Volleyball training method and system based on AI attitude capture

Abstract

The invention relates to the technical field of physical training, in particular to a volleyball training method and system based on AI posture capture. Comprising the following steps that a mixed posture capturing system collects training data of an athlete in the volleyball training process in real time; analyzing the training data based on an AI processing unit of a preset volleyball action database of athletes with different technical levels and body conditions to obtain analysis data; generating a personalized training scheme according to the analysis data in combination with historical data and personalized features of the athletes; the personalized training scheme is fed back to the sportsman in a visual interaction mode in real time, meanwhile, the sportsman is explained and guided through voice interaction according to personalized training, and the sportsman carries out volleyball training according to the visual interaction and the voice interaction. According to the volleyball training method and system, the generated personalized training scheme is more scientific and effective through the obtained analysis data, so that the training effect is improved when athletes carry out volleyball training.

Description

Technical Field

[0001] The present invention relates to the field of sports training technology, and in particular to a volleyball training method and system based on AI posture capture. Background Art

[0002] In modern volleyball training, traditional training methods and techniques face numerous challenges. Currently, most volleyball training data collection relies on a single device, such as using only a standard camera to record athlete movements. This method captures a single dimension of data and fails to fully capture the athlete's three-dimensional posture data, motion trajectory data, and joint angle change data during training, making it difficult to conduct in-depth analysis of the athlete's movements. Furthermore, the movement analysis process lacks a comprehensive, multi-dimensional consideration. Simply comparing movement forms without integrating mechanical models to analyze the force generation and impact point of the hitting action, nor does it consider the rationality of the athlete's movement and stance in actual tactical scenarios. This makes it impossible to accurately identify the root causes of athlete movement problems, and the training guidance lacks scientificity and effectiveness.

[0003] When it comes to developing training plans, existing technologies often adopt uniform training standards and models, ignoring individual differences in athletes' physical fitness, technical proficiency, learning abilities, and other characteristics. This makes it difficult for training plans to fully meet the actual needs of each athlete. This can lead to excessive training intensity, impacting athletes' health, or insufficient training intensity, failing to effectively improve athletes' skills, making it difficult to maximize training results.

[0004] In summary, in view of the technical problem in the prior art that the training effect on athletes during volleyball training is relatively low, the technical problem actually solved by the present invention is how to improve the training effect of athletes in volleyball training. Summary of the Invention

[0005] In order to overcome the technical defects of the above-mentioned prior art in terms of low training effect on athletes during volleyball training, the purpose of the present invention is to provide a volleyball training method and system based on AI posture capture, by analyzing the force mode information and hitting point information of the hitting action based on a mechanical model and the athlete's running information and standing position information in combination with tactical scenarios to obtain analysis data, so that the generated personalized training plan is more scientific and effective, so that athletes can improve the training effect when performing volleyball training.

[0006] The present invention discloses a volleyball training method based on AI posture capture, comprising the following steps:

[0007] A hybrid posture capture system is constructed using a fisheye camera, a high-definition camera, and an inertial sensor. The hybrid posture capture system collects real-time training data from athletes during volleyball training. The training data includes at least three-dimensional posture information, motion trajectory information, and joint angle change information.

[0008] An AI processing unit, which includes a pre-set database of volleyball movements for athletes of varying skill levels and physical conditions, analyzes training data to obtain analytical data. This data is obtained by analyzing the force generation and hitting point information of the hitting movement based on a mechanical model, as well as the athlete's movement and positioning information in combination with tactical scenarios.

[0009] Generate personalized training plans based on analyzed data and combined with the athlete's historical data and individual characteristics;

[0010] The personalized training plan is fed back to the athletes in real time through visual interaction. At the same time, the athletes are explained and guided through voice interaction based on the personalized training. The athletes perform volleyball training based on the visual interaction and voice interaction.

[0011] Preferably, when the fisheye camera, the high-definition camera, and the inertial sensor are installed, a unified three-dimensional coordinate system is established through a spatial calibration algorithm, so that the three-dimensional posture information, motion trajectory information, and joint angle change information collected by the fisheye camera, the high-definition camera, and the inertial sensor are in the same spatial reference;

[0012] The spatial calibration algorithm is based on Zhang's calibration method and combines the calibration parameters of the inertial sensor for joint calibration.

[0013] Preferably, the horizontal viewing angle of the fisheye camera is greater than or equal to 180°, the vertical viewing angle is greater than or equal to 150°, and the fisheye camera is equipped with an image distortion correction algorithm to perform real-time distortion correction on the fisheye image when the fisheye camera captures the fisheye image;

[0014] The HD camera has a resolution greater than or equal to 4K and a frame rate greater than or equal to 60fps;

[0015] The inertial sensor’s measurement accuracy is less than ±0.5° in angle error and less than ±2.1m / s in acceleration error. 2 .

[0016] Preferably, the fisheye camera is used to obtain panoramic data of the training ground, the high-definition camera is used to obtain the athlete's movement data, and the inertial sensor is used to obtain the athlete's joint movement data;

[0017] After preprocessing the panoramic data, the motion data, and the joint motion data, three-dimensional posture information, motion trajectory information, and joint angle change information are obtained based on the preprocessed panoramic data, and / or the motion data, and / or the joint motion data;

[0018] Preprocessing includes denoising and data compression.

[0019] Preferably, the step of analyzing the training data based on the AI ​​processing unit having a preset volleyball action database containing athletes of different skill levels and physical conditions to obtain analysis data further includes analyzing the angular velocity ω and angular acceleration α at the moment when the athlete hits the volleyball, so as to evaluate the stability of the athlete's hitting action based on the angular velocity ω and angular acceleration α. ​​When the angular velocity ω is greater than an angular velocity threshold and / or when the angular acceleration α is greater than an angular acceleration threshold, it is determined that the stability of the athlete's hitting action is insufficient.

[0020] The calculation formula for angular velocity ω is: The calculation formula for angular acceleration α is:

[0021] Among them, Δθ represents the change in joint angle, and Δt represents the time interval.

[0022] Preferably, after the step of generating a personalized training program based on the analyzed data and in combination with the athlete's historical data and personalized characteristics, the method further includes the following steps:

[0023] A training effect estimation report is generated based on the personalized training plan and the athlete's historical data and personalized characteristics. The analytic hierarchy process is used to assign weights to the various movement indicators of the athletes during volleyball training to build a comprehensive evaluation model. The athlete's training effect is then quantitatively evaluated based on the comprehensive evaluation model. The scoring formula is:

[0024]

[0025] Among them, w i Expressed as the weight of the i-th action indicator, x i It is represented as the score of the i-th action indicator, and n is represented as the total number of action indicators involved in the comprehensive evaluation calculation.

[0026] Preferably, a comprehensive scoring threshold is set in the comprehensive evaluation model;

[0027] When the comprehensive evaluation model calculates the athlete's quantitative evaluation score, the quantitative evaluation score is compared with the comprehensive score threshold. When the quantitative evaluation score is less than the comprehensive score threshold, the training data is re-acquired to generate a new personalized training plan.

[0028] In view of this, the second purpose of the present invention is to provide a volleyball training system based on AI posture capture, including an acquisition module, an analysis module, a training program generation module, and a training module; wherein,

[0029] Acquisition module: This module is used to collect real-time training data of athletes during volleyball training using a hybrid posture capture system composed of a fisheye camera, a high-definition camera, and an inertial sensor. The training data includes at least three-dimensional posture information, motion trajectory information, and joint angle change information.

[0030] Analysis module: This module is connected to the acquisition module and is used to analyze training data based on an AI processing unit that has a preset database of volleyball movements of players with different technical levels and physical conditions to obtain analytical data. The analytical data is obtained by analyzing the force generation method and hitting point information of the hitting movement based on the mechanical model, as well as the movement and stance information of the players in combination with tactical scenarios;

[0031] Training program generation module: connected to the analysis module, used to generate personalized training programs based on the analysis data and combined with the athlete's historical data and personalized characteristics;

[0032] Training module: connected to the training program generation module, used to provide real-time feedback of personalized training programs to athletes in a visual interactive manner, and at the same time explain and guide athletes based on personalized training through voice interaction. Athletes perform volleyball training based on visual interaction and voice interaction.

[0033] Preferably, it also includes a scoring calculation module; wherein,

[0034] Scoring calculation module: connected to the training program generation module, it is used to generate a training effect estimation report based on the personalized training program and the athlete's historical data and personalized characteristics. It also uses the hierarchical analysis method to assign weights to the various movement indicators of the athletes during volleyball training to build a comprehensive evaluation model. Based on the comprehensive evaluation model, it calculates the athlete's training effect for quantitative evaluation. The scoring calculation formula is:

[0035]

[0036] Among them, w i Expressed as the weight of the i-th action indicator, x i It is represented as the score of the i-th action indicator, and n is represented as the total number of action indicators involved in the comprehensive evaluation calculation.

[0037] Preferably, a comprehensive scoring threshold is provided in the comprehensive evaluation model, and the scoring calculation module is also used to compare the quantitative evaluation score with the comprehensive scoring threshold when the comprehensive evaluation model calculates the athlete's quantitative evaluation score. When the quantitative evaluation score is less than the comprehensive scoring threshold, the training data is re-acquired to generate a new personalized training plan.

[0038] After adopting the above technical solution, compared with the existing technology, the beneficial effect of the present invention is that by analyzing the force generation method information and hitting point information of the hitting action based on the mechanical model and the athlete's running information and standing position information in combination with tactical scenarios to obtain analysis data, the generated personalized training plan is more scientific and effective, so that athletes can improve the training effect when performing volleyball training. At the same time, when generating personalized training plans based on the analysis data, the historical data and personalized characteristics of individual athletes are combined, thereby further improving the targeting of individual athletes, so as to further improve the training effect of individual athletes when performing volleyball training; in addition, the present invention uses fisheye cameras, high-definition cameras and inertial sensors to form a hybrid posture capture system training data, which can perform in-depth analysis of the athlete's movements to improve the accuracy of the analysis data, indirectly improve the effectiveness of the personalized training plan, and then improve the training effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 The figure is a schematic diagram of the steps of a volleyball training method and system based on AI posture capture according to the present invention. DETAILED DESCRIPTION

[0040] The advantages of the present invention are further described below with reference to the accompanying drawings and specific embodiments.

[0041] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

[0042] The terms used in this disclosure are for the purpose of describing specific embodiments only and are not intended to limit the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an," "the," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0043] It should be understood that although the terms first, second, third, etc. may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining."

[0044] In the description of the present invention, it should be understood that the terms "longitudinal", "transverse", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention.

[0045] In the description of the present invention, unless otherwise specified and limited, it should be noted that the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a mechanical connection or an electrical connection, or it can be the internal communication between two components. It can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to the specific circumstances.

[0046] In the following description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present invention and have no specific meaning. Therefore, "module" and "component" can be used interchangeably.

[0047] The present embodiment provides a volleyball training method based on AI posture capture, comprising the following steps: utilizing a fisheye camera, a high-definition camera, and an inertial sensor to form a hybrid posture capture system, and collecting training data of athletes during volleyball training in real time based on the hybrid posture capture system, wherein the training data includes at least three-dimensional posture information, motion trajectory information, and joint angle change information; analyzing the training data based on an AI processing unit that has a preset volleyball action database of athletes with different technical levels and physical conditions to obtain analysis data, wherein the analysis data is obtained by analyzing the force generation method information and hitting point information of the hitting action based on a mechanical model, as well as the running information and standing information of the athlete in combination with tactical scenarios; generating a personalized training plan based on the analysis data and in combination with the athlete's historical data and personalized characteristics; feeding back the personalized training plan to the athlete in real time in a visual interactive manner, and explaining and guiding the athlete through voice interaction based on the personalized training, and the athlete performing volleyball training based on the visual interaction and voice interaction.

[0048] See Figure 1 As shown, in this embodiment, a volleyball training method based on AI posture capture is described in detail, which specifically includes the following steps:

[0049] Step S100: A hybrid posture capture system composed of a fisheye camera, a high-definition camera and an inertial sensor is used to collect the training data of athletes during volleyball training in real time. Specifically, the horizontal viewing angle of the fisheye camera is greater than or equal to 180°, the vertical viewing angle is greater than or equal to 150°, and the fisheye camera is deployed on the top of the training ground to obtain panoramic information. The resolution of the high-definition camera is greater than or equal to 4K, the frame rate is greater than or equal to 60fps, and the high-definition cameras are arranged around the training ground to focus on the details of the athletes' movements. The measurement accuracy of the inertial sensor is less than ±0.5° in terms of angle error and less than ±2.1m / s in terms of acceleration error. 2 , including but not limited to being worn on the athlete's shoulders, elbows, wrists and other key joints.

[0050] It should be noted that the training data includes at least three-dimensional posture information, motion trajectory information, and joint angle change information. The three-dimensional posture information is fused through triangulation of fisheye cameras and high-definition cameras with information obtained by inertial sensors to construct the three-dimensional coordinates of the athlete's joints throughout the body, thereby obtaining three-dimensional posture data. The motion trajectory information is obtained by tracking the athlete's movement trajectory in the training venue in real time using a fisheye camera combined with a venue coordinate system, thereby obtaining motion trajectory information. The joint angle change information is obtained by calculating the joint angle change based on the joint angular velocity ω and angular acceleration α collected in real time by the inertial sensors, thereby obtaining joint angle change information. The joint angle change Δθ is calculated as Δθ = ω·Δt.

[0051] It should be noted that the field coordinate system is a reference frame used to determine the positions of the players and volleyball on the training field. It is established in the following way:

[0052] Camera calibration: By setting up calibration objects at known locations in the training field, these objects are photographed by fisheye and HD cameras. Image processing algorithms are then used to calculate the internal parameters (such as focal length and principal point) and external parameters (such as rotation and translation) of the fisheye and HD cameras. This determines the position and attitude of the fisheye and HD cameras in the field coordinate system. This allows each pixel in the image captured by the camera to be mapped to its actual position in the field coordinate system through coordinate transformation.

[0053] Field feature point extraction: Identify fixed feature points on the training field, such as the four corners of the field and the intersection of the centerline and sidelines. The locations of these feature points in the field coordinate system are known. By detecting the locations of these feature points in images captured by fisheye cameras and high-definition cameras, a mapping relationship between image coordinates and field coordinates can be established.

[0054] Inertial sensor-assisted positioning: Inertial sensors worn at key joints provide information about the athlete's joint angular velocity and acceleration. By integrating and processing these angular velocity and acceleration, the positional changes of the athlete's joints in the local coordinate system are determined. Combined with the previously acquired field coordinates, the joint positions are converted to the field coordinate system, enabling precise tracking and positioning of the athlete's movements.

[0055] Step S200: An AI processing unit, based on a pre-set volleyball movement database containing athletes of varying skill levels and physical conditions, stores a vast and diverse array of movement data, serving as a "knowledge repository" for AI analysis. Using deep learning algorithms and a mechanical model, the AI ​​analyzes the force generation and impact point of the hitting motion. For example, using Newton's laws of mechanics, combined with information on changes in the athlete's joint angles and body posture, the AI ​​calculates the magnitude and direction of the force applied at the moment of impact to determine whether the force generation is appropriate. The AI ​​also analyzes the position and trajectory of the athlete's body parts during impact to determine the accuracy of the impact point. Furthermore, the AI ​​processing unit analyzes the athlete's movement and positioning based on tactical scenarios. Taking into account factors such as the current match situation and the opponent's tactical deployment, the AI ​​compares the athlete's actual movement and positioning with the pre-set optimal tactical plan to assess the athlete's tactical execution performance. Through comprehensive analysis of these multi-dimensional information, the AI ​​processing unit obtains analytical data. These analytical data can accurately locate the problems of athletes in technical movements and tactical understanding and execution, providing a scientific basis for the subsequent generation of personalized training plans, making training more targeted and effectively improving athletes' technical level and tactical qualities.

[0056] Step S300: In this step, the AI ​​processing unit generates a personalized training plan based on the analysis data obtained in step S200 above, combined with the athlete's historical data and personalized characteristics. Historical data records the athlete's past training performance, technical improvement, and injury recovery information. By analyzing this data, we can understand the athlete's training progress and potential. Personalized characteristics include the athlete's physical fitness (such as height, weight, physical condition, etc.), technical characteristics (such as preferred offensive methods, defensive weaknesses, etc.), and learning ability.

[0057] Reinforcement learning algorithms play a key role in this process. Based on analytical data and individual athlete differences, they select the most appropriate training plan for each athlete from a wide range of training movements and intensity combinations. For athletes with poor physical fitness but a strong technical foundation, physical training may be prioritized to gradually improve their endurance and strength while maintaining an appropriate intensity in technical training. For athletes with significant technical weaknesses, specialized training exercises will be designed to address these shortcomings. For example, for athletes with unstable serves, a dedicated serving posture correction and force training plan will be developed. This personalized training plan fully considers each athlete's uniqueness, maximizing their strengths, addressing deficiencies, and improving training effectiveness and efficiency.

[0058] Step S400: The generated personalized training plan is fed back to the athlete in real-time through visual interaction, while explanation and guidance are provided to the athlete through voice interaction. Visual interaction includes, but is not limited to, presentation via display screens or projection equipment installed at the training venue, or smart wearable devices worn by the athlete. This visual interaction uses intuitive graphics and animations to display information such as the standard posture of training movements, the progress of the training plan, and the differences between the athlete's current movements and the standard movements, allowing the athlete to quickly and clearly understand their training status and areas for improvement.

[0059] Voice interaction provides athletes with detailed explanations and real-time guidance. As athletes perform training exercises, the voice interaction module provides timely instructions and suggestions based on their actual performance, such as "Arm raise angle isn't high enough, raise it 10 degrees more," or "Pay attention to the rhythm of force at the moment of hitting the ball," helping athletes better understand and execute training plans. Athletes train using visual and voice interaction, receiving timely feedback and adjusting their movements throughout training. This creates a closed-loop "training-feedback-improvement" system, making training more efficient and helping athletes quickly master correct technical movements and tactical awareness, continuously improving their training level and competitive ability.

[0060] Furthermore, when the fisheye camera, high-definition camera, and inertial sensor are installed, a unified three-dimensional coordinate system is established through a spatial calibration algorithm, so that the three-dimensional posture information, motion trajectory information, and joint angle change information collected by the fisheye camera, high-definition camera, and inertial sensor are in the same spatial reference; the spatial calibration algorithm is based on Zhang's calibration method and combines the calibration parameters of the inertial sensor for joint calibration.

[0061] This example describes the installation of a hybrid posture capture system in detail. For example, during installation in a volleyball stadium, technicians employed Zhang's calibration method. Calibration plates were placed inside the venue, and the fisheye camera and high-definition camera captured images of the plates from different angles. Combined with the inertial sensor's factory calibration parameters, a joint calibration was performed, creating a unified 3D coordinate system. This ensured that the images captured by the fisheye and high-definition cameras corresponded to the joint angle change information collected by the inertial sensors, avoiding data errors caused by inconsistent coordinate systems and ensuring the accuracy and consistency of data acquisition.

[0062] Furthermore, the horizontal viewing angle of the fisheye camera is greater than or equal to 180°, the vertical viewing angle is greater than or equal to 150°, and the fisheye camera is equipped with an image distortion correction algorithm to perform real-time distortion correction on the fisheye image when the fisheye camera collects the fisheye image; the resolution of the high-definition camera is greater than or equal to 4K, and the frame rate is greater than or equal to 60fps; the measurement accuracy of the inertial sensor is less than ±0.5° in angle error and less than ±2.1m / s in acceleration error2 .

[0063] In this embodiment, the fisheye camera, high-definition camera and inertial sensor of the hybrid posture capture system will be described in detail.

[0064] Fisheye cameras use ultra-wide-angle lenses with a horizontal viewing angle greater than or equal to 180° and a vertical viewing angle greater than or equal to 150°. For example, some models offer a horizontal viewing angle of 200° and a vertical viewing angle of 160°, capable of covering the entire volleyball training field, allowing the AI ​​processing unit to fully grasp the players' positions on the field. Furthermore, their built-in image distortion correction algorithm processes images immediately after capture, restoring image distortion caused by wide-angle shooting to normal.

[0065] The HD camera has a resolution greater than or equal to 4K (3840×2160) and a frame rate greater than or equal to 60fps. When athletes perform quick spikes, saves, and other actions, it can clearly capture details such as the instantaneous force of the athletes' muscles and the subtle movements of their fingers.

[0066] The high precision of inertial sensors ensures data reliability. For example, when measuring the angle of an athlete's knee joint, the error is less than ±0.5°, and when measuring acceleration, the error is less than ±0.1m / s. 2 , enabling the system to accurately obtain the motion information of the athlete's joints.

[0067] Furthermore, the fisheye camera is used to obtain panoramic data of the training ground, the high-definition camera is used to obtain the athlete's motion data, and the inertial sensor is used to obtain the athlete's joint motion data; after pre-processing the panoramic data, motion data, and joint motion data, three-dimensional posture information, motion trajectory information, and joint angle change information are obtained based on the pre-processed panoramic data, and / or motion data, and / or joint motion data; the pre-processing includes denoising and data compression.

[0068] In this embodiment, the camera, high-definition camera, and inertial sensor in the hybrid posture capture system are described in detail. The camera, high-definition camera, and inertial sensor generate a large amount of information, so the various collected information needs to be preprocessed before transmission to improve transmission efficiency. Preprocessing includes, but is not limited to, denoising and data compression. For example, a fisheye camera captures panoramic data of the training ground, and a Kalman filter is used to remove noise interference from the panoramic data. The high-definition camera captures athlete motion data, and a Kalman filter is also used to remove noise interference from the motion data. After the inertial sensor captures the athlete's joint motion data, the joint motion data is compressed. The data compression process uses the H.265 video coding standard and a lossless compression algorithm to compress the data, thereby reducing transmission pressure and improving transmission efficiency.

[0069] Furthermore, in the step of analyzing the training data based on the AI ​​processing unit having a preset volleyball action database containing athletes of different technical levels and physical conditions to obtain analysis data, the step further includes analyzing the angular velocity ω and angular acceleration α at the moment when the athlete hits the volleyball, so as to evaluate the stability of the athlete's hitting action based on the angular velocity ω and angular acceleration α. ​​When the angular velocity ω is greater than an angular velocity threshold and / or when the angular acceleration α is greater than an angular acceleration threshold, it is determined that the stability of the athlete's hitting action is insufficient. The calculation formula of the angular velocity ω is: The calculation formula for angular acceleration α is: Among them, Δθ represents the change in joint angle, and Δt represents the time interval.

[0070] In this embodiment, step S200 will be described in detail again. In volleyball training, the stability of the hitting action is crucial to the effectiveness of the shot. Simply calculating the force applied at the moment of hitting and determining the deviation of the hitting point cannot fully assess the stability of the athlete's hitting action. For example, two athletes may generate the same force at the moment of hitting and have a small deviation in the hitting point, but due to different stability of joint motion during the force application process, there may be significant differences in the quality of the shot. Therefore, introducing angular velocity ω and angular acceleration α to analyze the stability of the hitting action from the perspective of the dynamic motion process can more accurately identify problems with the athlete's action, thereby formulating more effective training programs. More specifically, when evaluating the stability of the athlete's hitting action based on the angular velocity ω and angular acceleration α, an angle threshold and an angular acceleration threshold are set for judgment. When the angular velocity ω is greater than the angular velocity threshold and / or when the angular acceleration α is greater than the angular acceleration threshold, the athlete's hitting action is judged to be unstable.

[0071] The calculation formula for angular velocity ω is: The calculation formula for angular acceleration α is: Among them, Δθ represents the change in joint angle, and Δt represents the time interval.

[0072] Furthermore, after the step of generating a personalized training plan based on the analyzed data and in combination with the athlete's historical data and personalized characteristics, the method further includes the following steps: generating a training effect estimation report based on the personalized training plan and the athlete's historical data and personalized characteristics, and using the analytic hierarchy process to assign weights to various movement indicators of the athlete during volleyball training to construct a comprehensive evaluation model, and calculating the athlete's training effect according to the comprehensive evaluation model for quantitative evaluation. The scoring calculation formula is: Among them, w i Expressed as the weight of the i-th action indicator, x i It is represented as the score of the i-th action indicator, and n is represented as the total number of action indicators involved in the comprehensive evaluation calculation.

[0073] In this embodiment, the volleyball training method based on AI posture capture will be described in detail again. After step S300, step S500 is also included: a training effect prediction report is generated based on the personalized training plan generated in step S300 and the athlete's historical data and personalized characteristics, and a hierarchical analysis method is used to assign weights to the various action indicators of the athlete during volleyball training, thereby constructing a comprehensive prediction model, and calculating the athlete's training effect according to the comprehensive evaluation model for quantitative evaluation. The scoring calculation formula is Among them, w i Expressed as the weight of the i-th action indicator, x i represents the score of the i-th action indicator, and n represents the total number of action indicators included in the comprehensive evaluation. For example, the weight of batting technique is set to 0.4, the weight of physical fitness is set to 0.3, and the weight of tactical understanding is set to 0.3. By calculating the scores of each indicator, for example, if the batting technique score is 80 points, the physical fitness score is 75 points, and the tactical understanding score is 85 points, then the quantitative evaluation score calculated using the above formula is 81 points.

[0074] Furthermore, a comprehensive scoring threshold is set in the comprehensive evaluation model; when the comprehensive evaluation model calculates the athlete's quantitative evaluation score, the quantitative evaluation score is compared with the comprehensive scoring threshold. When the quantitative evaluation score is less than the comprehensive scoring threshold, the training data is re-acquired to generate a new personalized training plan.

[0075] In this embodiment, a comprehensive scoring threshold is set within the comprehensive evaluation model. After the quantitative evaluation score is calculated according to the above embodiment, the quantitative evaluation score is compared with the comprehensive scoring threshold. If the quantitative evaluation score is less than the comprehensive scoring threshold, training data is re-acquired to generate a new personalized training plan. For example, if the quantitative evaluation score calculated in the above embodiment is 81, and the comprehensive scoring threshold is 80, a new personalized training plan is generated according to steps 100 to 300. If the quantitative evaluation score is 82, the personalized training plan is considered reasonable and can be directly applied to the athlete for training.

[0076] This embodiment also provides a volleyball training system based on AI posture capture, including an acquisition module, an analysis module, a training program generation module, and a training module; wherein the acquisition module is used to utilize a hybrid posture capture system composed of a fisheye camera, a high-definition camera, and an inertial sensor, and to collect training data of athletes during volleyball training in real time based on the hybrid posture capture system, wherein the training data includes at least three-dimensional posture information, motion trajectory information, and joint angle change information; the analysis module is connected to the acquisition module and is used to analyze the training data based on an AI processing unit that has a preset volleyball action database containing athletes of different technical levels and physical conditions to obtain analysis data, wherein the analysis data is obtained by analyzing the force generation method information and hitting point information of the hitting action based on a mechanical model, as well as the running information and standing information of the athlete in combination with tactical scenarios; the training program generation module is connected to the analysis module and is used to generate a personalized training program based on the analysis data and in combination with the athlete's historical data and personalized characteristics; the training module is connected to the training program generation module and is used to provide real-time feedback of the personalized training program to the athlete in a visual interactive manner, and at the same time, explain and guide the athlete through voice interaction based on the personalized training, and the athlete performs volleyball training based on the visual interaction and voice interaction.

[0077] Acquisition module: used to collect the training data of athletes during volleyball training in real time through a hybrid posture capture system composed of a fisheye camera, a high-definition camera and an inertial sensor. Specifically, the horizontal viewing angle of the fisheye camera is greater than or equal to 180°, the vertical viewing angle is greater than or equal to 150°, and the fisheye camera is deployed on the top of the training ground to obtain panoramic information. The resolution of the high-definition camera is greater than or equal to 4K, the frame rate is greater than or equal to 60fps, and the high-definition cameras are arranged around the training ground to focus on the details of the athletes' movements. The measurement accuracy of the inertial sensor is less than ±0.5° in terms of angle error and less than ±2.1m / s in terms of acceleration error. 2 , including but not limited to being worn on the athlete's shoulders, elbows, wrists and other key joints.

[0078] It should be noted that the training data includes at least three-dimensional posture information, motion trajectory information, and joint angle change information. This three-dimensional posture information is fused through triangulation using fisheye cameras and high-definition cameras with information obtained by inertial sensors to construct the three-dimensional coordinates of the athlete's joints throughout the body, thereby obtaining three-dimensional posture data. The motion trajectory information is obtained by tracking the athlete's movement trajectory in the training field in real time using a fisheye camera combined with a field coordinate system, thereby obtaining motion trajectory information. The joint angle change information is obtained by calculating the joint angle change based on the joint angular velocity ω and angular acceleration α collected in real time by the inertial sensors, thereby obtaining joint angle change information. The calculation formula for the joint angle change Δθ is: Δθ = ω·Δt.

[0079] It should be noted that the field coordinate system is a reference frame used to determine the positions of the players and volleyball on the training field. It is established in the following way:

[0080] Camera calibration: By setting up calibration objects at known locations in the training field, these objects are photographed by fisheye and HD cameras. Image processing algorithms are then used to calculate the internal parameters (such as focal length and principal point) and external parameters (such as rotation and translation) of the fisheye and HD cameras. This determines the position and attitude of the fisheye and HD cameras in the field coordinate system. This allows each pixel in the image captured by the camera to be mapped to its actual position in the field coordinate system through coordinate transformation.

[0081] Field feature point extraction: Identify fixed feature points on the training field, such as the four corners of the field and the intersection of the centerline and sidelines. The locations of these feature points in the field coordinate system are known. By detecting the locations of these feature points in images captured by fisheye cameras and high-definition cameras, a mapping relationship between image coordinates and field coordinates can be established.

[0082] Inertial sensor-assisted positioning: Inertial sensors worn at key joints provide information about the athlete's joint angular velocity and acceleration. By integrating and processing these angular velocity and acceleration, the positional changes of the athlete's joints in the local coordinate system are determined. Combined with the previously acquired field coordinates, the joint positions are converted to the field coordinate system, enabling precise tracking and positioning of the athlete's movements.

[0083] The analysis module is designed to utilize an AI processing unit based on a pre-set volleyball movement database containing a wealth of diverse movement data from athletes of varying skill levels and physical conditions. This database serves as a "knowledge repository" for the AI ​​processing unit to analyze. Using deep learning algorithms and a mechanical model, it analyzes the force generation and impact point of the hitting motion. For example, using Newton's laws of mechanics, combined with information on changes in the athlete's joint angles and body posture, it calculates the magnitude and direction of the force applied at the moment of impact to determine whether the force generation is appropriate. The AI ​​processing unit also determines the accuracy of the impact point by analyzing the position and trajectory of various body parts during the hit. Furthermore, the AI ​​processing unit analyzes the athlete's movement and positioning based on tactical scenarios. Taking into account factors such as the current game situation and the opponent's tactical deployment, it compares the athlete's actual movement and positioning with the pre-set optimal tactical plan to assess the athlete's tactical execution performance. Through comprehensive analysis of these multi-dimensional information, the AI ​​processing unit obtains analytical data. These analytical data can accurately locate the problems of athletes in technical movements and tactical understanding and execution, providing a scientific basis for the subsequent generation of personalized training plans, making training more targeted and effectively improving athletes' technical level and tactical qualities.

[0084] Training plan generation module: The AI ​​processing unit generates a personalized training plan based on the analysis data obtained by the above analysis module, combined with the athlete's historical data and personalized characteristics. Historical data records the athlete's past training performance, technical improvement, injury recovery and other information. By analyzing this data, we can understand the athlete's training progress and potential. Personalized characteristics include the athlete's physical fitness (such as height, weight, physical condition, etc.), technical characteristics (such as their preferred offensive methods, defensive weaknesses, etc.), and learning ability.

[0085] Reinforcement learning algorithms play a key role in this process. Based on analytical data and individual athlete differences, they select the most appropriate training plan for each athlete from a wide range of training movements and intensity combinations. For athletes with poor physical fitness but a strong technical foundation, physical training may be prioritized to gradually improve their endurance and strength while maintaining an appropriate intensity in technical training. For athletes with significant technical weaknesses, specialized training exercises will be designed to address these shortcomings. For example, for athletes with unstable serves, a dedicated serving posture correction and force training plan will be developed. This personalized training plan fully considers each athlete's uniqueness, maximizing their strengths, addressing deficiencies, and improving training effectiveness and efficiency.

[0086] Training Module: This module provides real-time feedback to athletes on personalized training plans through interactive visual feedback, while also providing explanations and guidance via voice interaction. This interactive visualization includes, but is not limited to, presentations on display screens and projection equipment at the training venue, or on smart wearable devices worn by athletes. This intuitive graphics and animations showcase information such as standard training postures, training plan progress, and the differences between the athlete's current movements and standard movements, allowing athletes to quickly and clearly understand their training progress and areas for improvement.

[0087] Voice interaction provides athletes with detailed explanations and real-time guidance. As athletes perform training exercises, the voice interaction module provides timely instructions and suggestions based on their actual performance, such as "Arm raise angle isn't high enough, raise it 10 degrees more," or "Pay attention to the rhythm of force at the moment of hitting the ball," helping athletes better understand and execute training plans. Athletes train using visual and voice interaction, receiving timely feedback and adjusting their movements throughout training. This creates a closed-loop "training-feedback-improvement" system, making training more efficient and helping athletes quickly master correct technical movements and tactical awareness, continuously improving their training level and competitive ability.

[0088] Furthermore, a scoring calculation module is included; wherein the scoring calculation module is connected to the training program generation module and is used to generate a training effect estimation report based on the personalized training program and the athlete's historical data and personalized characteristics, and adopts the hierarchical analysis method to assign weights to various action indicators of the athlete during volleyball training to build a comprehensive evaluation model, and calculate the athlete's training effect according to the comprehensive evaluation model for quantitative evaluation. The scoring calculation formula is: Among them, w i Expressed as the weight of the i-th action indicator, x i It is represented as the score of the i-th action indicator, and n is represented as the total number of action indicators involved in the comprehensive evaluation calculation.

[0089] After the training program generation module, a scoring calculation module is also included. This module is used to generate a training effect prediction report based on the personalized training program generated by the training program generation module and the athlete's historical data and personalized characteristics. It also uses the hierarchical analysis method to assign weights to the various action indicators of the athletes during volleyball training, thereby building a comprehensive prediction model and calculating the athlete's training effect according to the comprehensive evaluation model for quantitative evaluation. The scoring calculation formula is Among them, w i Expressed as the weight of the i-th action indicator, x irepresents the score of the i-th action indicator, and n represents the total number of action indicators included in the comprehensive evaluation. For example, the weight of batting technique is set to 0.4, the weight of physical fitness is set to 0.3, and the weight of tactical understanding is set to 0.3. By calculating the scores of each indicator, for example, if the batting technique score is 80 points, the physical fitness score is 75 points, and the tactical understanding score is 85 points, then the quantitative evaluation score calculated using the above formula is 81 points.

[0090] Furthermore, a comprehensive scoring threshold is provided in the comprehensive evaluation model, and the scoring calculation module is also used to compare the quantitative evaluation score with the comprehensive scoring threshold when the comprehensive evaluation model calculates the quantitative evaluation score of the athlete. When the quantitative evaluation score is less than the comprehensive scoring threshold, the training data is re-acquired to generate a new personalized training plan.

[0091] In this embodiment, a comprehensive scoring threshold is set within the comprehensive evaluation model. After the quantitative evaluation score is calculated according to the above embodiment, the quantitative evaluation score is compared with the comprehensive scoring threshold. If the quantitative evaluation score is less than the comprehensive scoring threshold, training data is re-acquired to generate a new personalized training plan. For example, if the quantitative evaluation score calculated in the above embodiment is 81, and the comprehensive scoring threshold is 80, a new personalized training plan is generated according to steps 100 to 300. If the quantitative evaluation score is 82, the personalized training plan is considered reasonable and can be directly applied to the athlete for training.

[0092] It should be noted that the embodiments of the present invention have better practicability and do not impose any form of limitation on the present invention. Any technician familiar with the field may use the technical content disclosed above to change or modify it into an equivalent effective embodiment. However, any modification or equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A volleyball training method based on AI posture capture, characterized in that: The following steps are involved: A hybrid posture capture system is formed using a fisheye camera, a high-definition camera, and an inertial sensor. Training data of athletes during volleyball training is collected in real time based on the hybrid posture capture system. The training data includes at least three-dimensional posture information, motion trajectory information, and joint angle change information. An AI processing unit, based on a preset volleyball movement database containing athletes of different skill levels and physical conditions, analyzes the training data to obtain analytical data, wherein the analytical data is obtained by analyzing information on the force generation method and hitting point of the hitting movement based on a mechanical model, as well as information on the movement and stance of the athletes in combination with tactical scenarios; generating a personalized training program based on the analyzed data and in combination with the athlete's historical data and personalized characteristics; The personalized training program is fed back to the athlete in real time in a visual interactive manner, and the athlete is explained and guided by voice interaction according to the personalized training. The athlete performs volleyball training according to the visual interaction and the voice interaction.

2. The volleyball training method based on AI posture capture according to claim 1, characterized in that: When the fisheye camera, the high-definition camera, and the inertial sensor are installed, a unified three-dimensional coordinate system is established through a spatial calibration algorithm, so that the three-dimensional posture information, the motion trajectory information, and the joint angle change information collected by the fisheye camera, the high-definition camera, and the inertial sensor are in the same spatial reference; The spatial calibration algorithm performs joint calibration based on Zhang's calibration method and the calibration parameters of the inertial sensor.

3. The volleyball training method based on AI posture capture according to claim 2, characterized in that: The horizontal viewing angle of the fisheye camera is greater than or equal to 180°, and the vertical viewing angle is greater than or equal to 150°, and the fisheye camera is configured with an image distortion correction algorithm to perform real-time distortion correction on the fisheye image when the fisheye camera captures the fisheye image; The resolution of the high-definition camera is greater than or equal to 4K, and the frame rate is greater than or equal to 60fps; The inertial sensor has an angular error of less than ±0.5° and an acceleration error of less than ±2.1 m / s. 2 .

4. The volleyball training method based on AI posture capture according to claim 3 is characterized in that: The fisheye camera is used to obtain panoramic data of the training ground, the high-definition camera is used to obtain the athlete's movement data, and the inertial sensor is used to obtain the athlete's joint movement data; After preprocessing the panoramic data, the motion data, and the joint motion data, the three-dimensional posture information, the motion trajectory information, and the joint angle change information are obtained based on the preprocessed panoramic data, and / or the motion data, and / or the joint motion data; The preprocessing includes noise removal and data compression.

5. The volleyball training method based on AI gesture capture according to claim 1, characterized in that: The step of analyzing the training data to obtain analysis data based on an AI processing unit preset with a volleyball action database containing athletes of different skill levels and physical conditions further includes analyzing the angular velocity ω and angular acceleration α at the moment of the athlete hitting the volleyball, thereby evaluating the stability of the athlete's hitting action based on the angular velocity ω and the angular acceleration α, and determining that the athlete's hitting action is unstable when the angular velocity ω is greater than an angular velocity threshold and / or when the angular acceleration α is greater than an angular acceleration threshold; The calculation formula of the angular velocity ω is: The calculation formula of the angular acceleration α is: Among them, Δθ represents the change in joint angle, and Δt represents the time interval.

6. The volleyball training method based on AI gesture capture according to claim 1, characterized in that: After the step of generating a personalized training program based on the analyzed data and in combination with the athlete's historical data and personalized characteristics, the following steps are also included: A training effect estimation report is generated based on the personalized training program and the athlete's historical data and personalized characteristics. A hierarchical analysis method is used to assign weights to various movement indicators of the athlete during volleyball training to construct a comprehensive evaluation model. The athlete's training effect is calculated based on the comprehensive evaluation model for quantitative evaluation. The scoring calculation formula is: Among them, w i Expressed as the weight of the i-th action indicator, x i It is represented as the score of the i-th action indicator, and n is represented as the total number of action indicators involved in the comprehensive evaluation calculation.

7. The volleyball training method based on AI gesture capture according to claim 6, characterized in that: A comprehensive scoring threshold is provided within the comprehensive evaluation model; When the comprehensive evaluation model calculates the quantitative evaluation score of the athlete, the quantitative evaluation score is compared with the comprehensive scoring threshold. When the quantitative evaluation score is less than the comprehensive scoring threshold, the training data is re-acquired to generate a new personalized training plan.

8. A volleyball training system based on AI posture capture, characterized in that: It includes acquisition module, analysis module, training program generation module and training module; among them, The acquisition module is used to use a hybrid posture capture system composed of a fisheye camera, a high-definition camera, and an inertial sensor to collect training data of athletes during volleyball training in real time based on the hybrid posture capture system. The training data includes at least three-dimensional posture information, motion trajectory information, and joint angle change information; The analysis module is connected to the acquisition module and is used to analyze the training data based on an AI processing unit that has a preset volleyball action database containing athletes of different skill levels and physical conditions to obtain analysis data. The analysis data is obtained by analyzing the force generation method and hitting point information of the hitting action based on a mechanical model, as well as the movement and stance information of the athletes in combination with tactical scenarios; A training program generation module: connected to the analysis module, used to generate a personalized training program based on the analysis data and in combination with the athlete's historical data and personalized characteristics; Training module: connected to the training program generation module, used to provide real-time feedback of the personalized training program to the athlete in a visual interactive manner, and at the same time explain and guide the athlete through voice interaction based on the personalized training. The athlete performs volleyball training based on the visual interaction and the voice interaction.

9. The volleyball training system based on AI gesture capture according to claim 8, characterized in that: It also includes a scoring calculation module; wherein, The scoring calculation module is connected to the training program generation module and is used to generate a training effect estimation report based on the personalized training program and the athlete's historical data and personalized characteristics, and use the hierarchical analysis method to assign weights to various action indicators of the athlete during volleyball training to build a comprehensive evaluation model, and calculate the athlete's training effect according to the comprehensive evaluation model for quantitative evaluation. The scoring calculation formula is: Among them, w i Expressed as the weight of the i-th action indicator, x i It is represented as the score of the i-th action indicator, and n is represented as the total number of action indicators involved in the comprehensive evaluation calculation.

10. The volleyball training system based on AI gesture capture according to claim 9, characterized in that: A comprehensive scoring threshold is provided in the comprehensive evaluation model, and the scoring calculation module is further used to compare the quantitative evaluation score with the comprehensive scoring threshold when the comprehensive evaluation model calculates the quantitative evaluation score of the athlete. When the quantitative evaluation score is less than the comprehensive scoring threshold, the training data is reacquired to generate a new personalized training plan.

Images