A method and system for updating sports event scores

Abstract

The present application provides a method and system for updating the scoring of sports events. Among them, the mechanical data and multi-view video data of the sensors inside the sphere and the players' protective gear are obtained, the three-dimensional position coordinates are calculated based on the frequency change and the elastic coefficient of the sphere material, the force direction is reversely deduced through the phase parameter, and the motion state fusion data is aligned with the video time stamp. The three-dimensional spatial motion sequence of the sphere is generated by cross-frame matching, and the joint torque parameters are extracted in combination with the dynamic modeling of the skeleton. After verifying the consistency of the sphere trajectory, the joint torque and the force direction are fused to generate composite event parameters, and the score results are updated by matching the preset event rules database, and dynamically superimposed on the multi-view picture transmission terminal. The technical solution provided by this application realizes the precise synchronous analysis of motion trajectory and biomechanical parameters, supports the accuracy of event penalties and the visualization of multi-view live broadcast data.

Description

Technical Field

[0001] The present application relates to the technical field of score updating, and in particular to a method and system for updating scores of sports events. Background Art

[0002] Scoring updates for sports events must address issues such as the lack of real-time and authoritative judgments on controversial actions (such as whether the ice hockey puck has crossed the line or disputes over the completion of gymnastics movements), low efficiency in collaborative decision-making using multi-source heterogeneous data (referee voice commands, video streams, and electronic scoring equipment), and delays in synchronization of scoring status across platforms (referee systems, live streams, and on-site large screens). These issues require that scoring update methods be able to: parse the core semantics of highly dynamic competition actions (such as equipment contact boundaries and movement standardization); integrate unstructured data (emotional analysis of referee voice commands and semantic segmentation of video key frames) to generate a traceable judgment basis chain; and achieve millisecond-level state synchronization across multiple terminals to ensure seamless integration between referee decisions and event presentation systems.

[0003] The current mainstream solution adopts a voiceprint-driven visual attention network and distributed consensus protocol fusion architecture: by deploying a directional microphone array to capture the voiceprint characteristics of the referee's instructions, and combining it with the voice emotion recognition model to judge the judgment tendency; using the visual attention network to focus on controversial areas in the video stream (such as the boundary line of the ball landing point) to generate a high-precision semantic segmentation heat map; using the blockchain consensus mechanism to perform multi-node verification on the judgment results of the referee end, video analysis module, and scoring terminal, triggering the score update command through the smart contract, and synchronizing it to the live stream barrage engine and the on-site large-screen control system.

[0004] However, voiceprint emotion recognition models lack cross-lingual generalization for referee instructions in multiple languages ​​(for example, the dialect intonation of referees speaking minority languages ​​varies), leading to an increased rate of biased misjudgments. Visual attention networks also experience a sharp drop in semantic segmentation accuracy in low-light, high-speed, and blurred scenes (for example, identifying offside lines in nighttime soccer matches). Furthermore, the high throughput requirements of blockchain consensus mechanisms conflict with the real-time nature of match refereeing, necessitating manual intervention for synchronization. Summary of the Invention

[0005] The present application provides a method and system for updating scores of sports events, which are used to solve the problems of dynamic event determination delay and rule matching inaccuracy in the prior art.

[0006] In a first aspect, the present application provides a method for updating scores of sports events, comprising:

[0007] Acquire mechanical data from sensors deployed inside the ball and on the players' protective gear, as well as multi-view video data captured by multi-view cameras. The mechanical data includes frequency changes when the ball is impacted and phase change parameters of the players' limbs contacting the ball.

[0008] Based on the correlation between the frequency variation and a preset elastic coefficient of the sphere material, a three-dimensional position coordinate sequence of the sphere's motion trajectory is calculated. Simultaneously, a force direction parameter of the player's limb movement is reversely deduced using the phase variation parameter. The three-dimensional position coordinate sequence and the force direction parameter are aligned with the time stamp of the multi-view video data to obtain motion state fusion data.

[0009] Performing cross-frame matching of the sphere outline based on the motion state fusion data to generate a three-dimensional spatial motion sequence of the sphere's flight trajectory, and performing skeletal dynamic modeling of the player's posture based on the motion state fusion data to generate joint torque parameters of the player's action;

[0010] Performing trajectory consistency verification on the three-dimensional position coordinate sequence of the spherical motion trajectory and the three-dimensional space motion sequence, and generating composite event parameters by combining the joint torque parameters and the force direction parameters;

[0011] Based on the composite event parameters, rule matching is performed with a preset ball game rule database to update the game score result, and the updated score result is dynamically superimposed on the target screen of the multi-view video data and transmitted to the terminal display device.

[0012] Optionally, calculating a three-dimensional position coordinate sequence of the sphere's motion trajectory based on a correlation between the frequency variation and a preset elastic coefficient of the sphere material includes:

[0013] Obtaining a frequency change of the sphere when it is impacted from a sensor inside the sphere, wherein the frequency change is the difference between the vibration frequency of the sphere surface and the frequency of the initial static state;

[0014] According to the preset correlation relationship between the elastic coefficient of the sphere material, the frequency change is converted into the deformation recovery force of the sphere when it is impacted. The deformation recovery force is the force generated when the sphere material returns to its original shape after deformation;

[0015] Based on the deformation recovery force and the mass of the sphere, the initial acceleration of the sphere after the impact is calculated;

[0016] Calculate the instantaneous velocity of the sphere after the impact based on the initial acceleration and the motion state of the sphere before the impact;

[0017] Based on the instantaneous velocity and in combination with the time increment, the position coordinates of the sphere in the three-dimensional space are iteratively calculated to form a three-dimensional position coordinate sequence of the sphere's motion trajectory.

[0018] The reverse derivation of the force direction parameter of the player's limb movement by the phase change parameter includes:

[0019] Determining the contact time point and contact duration of the player's limbs and the ball based on the phase change parameters;

[0020] Based on the contact time point and contact duration, combined with the spatial distribution of protective gear sensors, the movement direction of the player's limb when contacting the ball is calculated;

[0021] According to the movement direction and the magnitude of the force applied by the protective gear sensor, the direction of the force applied by the player's limbs to the ball is reversely deduced.

[0022] Optionally, aligning the three-dimensional position coordinate sequence, the force direction parameter, and the time stamp of the multi-view video data to obtain motion state fusion data includes:

[0023] Extracting the time stamp of each frame in the multi-view video data, wherein the time stamp is the precise time stamp recorded by the video acquisition device;

[0024] Matching the three-dimensional position coordinate sequence and instantaneous velocity sequence of the sphere's motion trajectory with the time stamps of the multi-view video data according to the corresponding timestamps;

[0025] Match the force direction of the player's body movements with the time stamps of the multi-view video data according to the corresponding timestamps;

[0026] The matched three-dimensional position coordinates, instantaneous velocity and force direction are combined with the corresponding frames of the multi-view video data to form motion state fusion data.

[0027] Optionally, performing cross-frame matching of the sphere outline based on the motion state fusion data to generate a three-dimensional spatial motion sequence of the sphere's flight trajectory, and performing skeletal dynamic modeling of the player's posture based on the motion state fusion data to generate joint torque parameters of the player's action include:

[0028] Extracting the three-dimensional position coordinates of the sphere's flight trajectory from the motion state fusion data, and calculating the sphere's cross-frame motion trajectory sequence based on the sphere's contour shape differences between adjacent frames in the time stamps of the multi-view video data;

[0029] Based on the temporal continuity of the three-dimensional position coordinates of the sphere in the cross-frame motion trajectory sequence, a three-dimensional spatial motion sequence of the sphere's flight trajectory is generated, wherein the three-dimensional spatial motion sequence includes the displacement direction, instantaneous velocity, and acceleration parameters of the sphere;

[0030] Extracting force direction parameters of the player's movements from the motion state fusion data, combining them with the spatial displacements of the player's joint nodes at adjacent time markers in the multi-view video data, constructing a skeletal dynamic model of the player's posture, and determining a joint node angle change sequence and a bone length change sequence based on the skeletal dynamic model;

[0031] According to the angle change sequence of the joint node and the bone length change sequence, combined with the action vector of the force direction parameter, the joint torque parameter of the player's action is calculated. The joint torque parameter includes the joint rotation direction and the torque intensity ratio.

[0032] Optionally, extracting force direction parameters of the player's actions from the motion state fusion data, combining the spatial displacements of the player's joint nodes at adjacent time markers in the multi-view video data, and constructing a skeletal dynamic model of the player's posture; determining an angle change sequence and a bone length change sequence of the joint nodes based on the skeletal dynamic model, including:

[0033] Extracting a force direction parameter of the player's action from the motion state fusion data, and generating an action vector of the force direction parameter by obtaining a horizontal action angle and a vertical action angle of the force direction parameter in three-dimensional space;

[0034] Based on the spatial displacements of the joint nodes of the same player at consecutive time marks in the multi-view video data, a joint point displacement trajectory reflecting the movement trend is formed by correlating the vector changes of the spatial displacements of the joint nodes at adjacent time marks;

[0035] Based on the fluctuation characteristics of the action vector of the force direction parameter on the time axis and the continuity of the spatial displacement trajectory of the joint point, a skeletal dynamic model of the player's posture is constructed, wherein the skeletal dynamic model includes the initial angles of the joint nodes and the initial lengths of the bones;

[0036] generating an angle change sequence of the joint nodes based on the initial angles of the joint nodes in the skeletal dynamic model and the fluctuation amplitude of the displacement direction in the spatial displacement trajectory of the joint points;

[0037] A bone length change sequence is generated based on the initial length of the bone in the bone dynamic model and the cumulative difference of the displacement distance in the spatial displacement trajectory of the joint point.

[0038] Optionally, the three-dimensional position coordinate sequence of the spherical motion trajectory is subjected to trajectory consistency verification with the three-dimensional space motion sequence, and the joint torque parameter and the force direction parameter are combined to generate a composite event parameter, including:

[0039] Obtaining a displacement direction parameter by analyzing position changes of consecutive time stamps in the three-dimensional position coordinate sequence, and performing trajectory consistency comparison between the displacement direction parameter and the displacement direction parameter in the three-dimensional space motion sequence to output a trajectory matching degree;

[0040] When the trajectory matching degree is lower than a preset trajectory tolerance threshold, marking an abnormal trajectory segment in the three-dimensional space motion sequence that does not match the spherical motion trajectory, and updating the displacement direction parameter of the three-dimensional space motion sequence according to the abnormal trajectory segment;

[0041] Extracting the joint rotation direction parameter and the torque intensity ratio parameter from the joint torque parameter, and combining the action vector of the force direction parameter to calculate the spatial angle between the force direction parameter and the joint rotation direction parameter and the ratio difference between the torque intensity and the action vector;

[0042] When the spatial angle is less than the preset angle tolerance and the proportional difference is less than the preset strength tolerance, the joint torque parameter and the force direction parameter at the current moment are bound to form a valid action parameter group;

[0043] The updated displacement direction parameter and the effective action parameter group are spliced ​​according to the timestamp to generate a composite event parameter.

[0044] Optionally, matching the composite event parameters with a preset ball game rule database to update the game score result includes:

[0045] Extracting a valid action parameter group from the composite event parameters, and matching the valid action parameter group with the rule compliance of goal and foul determination conditions in a preset ball game rule database, and generating a corresponding trigger event mark when the rule compliance reaches a preset threshold;

[0046] When the trigger event mark satisfies the goal determination condition, a score change instruction is generated according to the overlapping relationship between the end position of the ball motion trajectory and the goal area, so as to update the event score result according to the score change instruction;

[0047] When the trigger event mark satisfies the foul determination condition, a foul level mark is generated according to the matching degree between the force direction parameter and the preset illegal action direction, so as to update the event score result according to the foul level mark.

[0048] In a second aspect, the present application provides a system for updating scores of sports events, including:

[0049] An acquisition module is used to obtain mechanical data fed back by sensors deployed inside the ball and on the player's protective gear, as well as multi-view video data collected by multi-view cameras. The mechanical data includes frequency changes when the ball is impacted and phase change parameters of the player's limb contact;

[0050] a calculation module, configured to calculate a three-dimensional position coordinate sequence of the ball's motion trajectory based on a correlation between the frequency variation and a preset elastic coefficient of the ball's material, and simultaneously reversely derive a force direction parameter of the player's limb movement using the phase variation parameter, and align the three-dimensional position coordinate sequence and the force direction parameter with a time stamp of the multi-view video data to obtain motion state fusion data;

[0051] a generation module, configured to perform cross-frame matching of the sphere outline based on the motion state fusion data to generate a three-dimensional spatial motion sequence of the sphere's flight trajectory, and perform skeletal dynamic modeling of the player's posture based on the motion state fusion data to generate joint torque parameters of the player's action;

[0052] a verification module, configured to verify the consistency of the three-dimensional position coordinate sequence of the spherical motion trajectory with the three-dimensional spatial motion sequence, and generate a composite event parameter by combining the joint torque parameter and the force direction parameter;

[0053] The matching module is used to match the rules of the composite event parameters with the preset ball game rule database to update the game score result, and dynamically superimpose the updated score result on the target screen of the multi-view video data to transmit it to the terminal display device.

[0054] In a third aspect, the present application provides a computing device comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for updating the score of a sports event as described in the first aspect above.

[0055] In a fourth aspect, the present application provides a computer storage medium storing a computer program, which, when executed by a computer, implements a method for updating sports event scores as described in the first aspect.

[0056] The present application obtains mechanical data fed back by sensors deployed inside the ball and on the player's protective gear, as well as multi-view video data collected by a multi-view camera device. The mechanical data includes the frequency change when the ball is impacted and the phase change parameters of the player's limb contact, so as to establish a multimodal data acquisition channel and synchronously capture physical motion characteristics; the three-dimensional position coordinate sequence of the ball's motion trajectory is calculated based on the correlation between the frequency change and the preset elastic coefficient of the ball material, and the force direction parameters of the player's limb movement are reversely deduced through the phase change parameters, and the three-dimensional position coordinate sequence, the force direction parameters and the time stamps of the multi-view video data are aligned to obtain motion state fusion data, so as to realize physical reconstruction of the ball's motion trajectory and reverse modeling of the player's force behavior; cross-frame matching of the ball contour is performed based on the motion state fusion data to generate The three-dimensional spatial motion sequence of the ball's flight trajectory is obtained, and the skeletal dynamic modeling of the player's posture is performed based on the motion state fusion data to generate joint torque parameters, which can improve the cross-view continuity of the ball's motion trajectory and quantify the mechanical correlation of limb movements; by verifying the trajectory consistency of the three-dimensional position coordinate sequence of the ball's motion trajectory with the three-dimensional spatial motion sequence, and combining the joint torque parameters with the force direction parameters to generate composite event parameters, it is possible to eliminate multi-source data errors and construct an event judgment basis that associates physical authenticity with action logic; by matching the composite event parameters with a preset ball competition rule database to update the competition score results, and dynamically superimposing the updated score results on the target screen of the multi-view video data and transmitting it to the terminal display device, it is possible to achieve real-time competition judgment and enhance the information fusion of the multi-view live broadcast screen.

[0057] Furthermore, by converting the frequency change of the ball when it is impacted into a deformation recovery force based on the elastic coefficient of the material, and combining the mass parameters to calculate the initial acceleration and instantaneous velocity, the dynamic coupling modeling of the impact force-motion state is realized, thereby improving the physical reality of the trajectory prediction; by iteratively calculating the three-dimensional position coordinate sequence, the influence of air resistance and rotation effect on the trajectory is dynamically corrected to ensure the spatial continuity and temporal synchronization of the motion trajectory; at the same time, the time point and direction of the player's limb contact are analyzed based on the phase change parameters, and the force direction is reversely deduced based on the spatial distribution of the protective gear sensor and the force data, accurately quantifying the mechanical correlation between limb movements and ball movement, and eliminating the subjective errors of motion capture in traditional visual analysis; finally, through multimodal data fusion and physical parameter iteration, a two-way verification mechanism for the ball's motion trajectory and the player's force behavior is constructed, which enhances the objectivity of event judgment and provides high-precision data support for three-dimensional trajectory reconstruction and motion reverse modeling.

[0058] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0060] Figure 1 A flowchart showing a method for updating sports event scores provided by the present application is shown;

[0061] Figure 2 A schematic diagram showing the structure of a sports event score updating system provided by the present application is shown;

[0062] Figure 3 A schematic structural diagram of a computing device provided by the present application is shown. DETAILED DESCRIPTION

[0063] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0064] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to be different types.

[0065] Researchers have found that existing sports event motion capture technology has problems such as temporal and spatial mismatch between sensor data and video images, insufficient accuracy in reverse derivation of motion parameters, and delayed rule matching, resulting in score determination relying on manual labor and limited real-time performance. Based on this, a method for dynamic scoring and multi-source fusion of ball events is provided. This method can achieve automated synchronization of motion parameters and rule determination through spatiotemporal alignment of mechanical data and multi-view videos and three-dimensional trajectory verification. The technical solution of this application can be applied to ball event scoring, motion analysis, and referee-assisted decision-making scenarios.

[0066] The entire R&D process embodies a deep collaborative mechanism of multi-source data fusion and dynamic rule matching, aiming to overcome the defects of existing solutions caused by data fragmentation leading to scoring lags, accumulated trajectory calculation errors, and disconnection between action parameters and rules. Through the precise alignment of mechanical data and video time stamps, the reliability bottleneck of traditional single sensors or visual analysis is broken through; based on the joint verification of cross-frame matching and skeletal modeling, the problems of motion trajectory drift and misjudgment of action intentions are solved; combined with the dynamic matching of composite event parameters and the rule database, the automation and real-time nature of scoring are achieved; and finally, through the superposition and transmission of multi-source data, millisecond-level synchronization of event results and screen presentation is achieved. This application forms a closed-loop optimization from data fusion to rule mapping, significantly improving the accuracy of scoring and the efficiency of event analysis.

[0067] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0068] Figure 1 The present invention provides a flowchart of a method for updating the score of a sports event, such as Figure 1 As shown, the method includes:

[0069] 101. Obtain mechanical data fed back by sensors deployed inside the ball and on the player's protective gear, as well as multi-view video data captured by a multi-view camera. The mechanical data includes frequency changes when the ball is impacted and phase change parameters of the player's limb contact.

[0070] In this step, mechanical data refers to the physical quantities of force or motion detected by sensors. Frequency change refers to the difference in vibration frequency of a ball relative to its initial static state when impacted. Phase change parameters refer to the temporal phase difference of force applied during a player's physical contact event as detected by sensors. Multi-view video data refers to the image streams captured synchronously by cameras at multiple angles.

[0071] 102. Based on the correlation between the frequency variation and the preset elastic coefficient of the sphere material, a three-dimensional position coordinate sequence of the sphere's motion trajectory is calculated. Simultaneously, a force direction parameter of the player's limb motion is reversely deduced using the phase variation parameter. The three-dimensional position coordinate sequence and the force direction parameter are aligned with the time stamp of the multi-view video data to obtain motion state fusion data.

[0072] In this step, the sphere's material elastic coefficient correlation refers to the mathematical relationship between material hardness and deformation recovery force. The 3D position coordinate sequence refers to the coordinate data sequence of the sphere's position in 3D space over time. The force direction parameter is a quantitative description of the direction of the force applied by the player's limbs on the sphere. The time stamp refers to the precise time stamp recorded during data acquisition. The motion state fusion data refers to the alignment data of the integrated trajectory, force direction, and video timeline.

[0073] 103. Perform cross-frame matching of the sphere outline based on the motion state fusion data to generate a three-dimensional spatial motion sequence of the sphere's flight trajectory, and perform skeletal dynamic modeling of the player's posture based on the motion state fusion data to generate joint torque parameters of the player's action;

[0074] In this step, cross-frame matching refers to the technology used to track the position of the ball across consecutive video frames. The sphere outline refers to the bounding shape of the sphere in the video. The 3D spatial motion sequence refers to the trajectory data (including displacement and velocity) of the sphere in the 3D coordinate system. Skeletal dynamic modeling involves reconstructing a virtual model of the player's skeletal joint motion from this data. The joint torque parameter refers to the product of the force required for joint rotation and the moment arm.

[0075] 104. Perform trajectory consistency verification on the three-dimensional position coordinate sequence of the spherical motion trajectory and the three-dimensional space motion sequence, and generate composite event parameters by combining the joint torque parameters and the force direction parameters;

[0076] In this step, trajectory consistency verification refers to the error checking process between trajectory data generated by different methods. Composite event parameters refer to the comprehensive event identifier that integrates trajectory, force direction, and rule judgment.

[0077] 105. Perform rule matching based on the composite event parameters and a preset ball game rule database to update the game score result, and dynamically superimpose the updated score result on the target screen of the multi-view video data and transmit it to the terminal display device.

[0078] In this step, the sports rules database refers to a standard dataset storing match rules. The target screen refers to the designated area of ​​the multi-view video where score information needs to be overlaid. The terminal display device refers to the user device that receives and displays the live video.

[0079] In summary, steps 101 to 105 achieve a breakthrough in multimodal data fusion and intelligent decision-making for scoring in sports events. By integrating ball sensor data with multi-view visual information, this method establishes a precise correlation model between mechanical parameters and motion trajectories, effectively overcoming the efficiency bottleneck of traditional reliance on video playback or manual judgment. The collaborative analysis of three-dimensional spatial motion sequences and skeletal dynamic modeling restores the physical interaction relationship between player movements and ball trajectories, providing a data-driven objective basis for scoring event determination. Through the multi-dimensional verification mechanism of composite event parameters, the accuracy and real-time performance of goal and foul determinations are significantly improved, forming a closed loop of automated processing from data collection to score updates.

[0080] In order to monitor the frequency changes and material elastic properties of the football when it is impacted, dynamically analyze the deformation recovery force and acceleration parameters under the action of impact force, and build a three-dimensional trajectory prediction model based on the mass and motion state of the football, and at the same time reverse model the force direction of the player's limb movements based on the contact time and protective gear sensor data, so as to accurately quantify the mechanical correlation between the football's motion trajectory and the player's limb movements in the impact event, provide multi-dimensional data support for athlete technical analysis, movement standardization evaluation and training strategy optimization, and realize a two-way dynamic mapping closed loop from mechanical response to movement intention.

[0081] In some embodiments, the step 102 of calculating the three-dimensional position coordinate sequence of the sphere's motion trajectory based on the correlation between the frequency variation and a preset elastic coefficient of the sphere material includes:

[0082] 201. Obtaining a frequency change of the sphere when it is impacted from a sensor inside the sphere, wherein the frequency change is a difference between the surface vibration frequency of the sphere and the initial static state frequency;

[0083] In step 201, the frequency variation refers to the difference between the surface vibration frequency of the ball when it is impacted and the initial resting state frequency. The initial resting state frequency refers to the reference vibration frequency when the ball is not touched.

[0084] In this embodiment, a vibration sensor is first installed inside the ball. When the ball is impacted (e.g., kicked), the sensor detects the vibration frequency of the ball's surface and compares this frequency with the initial frequency when the ball is stationary and untouched. This frequency change (i.e., the difference) is then recorded to synchronize subsequent steps. For example, if the ball's frequency is 100Hz when it's stationary and the sensor detects an increase to 150Hz after being kicked, the frequency change is 150 minus 100, which equals 50Hz. This value is immediately associated with the sensor's timestamp for later processing.

[0085] 202. Convert the frequency variation into a deformation recovery force of the sphere when it is impacted, based on a preset correlation relationship between the elastic coefficients of the sphere material. The deformation recovery force is a force generated when the sphere material recovers its original shape after deformation.

[0086] In step 202, the correlation relationship between the elastic coefficient of the sphere material refers to the mathematical correspondence between the sphere material characteristics and the deformation recovery force. The deformation recovery force refers to the force generated by the elasticity of the material to restore the sphere to its original shape after being impacted.

[0087] In the embodiment of the present application, based on the preset material properties of the sphere (such as the elastic coefficient corresponding to rubber, which is 0.8 here), the frequency change is converted into the force generated when the sphere returns to its original shape after deformation (called deformation recovery force). Specifically, the material parameter table and the conversion formula are used (for example, assuming that every 1 Hz frequency change corresponds to a 2 Newton recovery force). For example, when the frequency change is 50 Hz, the recovery force calculation formula is 50 Hz multiplied by 2 N / Hz, resulting in a recovery force of 100 Newtons. This force value is also bound to a timestamp for the next step.

[0088] 203. Calculate the initial acceleration of the sphere after the impact based on the deformation recovery force and the mass of the sphere;

[0089] In step 203, mass refers to the physical mass property of the sphere. Initial acceleration refers to the instantaneous acceleration value of the sphere when it is impacted.

[0090] In the embodiment of the present application, the initial acceleration after the impact is calculated by combining the mass of the sphere (read directly from the parameter library) using Newton's second law (force equals mass multiplied by acceleration). For example, if the mass of the sphere is 0.45 kg and the restoring force is 100 Newtons, then the acceleration is equal to the restoring force divided by the mass, 100 N ÷ 0.45 kg ≈ 222 m / s². This value represents the instantaneous acceleration change of the sphere at the moment of impact.

[0091] 204. Calculate the instantaneous velocity of the sphere after the impact based on the initial acceleration and the motion state of the sphere before the impact;

[0092] In step 204, the motion state refers to the motion characteristics of the ball before the impact, such as speed and direction. The instantaneous speed refers to the speed value of the ball at a certain moment after the impact.

[0093] In an embodiment of the present application, based on the motion state of the ball before the impact (such as the direction and magnitude of the velocity) and the initial acceleration just calculated, the velocity formula (velocity equals initial velocity plus acceleration multiplied by time) is used to calculate the instantaneous velocity after the impact. For example, before the impact, the football was moving eastward at 5m / s, the impact lasted for 0.01 seconds, and the acceleration was 222m / s². Then the instantaneous velocity is calculated as 5m / s (initial velocity) plus 222m / s² multiplied by 0.01 seconds, that is, 5+2.22≈7.22m / s, and the direction remains eastward. The new velocity value is associated with the timestamp to update the motion state.

[0094] 205. Based on the instantaneous velocity and in combination with the time increment, iteratively calculate the position coordinates of the sphere in the three-dimensional space to form a three-dimensional position coordinate sequence of the sphere's motion trajectory.

[0095] In step 205, the time increment refers to the time interval used in the iterative calculation. The three-dimensional position coordinate sequence refers to a position coordinate data set of the sphere in the three-dimensional space that changes with time.

[0096] In an embodiment of the present application, this instantaneous speed and a preset time interval (such as 0.01 seconds) are used for iterative calculations, and the position coordinates of the sphere in three-dimensional space are updated once every period of time (the new position is equal to the current position multiplied by the speed multiplied by the time interval) to generate a continuous motion trajectory sequence. For example, the initial position is set to (0, 0, 0), the first iteration speed is 7.22 m / s eastward (in the three-dimensional coordinate system, the X-axis velocity component is 7.22 m / s, and the Y and Z axes are 0), the time increment is 0.01 seconds, and the new coordinates are calculated as (0+7.22×0.01, 0, 0) = (0.0722, 0, 0), and then the process is repeated until the movement ends or a new impact occurs, eventually forming a sequence of hundreds of coordinate points per second for displaying or analyzing the trajectory.

[0097] The reverse derivation of the force direction parameter of the player's limb movement by the phase change parameter includes:

[0098] 206. Determine the contact time point and contact duration of the player's limb and the ball based on the phase change parameter;

[0099] In step 206, the phase change parameter refers to the phase difference of the waveform that reflects the contact characteristics of the limb. The contact time point refers to the precise moment when the limb and the ball begin to contact. The contact duration refers to the length of time the limb and the ball remain in contact.

[0100] In the embodiment of the present application, first, the player's protective gear is equipped with sensors. When the player's limbs (such as feet) touch the ball, these sensors will capture the change in pressure, which is manifested as a specific waveform change (that is, a phase change parameter). Using this waveform change, it is possible to accurately locate when the player's limbs touched the ball (such as 0.5 seconds in the video) and how long the kick lasted (for example, the kick started at 0.5 seconds and was not released until 0.7 seconds, lasting 0.2 seconds). At the same time, the time point is synchronized with the multi-angle video images of the scene to ensure that the timeline is completely consistent, which provides a critical time reference for the next step of analysis.

[0101] 207. Based on the contact time point and contact duration, combined with the spatial distribution of protective gear sensors, calculate the movement direction of the player's limb when contacting the ball;

[0102] In step 207, the spatial distribution of the protective gear sensors refers to the three-dimensional position layout of the sensors on the player's protective gear. The movement direction refers to the moving direction vector of the limb when it contacts the ball.

[0103] In the embodiment of the present application, after knowing the specific time point (0.5 seconds) and duration (0.2 seconds) of the kick, combined with the position distribution data of the protective gear sensors (for example, the sensors on the protective gear are distributed at multiple positions such as the ankles and heels to form a three-dimensional coordinate network), the actual position of the player's limbs can be found at the moment of kicking the ball (for example, the coordinates of the left foot in three-dimensional space are 1.2 meters on the X axis and 0.8 meters on the Y axis). Then, the change in the position of the protective gear before and after the kick is observed (for example, the position of the left foot is (1.2, 0.8) at 0.5 seconds, and becomes (1.5, 0.8) at 0.7 seconds). In this way, the direction of movement of the legs and feet during the kick can be calculated (for example, the foot moves horizontally from 1.2 meters to 1.5 meters horizontally, with no change on the Y axis, indicating pure horizontal movement, and the direction of movement is approximately 30 degrees east of the horizontal, which represents the general trend of limb movement).

[0104] 208. Based on the movement direction and the force magnitude of the protective gear sensor, reversely deduce the direction of the force applied by the player's limbs to the ball.

[0105] In step 208, the force magnitude refers to the force value measured by the sensor. Reverse deduction refers to the calculation method of inferring the direction of the force from the force result. The force direction refers to the direction vector of the force applied by the limb on the sphere.

[0106] In the embodiment of the present application, the direction of force is reversely deduced, that is, the force used by the player to kick the ball and the specific angle at which the force is applied to the ball are reversed. The method is to combine the limb movement direction just obtained (for example, a horizontal angle of 30 degrees) and the actual force measured by the protective gear sensor (for example, at the moment of kicking the ball, the maximum pressure measured by the left foot sensor is 200 Newtons), and decompose this force into different axes of the three-dimensional coordinate system according to the direction of movement (for example, the X-axis component is 200 Newtons multiplied by the cosine value of the 30-degree angle cos30°≈0.866, which is calculated as 200×0.866≈173 Newtons; the Y-axis component is 200 Newtons multiplied by the sine value of the 30-degree angle sin30°=0.5, which is calculated as 200×0.5=100 Newtons). In this way, the specific direction of the force can be determined (for example, the force direction is also about 30 degrees horizontally), and then this result is compared with the direction in which the ball was kicked to correct the predicted ball flight trajectory or analyze whether the player's action is legal.

[0107] In summary, steps 201 to 208 achieve high-precision analysis of kinematic parameters and trajectory reconstruction. By establishing a physical model conversion mechanism between frequency variation and material elastic coefficient, this method accurately restores the three-dimensional motion trajectory of the ball after impact, overcoming the error accumulation problem of traditional visual tracking under occlusion or high-speed motion. The reverse derivation algorithm for force direction parameters, combined with the spatial distribution characteristics of protective gear sensors, accurately captures the biomechanical characteristics of the player's movements, providing a reliable physical action parameter foundation for subsequent event determination, significantly improving the engineering applicability and data credibility of motion state analysis.

[0108] In some embodiments, in step 102, aligning the three-dimensional position coordinate sequence, the force direction parameter, and the time stamp of the multi-view video data to obtain motion state fusion data includes:

[0109] 301. Extract a time stamp of each frame in the multi-view video data, where the time stamp is an accurate time stamp recorded by a video acquisition device;

[0110] In step 301, the time stamp refers to the precise time mark when the video frame is captured. The precise time stamp refers to a time record value with millisecond-level accuracy.

[0111] In the embodiments of the present application, when establishing motion state fusion data, it is first necessary to unify the time base of the multi-source data. This process begins by extracting the time stamps of the multi-view video: the video acquisition device generates a precise timestamp (accurate to the millisecond level) for each frame. By parsing the metadata fields of the video stream (such as the time tags in the MPEG-TS stream), these timestamps are converted to a unified format and stored as an ordered sequence. For example, when a football match uses five cameras to capture a shot, the 120th frame from the left perspective may correspond to the absolute time 10:00:05.230. This timestamp becomes the reference point for data alignment.

[0112] 302. Match the three-dimensional position coordinate sequence and instantaneous velocity sequence of the spherical motion trajectory with the time stamp of the multi-view video data according to the corresponding timestamps;

[0113] In step 302, the instantaneous velocity sequence refers to a data set of the instantaneous velocity of the sphere changing with time. The time slice refers to a specific time slice in the data sequence.

[0114] In this embodiment of the present application, trajectory data matching is then performed: based on the trajectory data generated by the sensor inside the sphere (100 sets of three-dimensional coordinates and instantaneous speed values ​​per second), its timestamp is aligned with the video frame time stamp - if the trajectory time point 10:00:05.230 has actual coordinates (such as X=12.3m, Y=4.5m, Z=1.8m in the goal coordinate system) and speed value (such as 18.2m / s), it is directly bound to the 120th frame of the video; if the trajectory data has a coordinate point at 10:00:05.228 and 10:00:05.232 (assuming it is point A and point B), the intermediate coordinates at the time of 230 milliseconds are calculated by linear interpolation. For example, if the coordinates of point A are (12.0, 4.0, 1.7) and point B are (12.5, 4.7, 1.9), and the time interval is 4 milliseconds, then the coordinates at the time of 230 milliseconds are calculated by weight:

[0115] X=12.0+(12.5-12.0)×(230-228) / 4≈12.25m;

[0116] Y=4.0+(4.7-4.0)×(230-228) / 4≈4.35m;

[0117] Z=1.7+(1.9-1.7)×(230-228) / 4≈1.8m;

[0118] At this time, the interpolation result is used as the binding trajectory data of the 120th frame.

[0119] 303. Match the force direction of the player's body movement with the time stamp of the multi-view video data according to the corresponding time stamp;

[0120] In step 303, the corresponding timestamp refers to the corresponding relationship after the time axis is compressed proportionally.

[0121] In this embodiment, force direction matching is performed simultaneously: the force parameters recorded by the player's protective gear sensor (e.g., a 35° angle and a force of 220N when the foot contacts the ball) are aligned with the video frames using a nearest neighbor matching algorithm. For example, if the sensor records force data at 10:00:05.228, and the nearest video frame is frame 120 (10:00:05.230), the time difference is only 2 milliseconds and does not exceed the threshold. In this case, the force direction parameter is directly associated with frame 120.

[0122] 304. Combine the matched three-dimensional position coordinates, instantaneous velocity, and force direction with corresponding frames of the multi-view video data to form motion state fusion data.

[0123] In step 304, the motion state fusion data refers to a composite data set that integrates trajectory, velocity, force direction, and video frames. The corresponding frames refer to video frames with matching time stamps.

[0124] In this embodiment, the final step is data packaging: the bound data for frame 120 (interpolated coordinates (12.25, 4.35, 1.8), velocity 18.2 m / s, and force angle 35°) is packaged with the multi-angle video footage of that frame (e.g., the player's posture at the moment of the shot) into a fused data unit using a structured protocol (e.g., JSON). This entire process is iterated frame by frame, ultimately forming a fused data set containing spatiotemporally synchronized kinematic parameters and multi-view images. For example, a shooting motion analysis system can use this data to simultaneously retrieve the image of any frame, the ball's flight trajectory, and the player's force parameters.

[0125] In summary, steps 301 to 304 achieve millisecond-level temporal synchronization and spatial alignment of multi-source heterogeneous data. Through precise timestamp matching and frame-level correspondence of 3D coordinate sequences, this method resolves the temporal axis misalignment between sensor data and video streams, ensuring spatiotemporal consistency between mechanical parameters and visual imagery. The structured integration of motion state fusion data establishes a unified benchmark for collaborative analysis of cross-modal data, effectively eliminating the information silos inherent in traditional multi-source data fusion and providing high-precision temporal dimension assurance for subsequent trajectory reconstruction and motion modeling.

[0126] In some embodiments, in step 103, performing cross-frame matching of the sphere contour based on the motion state fusion data to generate a three-dimensional spatial motion sequence of the sphere's flight trajectory, and performing skeletal dynamic modeling of the player's posture based on the motion state fusion data to generate joint torque parameters of the player's action include:

[0127] 401. Extract the three-dimensional position coordinates of the sphere's flight trajectory from the motion state fusion data, and calculate a cross-frame motion trajectory sequence of the sphere based on the sphere's contour shape differences between adjacent frames in the time stamps of the multi-view video data.

[0128] In step 401, the inter-frame motion trajectory sequence refers to continuous motion trajectory data calculated by comparing the sphere's outline differences between adjacent video frames. Three-dimensional position coordinates refer to the sphere's coordinate values ​​in three-dimensional space (e.g., x, y, and z axis values). Sphere outline shape differences refer to the amount of change in the sphere's boundary shape between adjacent frames.

[0129] In an embodiment of the present application, first, the three-dimensional position coordinates of the sphere corresponding to each video frame are extracted from the motion state fusion data to ensure that the coordinate data is strictly aligned with the time stamp. Secondly, the contour shape difference analysis of the sphere in adjacent frames in the multi-view video data (such as image edge detection and feature point matching technology) is used to identify the contour change characteristics of the sphere under continuous time stamps. Then, the displacement direction and distance of the sphere in three-dimensional space are calculated based on the contour change characteristics (such as a triangulation algorithm based on the displacement of feature points), and the displacement trajectory of the sphere between adjacent frames is generated. Finally, the displacement trajectories of adjacent frames are superimposed in chronological order to form a complete cross-frame motion trajectory sequence, which provides basic data for the continuous analysis of the sphere's flight path.

[0130] 402. Generate a three-dimensional spatial motion sequence of the sphere's flight trajectory based on the temporal continuity of the sphere's three-dimensional position coordinates in the cross-frame motion trajectory sequence, wherein the three-dimensional spatial motion sequence includes the sphere's displacement direction, instantaneous velocity, and acceleration parameters;

[0131] In step 402, the 3D motion sequence refers to a set of sphere motion trajectory data, including displacement, velocity, and acceleration. Displacement direction refers to the sphere's motion direction vector in 3D space. Instantaneous velocity refers to the sphere's velocity at a given moment. Acceleration parameter refers to the rate of change of the sphere's velocity over time.

[0132] In the embodiment of the present application, first, based on the three-dimensional position coordinates of the cross-frame motion trajectory sequence, the temporal continuity of the coordinate points is verified through time series analysis (such as checking whether the time intervals of adjacent coordinates are consistent). Secondly, the displacement difference of the sphere between adjacent time points is calculated based on the continuous three-dimensional position coordinates, and the instantaneous velocity is derived based on the time difference (such as the displacement difference divided by the time difference). Then, the acceleration parameter is calculated based on the change in instantaneous velocity (such as the ratio of the velocity difference to the time difference). Finally, the displacement direction, instantaneous velocity and acceleration of all time points are integrated to generate a three-dimensional spatial motion sequence including complete kinematic parameters, accurately describing the dynamic characteristics of the sphere's flight trajectory.

[0133] 403. Extract force direction parameters of the player's movements from the motion state fusion data, combine them with the spatial displacements of the player's joint nodes at adjacent time markers in the multi-view video data, and construct a skeletal dynamic model of the player's posture. Determine angle change sequences and bone length change sequences of the joint nodes based on the skeletal dynamic model.

[0134] In step 403, the skeleton dynamic model refers to the virtual skeleton motion model reconstructed using joint node displacement data. The joint node angle change sequence refers to the sequence data of the joint rotation angle changing over time. The bone length change sequence refers to the sequence data of the distance between the two ends of the skeleton changing over time.

[0135] In an embodiment of the present application, first, the force direction parameters of the player's action (such as the direction vector when touching the ball) are extracted from the motion state fusion data. Secondly, the spatial displacement of the player's joint nodes at adjacent time marks in the multi-view video data (such as the knee joint moving from position A to position B) is used to construct a skeletal dynamic model of the player's posture using skeletal key point detection technology. Then, based on the dynamic model, the angle changes of each joint node in the time series (such as the increase or decrease trend of the hip joint bending angle over time) and the bone length changes (such as the change in the distance of arm extension or contraction) are analyzed. Finally, a joint angle change sequence and a bone length change sequence are generated to provide biomechanical input for joint torque analysis.

[0136] 404. Calculate the joint torque parameters of the player's action based on the angle change sequence of the joint node and the bone length change sequence, combined with the action vector of the force direction parameter. The joint torque parameters include the joint rotation direction and the torque intensity ratio.

[0137] In step 404, the joint rotation direction refers to the axial direction of the joint rotation (such as rotation around the x-axis). The torque strength ratio refers to the relative proportional relationship of the joint torque magnitude.

[0138] In an embodiment of the present application, first, the motion vector of the joint node (such as the direction and distance of movement) is calculated based on the angle change sequence of the joint node (such as the knee joint angle increases from 30° to 45° over time) and the bone length change sequence (such as the thigh bone stretches 0.2 meters). Secondly, combined with the action vector of the force direction parameter (such as the direction and strength of the foot touching the ball), the driving relationship of the force direction on the joint movement (such as the angle between the force direction and the joint movement direction) is analyzed through the vector projection algorithm. Then, based on the angle and the change in bone length, the biomechanical model is used to calculate the torque intensity ratio and rotation direction (such as the clockwise rotation torque of the knee joint accounts for 80%) that the joint bears during the movement. Finally, the joint torque parameters including the joint rotation direction and the torque intensity ratio are output to provide a quantitative basis for movement compliance assessment and injury risk analysis.

[0139] In summary, steps 401 to 404 implement cross-modal dynamic modeling of the ball's trajectory and the player's posture. By combining a cross-frame matching algorithm for the ball's contour and analysis of skeletal joint displacement trajectories, this method innovatively constructs a correlation model between three-dimensional spatial motion sequences and joint torque parameters. Combining physical and mechanical parameters with video analysis techniques, this method not only accurately reproduces the dynamic characteristics of the ball's flight trajectory but also quantifies the biomechanical parameters of the player's movements. This establishes a dual verification mechanism for spatial motion and mechanical interaction for complex event determination, significantly enhancing the dimensionality and depth of motion event analysis.

[0140] In some embodiments, step 403 includes extracting force direction parameters of the player's movements from the motion state fusion data, combining the spatial displacements of the player's joint nodes at adjacent time markers in the multi-view video data, constructing a skeletal dynamic model of the player's posture, and determining a joint node angle change sequence and a bone length change sequence based on the skeletal dynamic model, including:

[0141] 501. Extracting a force direction parameter of the player's action from the motion state fusion data, and obtaining a horizontal action angle and a vertical action angle of the force direction parameter in three-dimensional space to generate an action vector of the force direction parameter;

[0142] In step 501, the action vector refers to the vector representation of the force direction in three-dimensional space. The horizontal action angle refers to the projection angle of the force direction on the horizontal plane. The vertical action angle refers to the projection angle of the force direction on the vertical plane.

[0143] In an embodiment of the present application, first, the force direction parameters of the player's action are extracted from the motion state fusion data. The parameters include the horizontal action angle in three-dimensional space (such as the angle relative to the east direction of the court) and the vertical action angle (such as the elevation angle relative to the horizontal plane). Secondly, the horizontal angle and the vertical angle are converted into the action direction vector in three-dimensional space (such as the proportional relationship between the X-axis component, the Y-axis component, and the Z-axis component) through a three-dimensional vector synthesis algorithm. Then, the unit length of the direction vector is ensured according to the vector normalization process. Finally, a standardized force direction parameter action vector is generated to provide mechanical input for subsequent skeletal dynamic modeling.

[0144] 502. Based on the spatial displacements of the joint nodes of the same player at consecutive time marks in the multi-view video data, a joint point displacement trajectory reflecting the movement trend is formed by correlating the vector changes of the spatial displacements of the joint nodes at adjacent time marks;

[0145] In step 502, the joint point displacement trajectory refers to the motion path of the joint node in three-dimensional space that changes with time. The vector change refers to the dynamic adjustment of the joint displacement direction and size.

[0146] In an embodiment of the present application, first, the spatial displacement of the joint nodes (such as the knee joint and elbow joint) of the same player at consecutive time marks is extracted from the multi-view video data (such as the displacement coordinate difference between frame 1 and frame 2). Secondly, the displacement direction and distance of the joint nodes at adjacent time marks are calculated using a vector analysis algorithm (such as the movement vector of the knee joint from coordinate A to B is X+0.5m, Y+0m). Then, the displacement vectors are connected in chronological order to form a joint point displacement trajectory that reflects the movement trend of the joint node (such as a trajectory chain consisting of 20 displacement vectors per second). Finally, the trajectory data is stored for dynamic parameter derivation of the skeletal model.

[0147] 503. Based on the fluctuation characteristics of the action vector of the force direction parameter on the time axis and the continuity of the spatial displacement trajectory of the joint point, a skeletal dynamic model of the player's posture is constructed, wherein the skeletal dynamic model includes the initial angles of the joint nodes and the initial lengths of the bones;

[0148] In step 503, the fluctuation characteristic refers to the change pattern of the force direction parameter in the time dimension. Continuity refers to the uninterrupted coherence of the joint displacement trajectory. The initial angle refers to the starting rotation angle of the joint node in the skeletal dynamic model. The initial length of the bone refers to the distance between the initial ends of the bone in the skeletal dynamic model.

[0149] In an embodiment of the present application, first, based on the fluctuation characteristics of the action vector of the force direction parameter on the time axis in step 501 (such as the direction angle gradually increasing from 30° to 45°), the driving relationship of the force direction on the joint movement is analyzed. Secondly, combined with the continuity of the joint point displacement trajectory in step 502 (such as the knee joint displacement trajectory has no sudden jump), a skeletal dynamic model of the player's posture is constructed through the inverse kinematics algorithm. Then, based on the joint node position of the initial frame (such as the initial angle of the knee joint is 30°) and the bone length (such as the initial length of the thigh bone is 0.8m), the initial state of the model is set. Finally, a dynamic model including the initial joint angle and the initial bone length is generated to provide a benchmark for the analysis of angle and length changes.

[0150] 504. Generate an angle change sequence of the joint nodes based on the initial angles of the joint nodes in the skeletal dynamic model and the fluctuation amplitude of the displacement direction in the spatial displacement trajectory of the joint points;

[0151] In step 504, the angle variation sequence refers to a data sequence of the fluctuation of the rotation angle of the joint node over time. The fluctuation amplitude refers to the amplitude range of the angle variation.

[0152] In an embodiment of the present application, first, the initial angle of the joint node is extracted from the skeletal dynamic model of step 503 (such as the initial angle of the hip joint is 15°). Secondly, based on the fluctuation amplitude of the displacement direction in the joint point displacement trajectory of step 502 (such as the knee joint displacement direction fluctuates within the range of ±10°), the angle change sequence of the joint node is generated by angle difference calculation (such as the hip joint angle increases by 2° per second). Then, the angle change sequence is aligned with the time stamp to ensure synchronization with the action of the video screen. Finally, the angle change parameter table of the joint node is output for action compliance and biomechanical analysis.

[0153] 505. Generate a bone length change sequence based on the initial length of the bone in the skeletal dynamic model and the cumulative difference of the displacement distance in the spatial displacement trajectory of the joint point.

[0154] In step 505, the bone length change sequence refers to the data sequence of bone length adjustment over time. The cumulative difference refers to the cumulative change of joint displacement distance over time.

[0155] In an embodiment of the present application, first, the initial length of the bone is extracted from the skeletal dynamic model of step 503 (e.g., the initial length of the calf bone is 0.5 m). Secondly, based on the cumulative difference in displacement distance in the joint point displacement trajectory of step 502 (e.g., the ankle joint cumulatively displaces 3 m in 10 seconds), a bone length change sequence is generated by length difference calculation (e.g., the calf bone stretches 0.02 m per second). Next, the length change sequence is aligned with the time marker to ensure the temporal continuity of the data. Finally, a bone length change parameter table is output to provide a quantitative basis for joint torque calculation and sports injury risk assessment.

[0156] In summary, steps 501 to 505 achieve a refined dynamic analysis of a player's biomechanical characteristics. Through vector correlation analysis between force vectors and joint displacements, this method establishes a quantitative index system for angular and length changes in the skeletal dynamic model. Combined with analysis of fluctuation characteristics along the time axis, this method accurately captures the continuous evolution of motion posture, transcending the limitations of traditional video analysis, which focuses on single-dimensional analysis of surface movements. The generation mechanism of joint torque parameters effectively reveals the inherent mechanical laws of movement execution, providing a quantifiable biomechanical basis for foul determination and technical performance evaluation.

[0157] In some embodiments, in step 104, the three-dimensional position coordinate sequence of the ball motion trajectory and the three-dimensional space motion sequence are subjected to trajectory consistency verification, and the joint torque parameter and the force direction parameter are combined to generate a composite event parameter, including:

[0158] 601. Obtain a displacement direction parameter by analyzing position changes of consecutive time stamps in the three-dimensional position coordinate sequence, and perform trajectory consistency comparison between the displacement direction parameter and the displacement direction parameter in the three-dimensional space motion sequence to output a trajectory matching degree.

[0159] In step 601, the displacement direction parameter refers to the instantaneous motion direction calculated from the position changes of consecutive timestamps. Trajectory consistency comparison refers to the process of comparing the similarity of displacement direction data generated by different methods. Trajectory matching is a quantitative indicator of the degree of consistency between two displacement direction data.

[0160] In an embodiment of the present application, first, the position difference of consecutive timestamps is extracted from the three-dimensional position coordinate sequence, and the displacement direction of the sphere at adjacent time points (such as the motion vector from coordinate A to coordinate B) is calculated to generate a displacement direction parameter. Secondly, the displacement direction parameter is compared with the pre-stored displacement direction parameters in the three-dimensional space motion sequence for vector similarity (such as the cosine similarity algorithm). Then, the degree of consistency between the two directions is calculated (such as a similarity value of 0.9), and the trajectory matching degree is output (such as a matching degree of 90%). Finally, the trajectory matching degree is compared with a preset threshold to determine whether to trigger the subsequent exception handling process.

[0161] 602. When the trajectory matching degree is lower than a preset trajectory tolerance threshold, mark an abnormal trajectory segment in the three-dimensional space motion sequence that does not match the spherical motion trajectory, and update the displacement direction parameter of the three-dimensional space motion sequence according to the abnormal trajectory segment;

[0162] In step 602, the preset trajectory tolerance threshold refers to the maximum critical value of the allowed trajectory difference. The abnormal trajectory segment refers to the displacement data segment that does not match the actual motion trajectory of the ball. Updating the displacement direction parameter refers to the operation of correcting and adjusting the abnormal segment data.

[0163] In this embodiment of the present application, first, when the trajectory matching degree in step 601 falls below a preset threshold (e.g., matching degree <85%), the corresponding abnormal trajectory segment in the three-dimensional motion sequence (e.g., the trajectory with timestamps 10:00:05 to 10:00:06) is marked. Secondly, an interpolation algorithm (e.g., cubic spline interpolation) is used to correct the displacement direction parameters of the abnormal trajectory segment based on the displacement directions of the preceding and following normal trajectory segments. Next, the corrected displacement direction parameters are updated to the three-dimensional motion sequence, overwriting the original abnormal data. Finally, the trajectory matching degree is recalculated until the threshold requirement is met or the maximum number of iterations is completed.

[0164] 603. Extract the joint rotation direction parameter and the torque intensity ratio parameter from the joint torque parameter, and calculate the spatial angle between the force direction parameter and the joint rotation direction parameter and the ratio difference between the torque intensity and the action vector in combination with the action vector of the force direction parameter;

[0165] In step 603, the spatial angle refers to the angular difference between the force direction and the joint rotation direction. The proportional difference between the torque strength and the action vector refers to the degree of difference between the strength parameters of the two.

[0166] In an embodiment of the present application, first, the joint rotation direction parameter (such as clockwise rotation of the knee joint) and the torque intensity ratio parameter (such as the intensity ratio of 70%) are extracted from the joint torque parameter. Secondly, the action vector (such as a direction angle of 30° and a force of 200N) is obtained from the force direction parameter, and the spatial angle between it and the joint rotation direction vector is calculated (such as an angle of 15°). Then, the relative difference between the torque intensity and the force direction force is calculated by the proportional difference algorithm (such as the matching deviation between 70% intensity and 200N). Finally, the spatial angle and the proportional difference are stored in a temporary parameter table to provide input for validity judgment.

[0167] 604. When the spatial angle is less than the preset angle tolerance and the proportional difference is less than the preset strength tolerance, the joint torque parameter and the force direction parameter at the current moment are bound to form a valid action parameter group;

[0168] In step 604, the preset angle tolerance refers to the maximum deviation value allowed for the spatial angle. The preset strength tolerance refers to the maximum threshold value allowed for the difference in torque strength ratio. The valid action parameter set refers to the verified matching data combination of joint torque and force direction.

[0169] In the embodiment of the present application, first, the spatial angle of step 603 is compared with the preset angle tolerance (such as 20°). If the angle is less than the threshold, the direction is considered valid. Secondly, the proportional difference is compared with the preset strength tolerance (such as 10%). If the difference is less than the threshold, the strength is considered valid. Then, when both conditions are met, the joint torque parameter and the force direction parameter of the current timestamp are bound to a valid action parameter group (such as marked as "valid_timestamp 10:00:05"). Finally, the valid parameter group is written to the database, and the invalid parameter group triggers an alarm or data correction process.

[0170] 605. Concatenate the updated displacement direction parameter and the valid action parameter group according to the timestamp to generate a composite event parameter.

[0171] In step 605, timestamp splicing refers to the operation of aligning and combining different parameter data in chronological order. Composite event parameters refer to a complete event data set that integrates displacement direction and effective action parameters.

[0172] In this embodiment, first, the updated displacement direction parameter is obtained from step 602, and the valid action parameter set is obtained from step 604, both of which are timestamped. Next, the displacement direction parameter and the valid action parameter set are concatenated based on the timestamp (e.g., a timestamp of 10:00:05 corresponds to a displacement direction due east and valid parameter set A). Next, the concatenated parameters are integrated into a time-continuous composite event parameter (e.g., 100 parameter records per second) using a data encapsulation protocol. Finally, the composite event parameter is pushed to the event analysis system to support event assessment and visualization.

[0173] In summary, steps 601 to 605 implement intelligent verification and optimization of motion event determination parameters. Through trajectory consistency verification and spatial correlation analysis of mechanical parameters, this method establishes an intelligent verification mechanism for multi-dimensional data cross-validation. The dynamic identification and parameter update mechanism for abnormal trajectory segments effectively eliminates the influence of sensor errors or visual interference, while the spatial matching algorithm of joint torque and force vectors strengthens the logical consistency of physical interaction relationships, forming a composite event generation system with self-correction capabilities, significantly improving the fault tolerance and environmental adaptability of the scoring determination system.

[0174] In some embodiments, in step 105, matching the composite event parameters with a preset ball game rule database to update the game score result includes:

[0175] 701. Extracting a valid action parameter group from the composite event parameters, and matching the valid action parameter group with the rule compliance of goal and foul determination conditions in a preset ball game rule database. When the rule compliance reaches a preset threshold, a corresponding trigger event flag is generated.

[0176] In step 701, rule compliance refers to a quantitative indicator of the degree of matching between action parameters and competition rules. Trigger event marker refers to an event identifier generated when the rule conditions are met.

[0177] In an embodiment of the present application, first, an effective action parameter group including parameters such as force direction and joint torque is extracted from the composite event parameters. Secondly, the key parameters in the effective action parameter group (such as the coordinates of the end point of the ball and the angle of the force direction) are compared with the judgment conditions in the ball competition rule database item by item: for the goal rule, it is determined whether the coordinates of the end point of the ball are within the coverage of the goal area; for the foul rule, it is determined whether the angle of the force direction exceeds the allowable range or whether the strength of the joint torque is illegal. Then, the rule matching algorithm (such as logical condition layer by layer screening) is used to calculate the rule compliance of the parameter group with the rule (such as 80% for goal conditions and 90% for foul conditions). Finally, when the rule compliance exceeds the preset threshold (such as 75% for goal threshold and 85% for foul threshold), a corresponding trigger event mark (such as "goal valid" or "secondary foul") is generated, and the mark is associated with the video frame with the corresponding timestamp.

[0178] 702. When the trigger event flag satisfies the goal determination condition, generating a score change instruction based on the overlapping relationship between the end position of the ball motion trajectory and the goal area, so as to update the event score result according to the score change instruction;

[0179] In step 702, the end position refers to the final coordinate point of the ball's motion trajectory. The goal area coverage relationship refers to the inclusion determination of the end position and the goal space range. The score change instruction refers to the control signal that triggers the score change.

[0180] In this embodiment of the present application, first, when the trigger event flag in step 701 meets the goal determination criteria, the coordinates of the end point position are extracted from the ball's motion trajectory (e.g., ball coordinate X = inside the goal line). Next, a three-dimensional spatial coverage model of the goal area (e.g., goal line range definition) is used to determine whether the end point position is within the valid scoring area. Then, if a valid goal is determined (e.g., the coordinates are inside the goal line and there are no foul interferences), a score change instruction is generated (e.g., home team score +1). Finally, the instruction is sent to the event scoring system, updating the event score results in the database.

[0181] 703. When the trigger event mark satisfies the foul determination condition, a foul level mark is generated according to the matching degree between the force direction parameter and the preset illegal action direction, so as to update the event score result according to the foul level mark.

[0182] In step 703, the direction of the illegal action refers to the range of force directions prohibited by the rules. The foul level mark refers to the level mark divided according to the degree of the violation.

[0183] In this embodiment of the present application, first, when the trigger event flag in step 701 meets the foul determination criteria, the action direction is extracted from the force direction parameter (e.g., the foot contact direction overlaps with the prohibited angle range). Next, the force direction is compared with a preset illegal action direction (e.g., a back tackle angle threshold) for a match degree (e.g., a directional deviation of less than 10° is considered a violation). Next, a foul level is assigned based on the degree of deviation (e.g., a flagrant foul level 1, a technical foul level 2). Finally, the match score is updated based on the foul level (e.g., the opposing team is awarded a free throw opportunity), and the result is synchronized with the scoring system.

[0184] 704. Bind the updated match score result to a preset score display area in the target screen of the multi-view video data to generate dynamic overlay data including the score value and the foul level;

[0185] In step 704, the score display area refers to a preset score information display position in the screen. The dynamic overlay data refers to a set of score and foul information that needs to be rendered.

[0186] In an embodiment of the present application, first, the score value and foul level (such as a score of 2:1 and a foul level of two) are extracted from the updated event score results. Secondly, based on the preset score display area (such as the score column in the upper left corner of the screen) and the foul mark area (such as the icon in the lower right corner of the screen) in the multi-view video data, the score and foul information are bound by position (such as the score column showing "2:1" and the red card icon showing in the lower right corner). Then, dynamic overlay data including text and icons is generated through a graphic overlay algorithm (such as a transparent layer overlay in PNG format). Finally, the overlay data is associated with the video frame metadata to ensure that the displayed content is synchronized with the screen action.

[0187] 705. According to the transmission frame rate of the multi-view video data, synchronously render the dynamic overlay data to the target screen according to the frame timestamp, generate overlay video transmission data, and output it to the terminal display device.

[0188] In step 705, the transmission frame rate refers to the number of frames per second of the video data. Synchronous rendering refers to the rendering process of accurately matching the overlay data with the video frames in time. The overlayed video transmission data refers to the final output video stream including the score information.

[0189] In an embodiment of the present application, first, based on the transmission frame rate of the multi-view video data (such as 25 frames per second), the score and foul information of the corresponding time point are extracted from the dynamic overlay data according to the frame timestamp (such as 10:00:05.040). Secondly, the overlay data (such as the score "2:1") is drawn to the target screen (such as the fixed coordinates in the upper left corner) at a preset position through the video rendering engine. Then, the rendered video frame is encoded into a streaming format (such as H.264), and a synchronization timestamp and event marker are inserted. Finally, the overlay video transmission data is output to the terminal display device (such as a live TV stream or a mobile APP) to achieve a visual presentation of the event information.

[0190] In summary, steps 701 to 705 achieve intelligent matching and visual presentation of event rules and data. By establishing a dynamic mapping mechanism between composite event parameters and the rule database, this method achieves automated decision-making for goal and foul decisions. The generation logic for score change instructions and foul level markers effectively balances rule complexity with the need for decision-making, while dynamic overlay rendering technology ensures precise synchronization of decision results with multi-view images. This technical system not only significantly improves the processing efficiency of event decisions, but also provides viewers with instant and transparent penalty information through enhanced visualization technology, revolutionizing the digital adjudication model for traditional sports events.

[0191] Figure 2 The present invention provides a schematic diagram of a system for updating scores of sports events. Figure 2 As shown, the system includes:

[0192] An acquisition module 21 is configured to acquire mechanical data fed back by sensors deployed inside the ball and on the player's protective gear, as well as multi-view video data captured by a multi-view camera. The mechanical data includes frequency variation of the ball when impacted and phase variation parameters of the player's limb contact.

[0193] a calculation module 22 for calculating a three-dimensional position coordinate sequence of the ball's motion trajectory based on a correlation between the frequency variation and a preset elastic coefficient of the ball's material, and reversely deriving a force direction parameter of the player's limb motion using the phase variation parameter, and aligning the three-dimensional position coordinate sequence and the force direction parameter with the time stamp of the multi-view video data to obtain motion state fusion data;

[0194] a generation module 23 configured to perform cross-frame matching of the sphere outline based on the motion state fusion data to generate a three-dimensional spatial motion sequence of the sphere's flight trajectory, and perform skeletal dynamic modeling of the player's posture based on the motion state fusion data to generate joint torque parameters of the player's action;

[0195] A verification module 24 is configured to verify the consistency of the three-dimensional position coordinate sequence of the spherical motion trajectory with the three-dimensional spatial motion sequence, and generate a composite event parameter by combining the joint torque parameter and the force direction parameter;

[0196] The matching module 25 is used to match the rules of the composite event parameters with the preset ball game rule database to update the game score result, and dynamically superimpose the updated score result on the target screen of the multi-view video data to transmit it to the terminal display device.

[0197] Figure 2 The system for updating the scores of sports events can execute Figure 1 The implementation principles and technical effects of the sports event score updating method described in the illustrated embodiment are not further elaborated. The specific manner in which the various modules and units in the sports event score updating system in the aforementioned embodiment perform operations has been described in detail in the relevant embodiments of the method and will not be further elaborated here.

[0198] In one possible design, Figure 2 A sports event score updating system according to the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0199] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0200] The processing component 32 is used for the above Figure 1 The embodiment provides a method for updating scores of sports events.

[0201] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0202] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0203] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0204] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0205] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0206] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0207] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The embodiment shown is a method for updating scores of sports events.

[0208] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0209] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0210] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0211] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for updating scores of sports events, characterized in that: The following steps are involved: Acquire mechanical data from sensors deployed inside the ball and on the players' protective gear, as well as multi-view video data captured by multi-view cameras. The mechanical data includes frequency changes when the ball is impacted and phase change parameters of the players' limbs contacting the ball. Based on the correlation between the frequency variation and a preset elastic coefficient of the sphere material, a three-dimensional position coordinate sequence of the sphere's motion trajectory is calculated. Simultaneously, a force direction parameter of the player's limb movement is reversely deduced using the phase variation parameter. The three-dimensional position coordinate sequence and the force direction parameter are aligned with the time stamp of the multi-view video data to obtain motion state fusion data. Performing cross-frame matching of the sphere outline based on the motion state fusion data to generate a three-dimensional spatial motion sequence of the sphere's flight trajectory, and performing skeletal dynamic modeling of the player's posture based on the motion state fusion data to generate joint torque parameters of the player's action; Performing trajectory consistency verification on the three-dimensional position coordinate sequence of the spherical motion trajectory and the three-dimensional space motion sequence, and generating composite event parameters by combining the joint torque parameters and the force direction parameters; Based on the composite event parameters, rule matching is performed with a preset ball game rule database to update the game score result, and the updated score result is dynamically superimposed on the target screen of the multi-view video data and transmitted to the terminal display device.

2. The method according to claim 1, characterized in that Based on the correlation between the frequency variation and the preset elastic coefficient of the sphere material, a three-dimensional position coordinate sequence of the sphere motion trajectory is calculated, including: Obtaining a frequency change of the sphere when it is impacted from a sensor inside the sphere, wherein the frequency change is the difference between the vibration frequency of the sphere surface and the frequency of the initial static state; According to the preset correlation relationship between the elastic coefficient of the sphere material, the frequency change is converted into the deformation recovery force of the sphere when it is impacted. The deformation recovery force is the force generated when the sphere material returns to its original shape after deformation; Based on the deformation recovery force and the mass of the sphere, the initial acceleration of the sphere after the impact is calculated; Calculate the instantaneous velocity of the sphere after the impact based on the initial acceleration and the motion state of the sphere before the impact; Based on the instantaneous velocity and in combination with the time increment, the position coordinates of the sphere in the three-dimensional space are iteratively calculated to form a three-dimensional position coordinate sequence of the sphere's motion trajectory; The reverse derivation of the force direction parameter of the player's limb movement by the phase change parameter includes: Determining the contact time point and contact duration between the player's limb and the ball based on the phase change parameters; Based on the contact time point and contact duration, combined with the spatial distribution of protective gear sensors, the movement direction of the player's limb when contacting the ball is calculated; According to the movement direction and the magnitude of the force applied by the protective gear sensor, the direction of the force applied by the player's limbs to the ball is reversely deduced.

3. The method according to claim 1, characterized in that The step of aligning the three-dimensional position coordinate sequence, the force direction parameter, and the time stamp of the multi-view video data to obtain motion state fusion data includes: Extracting the time stamp of each frame in the multi-view video data, wherein the time stamp is the precise time stamp recorded by the video acquisition device; Matching the three-dimensional position coordinate sequence and instantaneous velocity sequence of the sphere's motion trajectory with the time stamps of the multi-view video data according to the corresponding timestamps; Match the force direction of the player's body movements with the time stamps of the multi-view video data according to the corresponding timestamps; The matched three-dimensional position coordinates, instantaneous velocity and force direction are combined with the corresponding frames of the multi-view video data to form motion state fusion data.

4. The method according to claim 1, wherein Based on the motion state fusion data, cross-frame matching of the sphere outline is performed to generate a three-dimensional spatial motion sequence of the sphere's flight trajectory; based on the motion state fusion data, skeletal dynamic modeling of the player's posture is performed to generate joint torque parameters of the player's action, including: Extracting the three-dimensional position coordinates of the sphere's flight trajectory from the motion state fusion data, and calculating the sphere's cross-frame motion trajectory sequence based on the sphere's contour shape differences between adjacent frames in the time stamps of the multi-view video data; Based on the temporal continuity of the three-dimensional position coordinates of the sphere in the cross-frame motion trajectory sequence, a three-dimensional spatial motion sequence of the sphere's flight trajectory is generated, wherein the three-dimensional spatial motion sequence includes the displacement direction, instantaneous velocity, and acceleration parameters of the sphere; Extracting force direction parameters of the player's movements from the motion state fusion data, combining them with the spatial displacements of the player's joint nodes at adjacent time markers in the multi-view video data, constructing a skeletal dynamic model of the player's posture, and determining a joint node angle change sequence and a bone length change sequence based on the skeletal dynamic model; According to the angle change sequence of the joint node and the bone length change sequence, combined with the action vector of the force direction parameter, the joint torque parameter of the player's action is calculated. The joint torque parameter includes the joint rotation direction and the torque intensity ratio.

5. The method according to claim 4, characterized in that Extracting force direction parameters of the player's movements from the motion state fusion data, combining the spatial displacements of the player's joint nodes at adjacent time markers in the multi-view video data, constructing a skeletal dynamic model of the player's posture, and determining a joint node angle change sequence and a bone length change sequence based on the skeletal dynamic model, including: Extracting a force direction parameter of the player's action from the motion state fusion data, and generating an action vector of the force direction parameter by obtaining a horizontal action angle and a vertical action angle of the force direction parameter in three-dimensional space; Based on the spatial displacements of the joint nodes of the same player at consecutive time marks in the multi-view video data, a joint point displacement trajectory reflecting the movement trend is formed by correlating the vector changes of the spatial displacements of the joint nodes at adjacent time marks; Based on the fluctuation characteristics of the action vector of the force direction parameter on the time axis and the continuity of the spatial displacement trajectory of the joint point, a skeletal dynamic model of the player's posture is constructed, wherein the skeletal dynamic model includes the initial angles of the joint nodes and the initial lengths of the bones; generating an angle change sequence of the joint nodes based on the initial angles of the joint nodes in the skeletal dynamic model and the fluctuation amplitude of the displacement direction in the spatial displacement trajectory of the joint points; A bone length change sequence is generated based on the initial length of the bone in the bone dynamic model and the cumulative difference of the displacement distance in the spatial displacement trajectory of the joint point.

6. The method according to claim 1, characterized in that The three-dimensional position coordinate sequence of the spherical motion trajectory is verified to be consistent with the three-dimensional space motion sequence, and the joint torque parameter and the force direction parameter are combined to generate a composite event parameter, including: Obtaining a displacement direction parameter by analyzing position changes of consecutive time stamps in the three-dimensional position coordinate sequence, and performing trajectory consistency comparison between the displacement direction parameter and the displacement direction parameter in the three-dimensional space motion sequence to output a trajectory matching degree; When the trajectory matching degree is lower than a preset trajectory tolerance threshold, marking an abnormal trajectory segment in the three-dimensional space motion sequence that does not match the spherical motion trajectory, and updating the displacement direction parameter of the three-dimensional space motion sequence according to the abnormal trajectory segment; Extracting the joint rotation direction parameter and the torque intensity ratio parameter from the joint torque parameter, and combining the action vector of the force direction parameter to calculate the spatial angle between the force direction parameter and the joint rotation direction parameter and the ratio difference between the torque intensity and the action vector; When the spatial angle is less than the preset angle tolerance and the proportional difference is less than the preset strength tolerance, the joint torque parameter and the force direction parameter at the current moment are bound to form a valid action parameter group; The updated displacement direction parameter and the effective action parameter group are spliced ​​according to the timestamp to generate a composite event parameter.

7. The method according to claim 1, characterized in that Matching the composite event parameters with a preset ball game rule database to update the game score result includes: Extracting a valid action parameter group from the composite event parameters, and matching the valid action parameter group with the rule compliance of goal and foul determination conditions in a preset ball game rule database, and generating a corresponding trigger event mark when the rule compliance reaches a preset threshold; When the trigger event mark satisfies the goal determination condition, a score change instruction is generated according to the overlapping relationship between the end position of the ball motion trajectory and the goal area, so as to update the event score result according to the score change instruction; When the trigger event mark satisfies the foul determination condition, a foul level mark is generated according to the matching degree between the force direction parameter and the preset illegal action direction, so as to update the event score result according to the foul level mark; Positionally binding the updated match score result to a preset score display area in the target screen of the multi-view video data to generate dynamic overlay data including the score value and the foul level; According to the transmission frame rate of the multi-view video data, the dynamic superimposed data is synchronously rendered to the target screen according to the frame timestamp, and the superimposed video transmission data is generated and output to the terminal display device.

8. A sports event score update system, characterized in that: The following steps are involved: An acquisition module is used to obtain mechanical data fed back by sensors deployed inside the ball and on the player's protective gear, as well as multi-view video data collected by multi-view cameras. The mechanical data includes frequency changes when the ball is impacted and phase change parameters of the player's limb contact; a calculation module, configured to calculate a three-dimensional position coordinate sequence of the ball's motion trajectory based on a correlation between the frequency variation and a preset elastic coefficient of the ball's material, and simultaneously reversely derive a force direction parameter of the player's limb movement using the phase variation parameter, and align the three-dimensional position coordinate sequence and the force direction parameter with a time stamp of the multi-view video data to obtain motion state fusion data; a generation module, configured to perform cross-frame matching of the sphere outline based on the motion state fusion data to generate a three-dimensional spatial motion sequence of the sphere's flight trajectory, and perform skeletal dynamic modeling of the player's posture based on the motion state fusion data to generate joint torque parameters of the player's action; a verification module, configured to verify the consistency of the three-dimensional position coordinate sequence of the spherical motion trajectory with the three-dimensional spatial motion sequence, and generate a composite event parameter by combining the joint torque parameter and the force direction parameter; The matching module is used to match the rules of the composite event parameters with the preset ball game rule database to update the game score result, and dynamically superimpose the updated score result on the target screen of the multi-view video data to transmit it to the terminal display device.

9. A computing device, characterized in that The method comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for updating a sports event score as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the method for updating the score of a sports event according to any one of claims 1 to 7 is implemented.

Images