically, it is set to 2 to ensure that the scoring results are within a reasonable range.

[0060] Specifically, the model uses a mathematical expression to weightedly fuse joint angular velocity stability and motion rhythm matching. First, the joint angular velocity stability coefficient is calculated as the ratio of the standard deviation to the mean. When this coefficient exceeds a preset threshold, a minimum value function is used to limit the lower limit of the score, preventing excessively poor stability from distorting the score. This is then combined with weighting coefficients. The impact of stability on dynamic scoring is emphasized. Secondly, the motion rhythm matching degree is calculated using the Pearson correlation coefficient to determine the linear correlation between joint angular velocity and equipment speed; the results are then normalized and compared with weighting coefficients. This involves combining and quantifying the contribution of human-machine collaboration to the scoring. Finally, constraints are applied. Ensure that the weights of the two dimensions are reasonably allocated to avoid bias in scoring towards a single factor. For example, when the trends of device speed and joint angular velocity change are highly consistent, the matching degree approaches 1, and the score for this part reaches its maximum value. If the fluctuation of joint angular velocity significantly exceeds the threshold, the stability score will be reduced proportionally, thereby dynamically reflecting the defects in attitude control.

[0061] Compared to existing technologies, current solutions typically analyze only single statistical indicators such as joint angles or velocities, failing to incorporate the matching of joint stability with the rhythm of device movement into a unified model. This invention, however, achieves multi-dimensional collaborative analysis of dynamic posture by integrating stability coefficients and matching degrees and introducing a weighting mechanism. For example, traditional methods may only determine the motion state based on the average angular velocity, but this invention further combines fluctuation levels and device coordination to more comprehensively identify instability or rhythm mismatch issues during movement.

[0062] Through the above technical solution, this invention solves the problem of a single dimension in dynamic posture scoring, enabling simultaneous evaluation of joint movement stability and human-machine rhythm matching, thus improving the comprehensiveness and accuracy of dynamic scoring. For example, in strength training, if the user's joint angular velocity fluctuates significantly but the equipment speed matches well, the model can balance the influence of both through weight allocation, avoiding misjudgments caused by a single indicator. In periodic exercises such as rowing machine training, the model can identify the degree of coordination between the user's exertion rhythm and the changes in equipment resistance by quantifying the rhythm matching degree, thereby providing targeted guidance for optimizing movements.

[0063] Preferably, the present invention further proposes a method for generating the attitude comprehensive score, specifically including: Step S41: Generate the force efficiency of the user when using the sports equipment based on the force data of the user; wherein, the force data includes the force vector applied by the user to the equipment, the motion velocity vector of the force point of the equipment, and the motion velocity vector of the user's limbs. Step S42: Based on the force efficiency, static posture score, and dynamic posture score, establish a comprehensive posture analysis model and generate a comprehensive posture score; The specific methods for generating the power efficiency include: Through the formula:

[0064] Generate power efficiency ; In the formula, This represents the force vector applied to the device by the user. This represents the velocity vector of the point on the equipment subjected to force. This represents the velocity vector of the user's limbs; The expression for the attitude synthesis analysis model is as follows:

[0065] In the expression, This indicates the efficiency of force exertion. This represents the static attitude score. This represents the attitude dynamic score. , All are weighting coefficients, and ; Among them, power efficiency This refers to the degree of consistency between the force vector applied by the user and the direction of movement of the device. Specifically, it can be achieved by performing a dot product operation on the vector data collected by the force sensor and the velocity sensor. This indicator is used to quantify the effectiveness of energy transfer. Force vector It refers to the direction and magnitude of the force applied by the user at the contact point of the device. Specifically, a six-dimensional force sensor can be installed on the grip part of the device for real-time measurement. The velocity vector of the point of force application on the equipment It refers to the instantaneous velocity direction and magnitude of the contact point when the equipment is pushed or pulled. Specifically, it can be obtained by using an encoder or an inertial measurement unit installed on the moving parts of the equipment. User's limb velocity vector It refers to the direction and magnitude of the movement speed of the part of the human body in contact with the equipment. Specifically, it can be collected by attaching an inertial measurement unit or an optical motion capture system to the end of the limb. The weighting coefficient refers to the ratio parameter that adjusts the weighting of static and dynamic scores according to different sports types. Specifically, it can be dynamically adjusted by preset empirical values ​​or machine learning models for different sports modes.

[0066] Specifically, by collecting force and velocity vector data, the force efficiency is calculated using the vector dot product formula. The efficiency value approaches its maximum when the applied force direction aligns with the equipment's movement direction and the limb velocity matches the equipment velocity. This calculation method incorporates vector direction matching into the evaluation system, overcoming the limitations of traditional methods that only focus on force magnitude. In the posture comprehensive analysis model, force efficiency acts as a product factor in the weighted sum of static and dynamic scores. The static score reflects the standardization of joint angles, while the dynamic score characterizes the coordination of movement rhythm. The introduction of weighting coefficients allows for adjustments to score weights based on different training objectives; for example, increasing the weight of static scores during strength training and emphasizing dynamic scores during endurance training. This model structure achieves a unified evaluation of movement standardization and force efficiency, deeply integrating human kinematic data with equipment dynamics data.

[0067] In some specific implementations, a six-dimensional force sensor can be installed on the grip of the fitness equipment to measure pushing and pulling force data, an encoder can be installed at both ends of the barbell to record the motion trajectory, and an inertial measurement unit can be strapped to the user's wrist to acquire end-limb velocity. The data processing unit calculates the force efficiency in real time and merges the static and dynamic scores according to preset weighting coefficients. When performing bench press training, the system automatically selects the strength training mode. Set it to 0.6. Set to 0.4 to emphasize the importance of standard joint angles. Switch to aerobic mode when training on the elliptical machine. Adjusted to 0.4 Set to 0.6 to focus on assessing the continuity of the movement rhythm.

[0068] Compared to existing technologies, traditional methods assess posture using only single-dimensional data, such as visual systems detecting joint angles or sensors measuring force magnitude, failing to establish a correlation between force direction and equipment motion. This invention quantifies force efficiency through vector operations, combining equipment motion data with human dynamics data, thus solving the problem of missing human-machine coordination assessment in existing technologies. Compared to methods that only use scalar values ​​to calculate power, this invention significantly improves the accuracy of force efficiency analysis through directional matching assessment.

[0069] Through the above technical solution, this invention achieves precise quantitative evaluation of force exertion efficiency, effectively solving the energy loss problem caused by the mismatch between the force exertion direction and the equipment movement direction in existing technologies. By establishing a coupled model of force exertion efficiency and posture scoring, it can simultaneously evaluate movement standardization and force exertion effectiveness, providing data support for correcting erroneous movement patterns and optimizing training programs. This solution overcomes the limitations of traditional single-dimensional evaluation systems and has practical application value in fields such as strength training and rehabilitation medicine, accurately identifying abnormal human postures with low force exertion efficiency.

[0070] Please see Figure 2 The present invention further proposes a human posture analysis system for sports equipment based on big data. This system is used in the above-mentioned human posture analysis method for sports equipment based on big data, and specifically includes: The data acquisition unit 10 is used to acquire joint angle data of the user of the sports equipment, the speed generated by the sports equipment, and the force data of the user when using the sports equipment; wherein, the joint angle data includes joint angle values ​​and joint angular velocities; The static analysis unit 20 is used to establish a static posture analysis model based on the joint angle values ​​and generate a static posture score. The dynamic analysis unit 30 is used to establish a dynamic posture analysis model and generate a dynamic posture score based on the joint angular velocity and equipment data generated by the sports equipment. The comprehensive analysis unit 70 is used to establish a comprehensive posture analysis model and generate a comprehensive posture score based on the force data, static posture score and dynamic posture score of the user when using sports equipment. The human posture analysis unit 50 is used to analyze human posture based on the overall posture score.

[0071] Preferably, the present invention further proposes that the static analysis unit 20 specifically includes: The deviation analysis module is used to generate joint angle deviation scores based on joint angle values. The posture static score generation module is used to establish a posture static analysis model based on joint angle deviation scores and generate a posture static score.

[0072] Preferably, the present invention further proposes that the dynamic analysis unit 30 specifically includes: The stability analysis module is used to generate joint angular velocity stability coefficients based on joint angular velocities. The matching degree analysis module is used to generate the motion rhythm matching degree based on the joint angular velocity and the equipment speed generated by the sports equipment; The posture dynamic score generation module is used to establish a posture dynamic analysis model based on the joint angular velocity stability coefficient and the motion rhythm matching degree, and generate a posture dynamic score.

[0073] Preferably, the present invention further proposes that the comprehensive analysis unit 40 specifically includes: The force analysis module is used to generate the force efficiency of a user when using sports equipment based on the force data of the user. The force data includes the force vector applied by the user to the equipment, the velocity vector of the force point on the equipment, and the velocity vector of the user's limbs. The comprehensive score generation module is used to establish a comprehensive attitude analysis model based on force efficiency, static attitude score, and dynamic attitude score, and generate a comprehensive attitude score.

[0074] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for analyzing human posture in sports equipment based on big data, characterized in that, Specifically, the following steps are included: Acquire joint angle data of sports equipment users, equipment speed generated by the sports equipment, and force data of users when using the sports equipment; among which, joint angle data includes joint angle values ​​and joint angular velocities; Based on the joint angle values, a static posture analysis model is established to generate a static posture score. Based on joint angular velocity and equipment data generated by sports equipment, a dynamic posture analysis model is established to generate a dynamic posture score. Based on the force data, static posture score, and dynamic posture score of users when using sports equipment, a comprehensive posture analysis model is established to generate a comprehensive posture score. Human posture is analyzed based on a comprehensive posture score.

2. The method for analyzing human posture in sports equipment based on big data according to claim 1, characterized in that, The method for generating the static attitude score specifically includes: Based on the joint angle values, a joint angle deviation score is generated; A static posture analysis model is established based on the joint angle deviation score, and a static posture score is generated.

3. The method for analyzing human posture in sports equipment based on big data according to claim 2, characterized in that, The specific methods for generating the joint angle deviation score include: Through the formula: Generate joint angle deviation score ; In the formula, This represents the angle value of the i-th joint. This represents the standard angle value of the i-th joint. This represents the maximum permissible deviation angle of the i-th joint, and k is the attenuation coefficient.

4. The method for analyzing human posture in sports equipment based on big data according to claim 2, characterized in that, The expression for the attitude static analysis model is as follows: In the expression, This represents the static attitude score. This represents the joint angle deviation score. This represents the weight coefficient of the i-th joint, where n is the total number of joints.

5. The method for analyzing human posture in sports equipment based on big data according to claim 1, characterized in that, The specific methods for generating the dynamic attitude score include: Based on the joint angular velocity, generate the joint angular velocity stability coefficient; The motion rhythm matching degree is generated based on the joint angular velocity and the equipment speed generated by the sports equipment; A dynamic posture analysis model is established based on the joint angular velocity stability coefficient and the motion rhythm matching degree, and a dynamic posture score is generated.

6. The method for analyzing human posture in sports equipment based on big data according to claim 5, characterized in that, The specific methods for generating the joint angular velocity stability coefficient include: Through the formula: Generate joint angular velocity stability coefficients ; In the formula, This represents the standard deviation of the joint angular velocity. It represents the arithmetic mean of the joint angular velocities.

7. The method for analyzing human posture in sports equipment based on big data according to claim 5, characterized in that, The specific methods for generating the motion rhythm matching degree include: Through the formula: Generate motion rhythm matching degree ; In the formula, This represents the joint angular velocity at the j-th sampling point. This represents the average joint angular velocity across all sampling points. This represents the device speed at the j-th sampling point. This represents the average speed of the device across all sampling points, where m represents the number of sampling points.

8. The method for analyzing human posture in sports equipment based on big data according to claim 5, characterized in that, The expression for the attitude dynamic analysis model is as follows: In the expression, This represents the joint angular velocity stability coefficient. This represents the threshold value of the joint angular velocity stability coefficient. This represents the degree of rhythm matching in motion, where h is a constant. , All are weighting coefficients, and .

9. The method for analyzing human posture in sports equipment based on big data according to claim 1, characterized in that, The specific methods for generating the overall attitude score include: Based on the force exertion data of users when using sports equipment, the force exertion efficiency of users when using sports equipment is generated; the force exertion data includes the force vector applied by the user to the equipment, the motion velocity vector of the force point on the equipment, and the motion velocity vector of the user's limbs. Based on force efficiency, static posture score, and dynamic posture score, a comprehensive posture analysis model is established to generate a comprehensive posture score. The specific methods for generating the power efficiency include: Through the formula: Generate power efficiency ; In the formula, This represents the force vector applied to the device by the user. This represents the velocity vector of the point on the equipment subjected to force. This represents the velocity vector of the user's limbs; The expression for the attitude synthesis analysis model is as follows: In the expression, This indicates the efficiency of force exertion. This represents the static attitude score. This represents the attitude dynamic score. , All are weighting coefficients, and .

10. A human posture analysis system for sports equipment based on big data, characterized in that, The system is used to execute the big data-based human posture analysis method for sports equipment as described in any one of claims 1-9, specifically including: The data acquisition unit is used to acquire joint angle data of the user of the sports equipment, the speed generated by the sports equipment, and the force data of the user when using the sports equipment; among which, the joint angle data includes joint angle values ​​and joint angular velocities; The static analysis unit is used to establish a static posture analysis model based on joint angle values ​​and generate a static posture score. The dynamic analysis unit is used to establish a dynamic posture analysis model and generate a dynamic posture score based on joint angular velocity and equipment data generated by sports equipment. The comprehensive analysis unit is used to establish a comprehensive posture analysis model and generate a comprehensive posture score based on the force data, static posture score and dynamic posture score of the user when using sports equipment. The human posture analysis unit is used to analyze human posture based on a comprehensive posture score.

Images