A tennis-based shot evaluation method, system, terminal and storage medium

Abstract

The application relates to the technical field of ball hitting scoring, and discloses a tennis-based ball hitting evaluation method, system, terminal and storage medium, the method comprising the following steps: acquiring a complete ball hitting process of a target user, extracting three-dimensional position coordinates, serving information and return ball information; constructing a virtual opponent in a target court, simulating the return hit of the virtual opponent according to the three-dimensional position coordinates, the serving information and the return ball information, obtaining the return ball quality and the ball hitting difficulty of the user, and obtaining the return ball score of the target user after multiple weighted fusions. The application introduces a dynamic confrontation evaluation system based on a virtual opponent model, combines multi-source sports information fusion, realizes tactical-level quantitative evaluation and adaptive training control of the ball hitting quality, can avoid the evaluation deviation of simple high-score or difficult low-score, and improves the evaluation accuracy.

Description

Technical Field

[0001] This invention relates to the field of ball-striking scoring technology, and more particularly to a ball-striking evaluation method, system, terminal, and computer-readable storage medium based on tennis. Background Technology

[0002] In traditional tennis training, ball machines only offer fixed or random serve patterns, lacking quantitative feedback on training quality and failing to assess the user's hitting ability.

[0003] While some existing intelligent training systems can record the trajectory or landing point of the ball, they are usually based on only one dimension. This makes it impossible to integrate the two-way motion information of the incoming ball (the ball machine) and the returning ball (the user), which significantly reduces the accuracy of the ball scoring results.

[0004] Therefore, existing technologies still need to be improved and developed. Summary of the Invention

[0005] The main objective of this invention is to provide a tennis-based shot evaluation method, system, terminal, and computer-readable storage medium, aiming to solve the problem that existing technologies evaluate user returns from a single perspective, failing to integrate multiple aspects of information, thus reducing the accuracy of the evaluation.

[0006] To achieve the above objectives, the present invention provides a tennis-based shot evaluation method, which includes the following steps: The system acquires multiple complete shot sequences of a target user within a target court and extracts the target user's three-dimensional position coordinates, the ball machine's serving information, and the target user's return information from each complete shot sequence. Within the target court, a virtual opponent for the target user is constructed. For each complete shot sequence, the virtual opponent is simulated to return the ball to the target user based on the serve information and the return information, thereby obtaining the return quality in multiple dimensions. Based on the three-dimensional position coordinates and the ball-serving information, the difficulty of hitting the ball by the ball-serving machine in multiple dimensions is evaluated; After multiple weighted fusions of all the stated shot difficulty and all the stated return quality, the return score of the target user at the time of the shot is obtained.

[0007] Optionally, the tennis-based shot evaluation method, wherein acquiring multiple complete shot sequences of the target user within the target court, and extracting the target user's three-dimensional position coordinates, the ball machine's serve information, and the target user's return information from each complete shot sequence, specifically includes: The system acquires multiple complete hitting sequences of a target user within a target court, wherein each complete hitting sequence represents the moment when the target user hits and returns the tennis ball after it is served by the ball machine and the tennis ball lands on the ground. For each complete hitting process, the complete hitting process is analyzed to obtain the ball speed of the ball machine and the return speed of the target user; A court coordinate system is constructed with the center point of the target court as the origin, and the three-dimensional position coordinates of the target user, the ball landing point of the ball machine, and the return ball landing point of the target user are extracted from the court coordinate system.

[0008] Optionally, the tennis-based shot evaluation method further includes, after constructing a court coordinate system with the center point of the target court as the origin, and extracting the three-dimensional position coordinates of the target user, the serve landing point of the ball machine, and the return ball landing point of the target user from the court coordinate system, the method further includes: For each complete hitting process, a hitting event model is constructed based on the three-dimensional position coordinates, the serve landing point, the serve speed, the return ball landing point, and the return ball speed; The three-dimensional position coordinates are represented as follows: ; in, Represents the three-dimensional location of the target user. The x-coordinate represents the target user. The vertical axis represents the target user. The vertical coordinate represents the target user; The coordinates of the serve landing point are represented as follows: ; in, Indicates the landing point of the serve. The x-coordinate represents the point where the serve lands. The vertical coordinate represents the landing point of the serve. The vertical coordinate of the serve's landing point; The serve speed Represented as: ; in, The lateral velocity representing the speed of the serve. The longitudinal velocity representing the speed of the serve. Vertical velocity representing the speed of the serve; The coordinates of the ball's landing point are represented as follows: ; in, Indicates the landing point of the return ball. The x-coordinate represents the point where the ball lands. The vertical coordinate represents the point where the ball lands. The vertical coordinate of the ball's landing point; The ball return speed Represented as: ; in, The lateral velocity representing the speed of the return ball. The longitudinal velocity representing the speed of the return ball. The vertical velocity representing the speed of the ball's return.

[0009] Optionally, the tennis-based shot evaluation method, wherein constructing a virtual opponent for the target user within the target court, and for each complete shot sequence, simulating the virtual opponent's return shot to the target user based on the serve information and the return shot information, to obtain the return shot quality in multiple dimensions, specifically includes: A virtual opponent for the target user is constructed on one side of the ball machine within the target court, wherein the virtual opponent is fixed at the center of the baseline of the target court; The difficulty level is set according to user needs. Based on the difficulty level and the coordinates of the virtual opponent, the catchable area of ​​the virtual opponent is constructed: ; ; in, Indicates the location of the virtual opponent. The coordinates representing the center position of the baseline. Indicates the area where the ball can be received. This represents any possible position the ball can reach in the next moment. This indicates the maximum movement speed of the virtual opponent. This represents the sum of the virtual opponent's reaction time and startup time. A shot quality assessment model is constructed, and the receiveable area, the serve information, and the return information of each complete shot process are input into the shot quality assessment model for comprehensive analysis to obtain the return quality of the ball in multiple dimensions.

[0010] Optionally, in the tennis-based shot evaluation method, the return quality includes: shot breakthrough quality, shot direction quality, shot depth quality, and stability quality. The construction of the shot quality assessment model involves inputting the receiveable area, serve information, and return information for each complete shot sequence into the model for comprehensive analysis, thereby obtaining the return quality in multiple dimensions, specifically including: Construct a shot quality assessment model by inputting the receiveable area, the serve information, and the return information into the shot quality assessment model; Determine whether the return ball landing point in the return ball information is within the reachable area. If not, the shot breakthrough quality is determined to be high. If so, the shot breakthrough quality is calculated based on the position of the return ball landing point within the reachable area. ; in, Indicates the quality of the shot's breakthrough. Indicates the landing point of the return ball. This indicates the quality of the shot when the return ball lands within the reachable area. express The boundary, Represents the distance function. Parameters that control the rate of decay; The return direction vector is calculated based on the return ball landing point in the return ball information and the service ball landing point in the service ball information. The opponent direction vector is calculated based on the position of the virtual opponent and the service ball landing point. The quality of the hitting direction of the return ball is evaluated based on the return ball direction vector and the opponent direction vector. ; ; ; in, This represents the direction vector of the ball's return. Indicates the landing point of the serve. This represents the opponent's direction vector. Indicates the quality of the direction of the shot. Indicates the angle of the return shot; Calculate the depth mass of the return shot based on the vertical coordinate of the ball's landing point and the vertical coordinate of the virtual opponent: ; in, Indicates the depth and quality of the shot. The vertical coordinate represents the landing point of the ball; By calculating the variance of all return ball landing points in each complete shot sequence, the stability quality of the target user during the return shot can be obtained. ; in, Indicates stability quality. This indicates variance calculation.

[0011] Optionally, in the tennis-based shot evaluation method, the shot difficulty includes: spatial distance difficulty, speed difficulty, and time pressure difficulty; The step of evaluating the hitting difficulty of the ball-serving machine in multiple dimensions based on the three-dimensional position coordinates and the ball-serving information specifically includes: Based on the three-dimensional position coordinates and the serve landing point in the serve information, the spatial distance difficulty is constructed as follows: ; in, Indicating spatial distance difficulty Indicates the landing point of the serve. The x-coordinate represents the point where the serve lands. The vertical coordinate represents the landing point of the serve. The vertical coordinate of the serve landing point. Represents the three-dimensional location of the target user. The x-coordinate represents the target user. The vertical axis represents the target user. The vertical coordinate represents the target user; Calculate the speed value of the serve speed in the serve information, and define the speed value as the speed difficulty: ; in, The speed value represents the difficulty or speed of the serve. Indicates the speed of the serve. The lateral velocity representing the speed of the serve. The longitudinal velocity representing the speed of the serve. Vertical velocity representing the speed of the serve; Calculate the target user's maximum reaction time throughout all complete shot sequences, and calculate the time pressure difficulty for each return shot by the target user based on the maximum reaction time: ; in, This indicates the time pressure and difficulty of the current return shot. This indicates the maximum reaction time for the current return ball. This indicates the reaction time for the current return ball.

[0012] Optionally, the tennis-based shot evaluation method, wherein obtaining the target user's shot return score after multiple weighted fusions of all shot difficulties and all return quality specifically includes: The overall shot quality is obtained by weighting and integrating the quality of the shot's breakthrough, direction, depth, and stability. ; in, Indicates the overall quality of the shot. Indicates the quality of the shot's breakthrough. Indicates the quality of the direction of the shot. Indicates the depth and quality of the shot. Indicates stability quality. , , and They represent , , and Quality weights; The overall difficulty of hitting the ball is obtained by weighting and combining the difficulties of spatial distance, speed, and time pressure separately: ; in, Indicates the overall difficulty of the shot. Indicating spatial distance difficulty The speed value represents the difficulty or speed of the serve. This indicates the time pressure and difficulty of the current return shot. , and They represent , and Difficulty weighting; Calculate the return score of the target user's shots in all complete shot sequences based on the overall shot quality and the overall shot difficulty: ; in, Indicates the score for the return shot. This indicates the difficulty reward factor.

[0013] Furthermore, to achieve the above objectives, the present invention also provides a tennis-based shot evaluation system, wherein the tennis-based shot evaluation system includes: The information extraction module is used to acquire multiple complete hitting processes of the target user in the target court, and extract the three-dimensional position coordinates of the target user, the ball machine's serving information, and the target user's return information from each complete hitting process; The quality assessment module is used to construct a virtual opponent for the target user within the target court. For each complete shot, the module simulates the virtual opponent's return shot to the target user based on the serve information and the return shot information, and obtains the return shot quality in multiple dimensions. The difficulty assessment module is used to assess the hitting difficulty of the ball-serving machine in multiple dimensions based on the three-dimensional position coordinates and the ball-serving information. The ball return scoring module is used to obtain the ball return score of the target user when hitting the ball by performing multiple weighted fusions of all the said hitting difficulties and all the said ball return quality.

[0014] Furthermore, to achieve the above objectives, the present invention also provides a terminal, wherein the terminal includes: a memory, a processor, and a tennis-based hit evaluation program stored in the memory and executable on the processor, wherein when the tennis-based hit evaluation program is executed by the processor, it implements the steps of the tennis-based hit evaluation method as described above.

[0015] Furthermore, to achieve the above objectives, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a tennis-based hit evaluation program, which, when executed by a processor, implements the steps of the tennis-based hit evaluation method as described above.

[0016] In this invention, multiple complete hitting sequences of a target user within a target court are acquired, and the target user's three-dimensional position coordinates, the ball machine's serving information, and the target user's return information are extracted from each complete hitting sequence. A virtual opponent for the target user is constructed within the target court. For each complete hitting sequence, the virtual opponent simulates returning the target user's shot based on the serving and return information, obtaining the return quality in multiple dimensions. The hitting difficulty of the ball machine in multiple dimensions is evaluated based on the three-dimensional position coordinates and the serving information. After multiple weighted fusions of all hitting difficulties and all return qualities, the target user's return score at the time of hitting is obtained. This invention introduces a dynamic adversarial evaluation system based on a virtual opponent model and combines multi-source motion information fusion to achieve tactical-level quantitative evaluation and adaptive training control of hitting quality. This avoids evaluation biases such as high scores for easy shots or low scores for difficult shots, improving evaluation accuracy. Attached Figure Description

[0017] Figure 1 This is a flowchart of a preferred embodiment of the tennis-based shot evaluation method of the present invention; Figure 2 This is a structural diagram of a preferred embodiment of the tennis-based shot evaluation system of the present invention; Figure 3 This is a structural diagram of a preferred embodiment of the terminal of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] In traditional tennis training, ball machines only provide fixed or random serving patterns, lacking quantitative feedback on training quality and failing to assess the user's hitting ability. Therefore, this invention discloses a method that can integrate multi-dimensional motion information to quantitatively score the quality of hitting.

[0020] The preferred embodiment of the tennis-based shot evaluation method of the present invention, such as... Figure 1 As shown, the tennis-based shot evaluation method includes the following steps: Step S10: Obtain multiple complete hitting processes of the target user within the target court, and extract the three-dimensional position coordinates of the target user, the ball machine's serving information, and the target user's return information from each complete hitting process.

[0021] Specifically, for the target's complete hitting process on the court, the location of this complete hitting process, the speed of the serve, the landing point of the serve (including the point where the serve lands on the user's racket), as well as the target user's return speed and return landing point are collected. Based on this information, the quality of this hit is further evaluated.

[0022] Specifically, the process involves acquiring multiple complete hitting sequences of a target user within a target court, where each complete hitting sequence represents the moment when the target user hits and returns the tennis ball after it is launched from the ball machine, and the tennis ball lands on the ground. For each complete hitting process, the complete hitting process is analyzed to obtain the ball speed of the ball machine and the return speed of the target user; A court coordinate system is constructed with the center point of the target court as the origin, and the three-dimensional position coordinates of the target user, the ball landing point of the ball machine, and the return ball landing point of the target user are extracted from the court coordinate system.

[0023] The process of launching a tennis ball from the ball machine to the ball landing on the ground or onto the user's racket is defined as a complete hitting process. Once a complete hitting process is obtained, it can be modeled to obtain a hitting event. : .

[0024] Furthermore, for each complete hitting process, a hitting event model is constructed based on the three-dimensional position coordinates, the serve landing point, the serve speed, the return ball landing point, and the return ball speed; The three-dimensional position coordinates are represented as follows: ; in, Represents the three-dimensional location of the target user. The x-coordinate represents the target user. The vertical axis represents the target user. The vertical coordinate represents the target user; The coordinates of the serve landing point are represented as follows: ; in, Indicates the landing point of the serve. The x-coordinate represents the point where the serve lands. The vertical coordinate represents the landing point of the serve. The vertical coordinate of the serve's landing point; The serve speed Represented as: ; in, The lateral velocity representing the speed of the serve. The longitudinal velocity representing the speed of the serve. Vertical velocity representing the speed of the serve; The coordinates of the ball's landing point are represented as follows: ; in, Indicates the landing point of the return ball. The x-coordinate represents the point where the ball lands. The vertical coordinate represents the point where the ball lands. The vertical coordinate of the ball's landing point; The ball return speed Represented as: ; in, The lateral velocity representing the speed of the return ball. The longitudinal velocity representing the speed of the return ball. The vertical velocity representing the speed of the ball's return.

[0025] This invention uses the complete ball-hitting process as the smallest unit of analysis, ensuring a closed-loop data flow from the ball machine's ball release to the user's return ball landing. This provides uninterrupted original data for subsequent analysis, avoiding temporal misalignment or missing actions caused by fragmented data collection. By independently analyzing the ball speed and return speed of each complete process, the invention can accurately quantify the user's reaction time and hitting power, eliminating interference from random fluctuations in the ball machine and making speed indicators comparable across different systems. Furthermore, by constructing a three-dimensional coordinate system with the center of the court as the origin and uniformly extracting the user's position, ball landing point, and return ball landing point, the invention incorporates the user's movement trajectory, the machine's ball release strategy, and the ball's landing distribution into the same reference framework from a spatial perspective. This facilitates subsequent calculations of displacement distance, angle deviation, and landing point clustering, thereby supporting a systematic diagnosis of user technical shortcomings, ball machine tactical adjustments, and even court area coverage. Overall, this process balances temporal continuity, speed accuracy, and spatial consistency, providing a reusable structured data foundation for intelligent training and evaluation.

[0026] Step S20: Construct a virtual opponent for the target user within the target court. For each complete shot sequence, simulate the virtual opponent's return shot to the target user based on the serve information and the return shot information to obtain the return shot quality in multiple dimensions.

[0027] Based on these constructed complete hitting processes, this invention further introduces a virtual opponent model to replace the traditional target area setting. It then uses the virtual opponent's ability to catch the user's return shot, or the state in which the virtual opponent catches the user's return shot, to inversely evaluate the quality of the user's return shot. The threat of the return shot is quantified based on whether the virtual opponent can catch the ball or their posture when catching it (such as movement distance and hitting difficulty), rather than solely relying on the landing point coordinates. This comprehensively reflects the actual impact of speed, angle, and depth on the virtual opponent. Simultaneously, the model can flexibly adjust the opponent's ability level, enabling differentiated diagnosis for users of different skill levels. It also uses failed catches or passive states to clearly identify technical shortcomings, guiding users to optimize their hitting strategies and improving the targeting and intelligence of training.

[0028] Specifically, a virtual opponent for the target user is constructed on one side of the ball machine within the target court, wherein the virtual opponent is fixed at the center of the baseline of the target court; The difficulty level is set according to user needs. Based on the difficulty level and the coordinates of the virtual opponent, the catchable area of ​​the virtual opponent is constructed: ; ; in, Indicates the location of the virtual opponent. The coordinates representing the center position of the baseline. Indicates the area where the ball can be received. This represents any possible position the ball can reach in the next moment. This indicates the maximum movement speed of the virtual opponent. This represents the sum of the virtual opponent's reaction time and startup time. A shot quality assessment model is constructed, and the receiveable area, the serve information, and the return information of each complete shot process are input into the shot quality assessment model for comprehensive analysis to obtain the return quality of the ball in multiple dimensions.

[0029] Among them, the quality of the return shot includes: the quality of the shot breakthrough, the quality of the shot direction, the quality of the shot depth, and the quality of the stability. After creating a virtual opponent, the reachable area of ​​the virtual opponent can be constructed according to the user-defined difficulty level. This area is the area that the virtual opponent can reach at its maximum speed within the time difference between the user's reaction time from returning the ball to starting to act and the start time from the user's return ball landing. If the virtual opponent can easily receive the user's return ball, it means that the return ball is easy to receive, thus judging that the quality of the return ball is not high. If the virtual opponent does not receive the return ball within the bounds of the court, it can be judged that the quality of the return ball is higher than the difficulty level used for customization.

[0030] By introducing the time difference between reaction time and start-up time, and constructing a reachable area based on maximum speed, the movement ability of the virtual opponent is quantified as a dynamic receiving boundary. This shifts the evaluation of return shots from static landing points to realistic defensive timeliness, making it closer to the logic of actual combat. It supports user-defined difficulty levels, allowing flexible adjustment of the opponent's reaction speed and movement speed to achieve a precise match between training intensity and user skill level, avoiding a one-size-fits-all evaluation standard. By judging whether the opponent can receive the return shot without it going out of bounds, it not only intuitively distinguishes between "easily received" and "unable to receive," but also reversely determines whether the quality of the return shot is higher than the current difficulty level, thus providing clear feedback for advancement or demotion, effectively guiding users to improve weaknesses such as speed, angle, or depth.

[0031] Furthermore, a shot quality assessment model is constructed, and the receiveable area, the serve information, and the return information are input into the shot quality assessment model; Determine whether the return ball landing point in the return ball information is within the reachable area. If not, the shot breakthrough quality is determined to be high. If so, the shot breakthrough quality is calculated based on the position of the return ball landing point within the reachable area. ; in, Indicates the quality of the shot's breakthrough. Indicates the landing point of the return ball. This indicates the quality of the shot when the return ball lands within the reachable area. express The boundary, Represents the distance function. Parameters that control the rate of decay; The return direction vector is calculated based on the return ball landing point in the return ball information and the service ball landing point in the service ball information. The opponent direction vector is calculated based on the position of the virtual opponent and the service ball landing point. The quality of the hitting direction of the return ball is evaluated based on the return ball direction vector and the opponent direction vector. ; ; ; in, This represents the direction vector of the ball's return. Indicates the landing point of the serve. This represents the opponent's direction vector. Indicates the quality of the direction of the shot. Indicates the angle of the return shot; Calculate the depth mass of the return shot based on the vertical coordinate of the ball's landing point and the vertical coordinate of the virtual opponent: ; in, Indicates the depth and quality of the shot. The vertical coordinate represents the landing point of the ball; By calculating the variance of all return ball landing points in each complete shot sequence, the stability quality of the target user during the return shot can be obtained. ; in, Indicates stability quality. This indicates variance calculation.

[0032] In addition to defining the quality of the shot that the virtual opponent can achieve, the system also defines the quality of the shot direction based on the angle between the return shot direction and the virtual opponent's direction. The greater the directional advantage, the further the return shot deviates from the virtual opponent's frontal defensive area, the stronger the attack, and therefore the higher the shot direction quality. Using the baseline as a reference, a shot depth index is defined. The closer the return shot is to the baseline area, the stronger the suppression, the higher the score, and therefore the higher the shot depth quality. Finally, the system calculates the variance of the user's return shot landing point in each complete shot sequence to obtain the shot direction quality, which can be used to judge the stability and consistency of the user's return shot.

[0033] This invention defines the quality of the shot direction by calculating the angle between the return shot direction and the virtual opponent's direction. Its advantage lies in quantifying the degree to which the return shot utilizes the virtual opponent's defensive gaps; a larger angle indicates a greater deviation from direct defense and stronger attack, thus directly reflecting the tactical threat and penetrating power of the return shot. It defines the quality of the shot depth based on the baseline, linking the return shot's landing point to the baseline distance. The closer the landing point is to the baseline, the stronger the suppression and the higher the scoring probability, effectively simulating depth control in actual matches, encouraging users to actively challenge the baseline area, and enhancing the offensive pressure. Furthermore, it assesses the stability of the direction quality by statistically analyzing the variance of the return shot's landing point across multiple complete shot sequences. This variance objectively reflects the consistency and repeatability of the user's movements, avoiding interference from single, accidental factors, and helping to identify systematic deviations in power generation, racket face, or footwork adjustments, providing clear data for subsequent targeted training and tactical correction.

[0034] Step S30: Based on the three-dimensional position coordinates and the ball-serving information, evaluate the ball-serving machine's hitting difficulty in multiple dimensions.

[0035] In addition to the quality level of the user's return, this invention also introduces the difficulty of a virtual opponent catching the return to judge the return score, thereby avoiding the evaluation bias of "high score for easy ball" or "low score for difficult ball", making the score more in line with the training incentive mechanism; the difficulty of hitting the ball includes: spatial distance difficulty, speed difficulty and time pressure difficulty.

[0036] Specifically, based on the three-dimensional position coordinates and the serve landing point in the serve information, the spatial distance difficulty is constructed as follows: ; in, Indicating spatial distance difficulty Indicates the landing point of the serve. The x-coordinate represents the point where the serve lands. The vertical coordinate represents the landing point of the serve. The vertical coordinate of the serve landing point. Represents the three-dimensional location of the target user. The x-coordinate represents the target user. The vertical axis represents the target user. The vertical coordinate represents the target user; Calculate the speed value of the serve speed in the serve information, and define the speed value as the speed difficulty: ; in, The speed value represents the difficulty or speed of the serve. Indicates the speed of the serve. The lateral velocity representing the speed of the serve. The longitudinal velocity representing the speed of the serve. Vertical velocity representing the speed of the serve; Calculate the target user's maximum reaction time throughout all complete shot sequences, and calculate the time pressure difficulty for each return shot by the target user based on the maximum reaction time: ; in, This indicates the time pressure and difficulty of the current return shot. This indicates the maximum reaction time for the current return ball. This indicates the reaction time for the current return ball.

[0037] The spatial distance difficulty can be calculated based on the hitting point and the return point. To unify the dimensions of all difficulties, the spatial distance difficulty is further normalized by dividing it by the maximum movable distance of the target court. Similarly, the normalized result for speed difficulty is obtained by dividing it by the maximum speed of the ball machine. The normalized result for time pressure difficulty requires analyzing the experience value (i.e., the average reaction time of the virtual opponent in multiple returns by the user). Dividing the time pressure difficulty by the experience value yields its normalized result.

[0038] The system incorporates a virtual opponent's ball-receiving difficulty as a scoring criterion, effectively correcting the traditional bias of "high scores for easy balls and low scores for difficult balls." This makes the ball return scores more closely reflect the actual competitive effect, enhancing the fairness and relevance of training incentives. The system constructs spatial distance difficulty based on three-dimensional coordinates and the ball's landing point, normalizing it by dividing by the maximum movable distance of the court. This accurately quantifies the user's running coverage requirements while eliminating differences in court size, ensuring comparability across different scenarios. Speed ​​difficulty is defined by the vector magnitude of the ball's speed and normalized to the maximum speed of the ball machine, objectively reflecting the compression effect of ball speed on reaction and movement, avoiding inflated scores for low-speed balls. Time pressure difficulty is calculated by using the ratio of maximum reaction time to actual reaction time, normalized by dividing by the user's historical average reaction time (experience value), dynamically adapting to individual user differences and more realistically reflecting time urgency. Normalizing the three heterogeneous dimensions and unifying their dimensions facilitates weighted fusion, thereby comprehensively evaluating the overall picture of the shot and improving the scientific and personalized level of the evaluation system.

[0039] Step S40: After multiple weighted fusions of all the said hitting difficulties and all the said return quality, the return score of the target user when hitting the ball is obtained.

[0040] Specifically, the overall shot quality is obtained by weighting and integrating the quality of the shot's breakthrough, direction, depth, and stability. ; in, Indicates the overall quality of the shot. Indicates the quality of the shot's breakthrough. Indicates the quality of the direction of the shot. Indicates the depth and quality of the shot. Indicates stability quality. , , and They represent , , and Quality weights; The overall difficulty of hitting the ball is obtained by weighting and combining the difficulties of spatial distance, speed, and time pressure separately: ; in, Indicates the overall difficulty of the shot. Indicating spatial distance difficulty The speed value represents the difficulty or speed of the serve. This indicates the time pressure and difficulty of the current return shot. , and They represent , and Difficulty weighting; Calculate the return score of the target user's shots in all complete shot sequences based on the overall shot quality and the overall shot difficulty: ; in, Indicates the score for the return shot. This indicates the difficulty reward factor.

[0041] Specifically, for all return ball quality, this invention constructs corresponding quality weights for weighted fusion, and each quality weight is adjusted according to the training objective, user skill level, and training mode. By setting different weights, this invention can adapt to different training objectives within the same scoring framework, thereby improving the flexibility, interpretability, and applicability of the scoring model; for example, in offensive training mode, it can improve... and The system places greater emphasis on whether the user can produce tactically threatening returns, such as hitting shots to areas difficult for the opponent to cover, or creating an offensive advantage through changes in direction or angles. In the basic training mode, this can improve... The weighting, and appropriately reduced and The weighting is adjusted so that the training goal is not to pursue extremely strong aggression, but to help users develop stable movements and basic ball return ability.

[0042] Furthermore, for different levels of difficulty, this invention also constructs corresponding difficulty weights for weighted fusion to obtain a comprehensive difficulty. Finally, based on the comprehensive shot quality and comprehensive shot difficulty, the target user's return score for shots in all complete shot sequences is calculated. As a difficulty reward factor, it is used to adjust the degree of influence of the hitting difficulty on the final score. The hitting quality mainly reflects the quality of the user's return, while the hitting difficulty reflects the challenge level of the scenario in which the user completed the hit. If the hitting difficulty and hitting quality are directly combined in a fixed ratio, it may result in the difficulty factor having too strong or too weak an influence on the final score. Therefore, this parameter will be adjusted according to the training difficulty set by the user. Beginner users can get reasonable rewards for completing the return at an appropriate difficulty level; advanced users need to complete high-quality returns at a higher difficulty level to get high scores.

[0043] This invention employs a weighted fusion method with adjustable quality weights. Its advantage lies in the ability to flexibly adjust the weights of each dimension based on training objectives, user skill level, and training mode. This allows the same scoring framework to highlight offensive tactical threats (e.g., increasing directional and depth weights) while also strengthening the stability training of novice users (e.g., increasing consistency weights), significantly improving the model's adaptability and personalization. Independent weights are set for spatial, speed, and time difficulties, which are then fused into a comprehensive difficulty level. An adjustable difficulty reward factor is introduced to regulate the impact of difficulty on the final score. This avoids biases caused by fixed-ratio combinations, ensuring that novice users receive positive incentives at reasonable difficulty levels, while advanced users need to achieve high-quality returns under greater challenges to obtain higher scores, thus achieving a dynamic match between the score and the user's ability. Finally, the return score is calculated based on a weighted relationship between comprehensive shot quality and comprehensive difficulty. This approach balances the quality of the result with the challenge of the process, encouraging tactical advancement while not neglecting fundamental accumulation, making the evaluation system more scientific, fair, and interpretable, effectively supporting tiered training and progressive improvement.

[0044] This invention introduces a dynamic adversarial evaluation system based on a virtual opponent model, and combines it with multi-source motion information fusion to achieve tactical-level quantitative evaluation and adaptive training control of shot quality. This can avoid evaluation bias of high scores for simple shots or low scores for difficult shots, and improve evaluation accuracy.

[0045] Furthermore, such as Figure 2 As shown, based on the above-described tennis-based shot evaluation method, the present invention also provides a tennis-based shot evaluation system, wherein the tennis-based shot evaluation system includes: Information extraction module 51 is used to acquire multiple complete hitting processes of the target user in the target court, and extract the three-dimensional position coordinates of the target user, the ball machine's serving information and the target user's return information from each complete hitting process; The quality assessment module 52 is used to construct a virtual opponent of the target user within the target court, and for each complete shot process, simulate the virtual opponent's return shot of the target user based on the serve information and the return shot information to obtain the return shot quality in multiple dimensions; The difficulty assessment module 53 is used to assess the hitting difficulty of the ball-serving machine in multiple dimensions based on the three-dimensional position coordinates and the ball-serving information. The ball return scoring module 54 is used to obtain the ball return score of the target user when hitting the ball by performing multiple weighted fusions of all the said hitting difficulties and all the said ball return quality.

[0046] Furthermore, such as Figure 3 As shown, based on the above-mentioned tennis-based shot evaluation method and system, the present invention also provides a terminal, which includes a processor 10, a memory 20 and a display 30. Figure 3 Only some of the terminal components are shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0047] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory. In other embodiments, the memory 20 may be an external storage device of the terminal, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Further, the memory 20 may include both internal and external storage devices. The memory 20 is used to store application software and various types of data installed on the terminal, such as the program code installed on the terminal. The memory 20 can also be used to temporarily store data that has been output or will be output. In one embodiment, the memory 20 stores a tennis-based hit evaluation program 40, which can be executed by the processor 10 to implement the tennis-based hit evaluation method of this application.

[0048] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor, or other data processing chip, used to run program code stored in the memory 20 or process data, such as executing the tennis-based shot evaluation method.

[0049] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display 30 is used to display information on the terminal and to display a visual user interface. The components of the terminal communicate with each other via a system bus.

[0050] In one embodiment, when the processor 10 executes the tennis-based hit evaluation program 40 in the memory 20, it implements the steps of the tennis-based hit evaluation method as described above.

[0051] The present invention also provides a computer-readable storage medium storing a tennis-based hit evaluation program, which, when executed by a processor, implements the steps of the tennis-based hit evaluation method as described above.

[0052] In summary, this invention provides a tennis-based shot evaluation method and related equipment. The method includes: acquiring multiple complete shot sequences of a target user on a target court, and extracting the target user's three-dimensional position coordinates, the ball machine's serving information, and the target user's return information from each complete shot sequence; constructing a virtual opponent for the target user on the target court, and simulating the virtual opponent's return shot of the target user based on the serving and return information for each complete shot sequence, obtaining the return quality of the return shot in multiple dimensions; evaluating the shot difficulty of the ball machine in multiple dimensions based on the three-dimensional position coordinates and the serving information; and obtaining the target user's return score at the time of the shot by performing multiple weighted fusions of all shot difficulties and all return qualities. This invention introduces a dynamic adversarial evaluation system based on a virtual opponent model and combines multi-source motion information fusion to achieve tactical-level quantitative evaluation and adaptive training control of shot quality, avoiding evaluation bias such as high scores for easy shots or low scores for difficult shots, and improving evaluation accuracy.

[0053] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal that includes that element.

[0054] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.). The program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The computer-readable storage medium can be a memory, magnetic disk, optical disk, etc.

[0055] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A tennis-based shot evaluation method, characterized in that, The tennis-based shot evaluation method includes: The system acquires multiple complete shot sequences of a target user within a target court and extracts the target user's three-dimensional position coordinates, the ball machine's serving information, and the target user's return information from each complete shot sequence. Within the target court, a virtual opponent for the target user is constructed. For each complete shot sequence, the virtual opponent is simulated to return the ball to the target user based on the serve information and the return information, thereby obtaining the return quality in multiple dimensions. Based on the three-dimensional position coordinates and the ball-serving information, the difficulty of hitting the ball by the ball-serving machine in multiple dimensions is evaluated; After multiple weighted fusions of all the stated shot difficulty and all the stated return quality, the return score of the target user at the time of the shot is obtained.

2. The tennis-based shot evaluation method according to claim 1, characterized in that, The process of acquiring multiple complete shot sequences of the target user within the target court, and extracting the target user's three-dimensional position coordinates, the ball machine's serve information, and the target user's return information from each complete shot sequence, specifically includes: The system acquires multiple complete hitting sequences of a target user within a target court, wherein each complete hitting sequence represents the moment when the target user hits and returns the tennis ball after it is served by the ball machine and the tennis ball lands on the ground. For each complete hitting process, the complete hitting process is analyzed to obtain the ball speed of the ball machine and the return speed of the target user; A court coordinate system is constructed with the center point of the target court as the origin, and the three-dimensional position coordinates of the target user, the ball landing point of the ball machine, and the return ball landing point of the target user are extracted from the court coordinate system.

3. The tennis-based shot evaluation method according to claim 2, characterized in that, The process involves constructing a court coordinate system with the center point of the target court as the origin, and extracting the three-dimensional position coordinates of the target user, the ball landing point of the ball machine, and the return ball landing point of the target user from the court coordinate system. The process further includes: For each complete hitting process, a hitting event model is constructed based on the three-dimensional position coordinates, the serve landing point, the serve speed, the return ball landing point, and the return ball speed; The three-dimensional position coordinates are represented as follows: ； in, Represents the three-dimensional location of the target user. The x-coordinate represents the target user. The vertical axis represents the target user. The vertical coordinate represents the target user; The coordinates of the serve landing point are represented as follows: ； in, Indicates the landing point of the serve. The x-coordinate represents the point where the serve lands. The vertical coordinate represents the landing point of the serve. The vertical coordinate of the serve's landing point; The serve speed Represented as: ； in, Lateral velocity, representing the speed of the serve. The longitudinal velocity representing the speed of the serve. Vertical velocity representing the speed of the serve; The coordinates of the ball's landing point are represented as follows: ； in, Indicates the landing point of the return ball. The x-coordinate represents the point where the ball lands. The vertical coordinate represents the landing point of the ball. The vertical coordinate of the ball's landing point; The ball return speed Represented as: ； in, The lateral velocity representing the speed of the return ball. The longitudinal velocity representing the speed of the return ball. The vertical velocity representing the speed of the ball's return.

4. The tennis-based shot evaluation method according to claim 1, characterized in that, The process involves constructing a virtual opponent for the target user within the target court. For each complete shot sequence, the virtual opponent simulates the return shot from the target user based on the serve information and the return shot information, thereby obtaining the return shot quality in multiple dimensions, specifically including: A virtual opponent for the target user is constructed on one side of the ball machine within the target court, wherein the virtual opponent is fixed at the center of the baseline of the target court; The difficulty level is set according to user needs. Based on the difficulty level and the coordinates of the virtual opponent, the catchable area of ​​the virtual opponent is constructed: ； ； in, Indicates the location of the virtual opponent. The coordinates representing the center position of the baseline. Indicates the area where the ball can be received. This represents any possible position the ball can reach in the next moment. This indicates the maximum movement speed of the virtual opponent. This represents the sum of the virtual opponent's reaction time and startup time. A shot quality assessment model is constructed, and the receiveable area, the serve information, and the return information of each complete shot process are input into the shot quality assessment model for comprehensive analysis to obtain the return quality of the ball in multiple dimensions.

5. The tennis-based shot evaluation method according to claim 4, characterized in that, The quality of the return shot includes: the quality of the shot's breakthrough, the quality of the shot's direction, the quality of the shot's depth, and the quality of its stability. The construction of the shot quality assessment model involves inputting the receiveable area, serve information, and return information for each complete shot sequence into the model for comprehensive analysis, thereby obtaining the return quality in multiple dimensions, specifically including: Construct a shot quality assessment model by inputting the receiveable area, the serve information, and the return information into the shot quality assessment model; Determine whether the return ball landing point in the return ball information is within the reachable area. If not, the shot breakthrough quality is determined to be high. If so, the shot breakthrough quality is calculated based on the position of the return ball landing point within the reachable area. ； in, Indicates the quality of the shot's breakthrough. Indicates the landing point of the return ball. This indicates the quality of the shot when the return ball lands within the reachable area. express The boundary, Represents the distance function. Parameters that control the rate of decay; The return direction vector is calculated based on the return ball landing point in the return ball information and the service ball landing point in the service ball information. The opponent direction vector is calculated based on the position of the virtual opponent and the service ball landing point. The quality of the hitting direction of the return ball is evaluated based on the return ball direction vector and the opponent direction vector. ； ； ； in, This represents the direction vector of the ball's return. Indicates the landing point of the serve. This represents the opponent's direction vector. Indicates the quality of the direction of the shot. Indicates the angle of the return shot; Calculate the depth mass of the return shot based on the vertical coordinate of the ball's landing point and the vertical coordinate of the virtual opponent: ； in, Indicates the depth and quality of the shot. The vertical coordinate represents the landing point of the ball; By calculating the variance of all return ball landing points in each complete shot sequence, the stability quality of the target user during the return shot can be obtained. ； in, Indicates stability quality. This indicates variance calculation.

6. The tennis-based shot evaluation method according to claim 1, characterized in that, The difficulty of hitting the ball includes: spatial distance difficulty, speed difficulty, and time pressure difficulty; The step of evaluating the hitting difficulty of the ball-serving machine in multiple dimensions based on the three-dimensional position coordinates and the ball-serving information specifically includes: Based on the three-dimensional position coordinates and the serve landing point in the serve information, the spatial distance difficulty is constructed as follows: ； in, Indicating spatial distance difficulty Indicates the landing point of the serve. The x-coordinate represents the point where the serve lands. The vertical coordinate represents the landing point of the serve. The vertical coordinate of the serve landing point. Represents the three-dimensional location of the target user. The x-coordinate represents the target user. The vertical axis represents the target user. The vertical coordinate represents the target user; Calculate the speed value of the serve speed in the serve information, and define the speed value as the speed difficulty: ； in, The speed value represents the difficulty or speed of the serve. Indicates the speed of the serve. The lateral velocity representing the speed of the serve. The longitudinal velocity representing the speed of the serve. Vertical velocity representing the speed of the serve; Calculate the target user's maximum reaction time throughout all complete shot sequences, and calculate the time pressure difficulty for each return shot by the target user based on the maximum reaction time: ； in, This indicates the time pressure and difficulty of the current return shot. This indicates the maximum reaction time for the current return ball. This indicates the reaction time for the current return ball.

7. The tennis-based shot evaluation method according to claim 1, characterized in that, The target user's return score is obtained by weighting and fusing all the shot difficulties and all the return quality multiple times, specifically including: The overall shot quality is obtained by weighting and integrating the quality of the shot's breakthrough, direction, depth, and stability. ； in, Indicates the overall quality of the shot. Indicates the quality of the shot's breakthrough. Indicates the quality of the direction of the shot. Indicates the depth and quality of the shot. Indicates stability quality. , , and They represent , , and Quality weights; The overall difficulty of hitting the ball is obtained by weighting and combining the difficulties of spatial distance, speed, and time pressure separately: ； in, Indicates the overall difficulty of the shot. Indicating spatial distance difficulty The speed value represents the difficulty or speed of the serve. This indicates the time pressure and difficulty of the current return shot. , and They represent , and Difficulty weighting; Calculate the return score of the target user's shots in all complete shot sequences based on the overall shot quality and the overall shot difficulty: ； in, Indicates the score for the return shot. This indicates the difficulty reward factor.

8. A tennis-based shot evaluation system, characterized in that, The tennis-based shot evaluation system is used to implement the tennis-based shot evaluation method as described in any one of claims 1-7, wherein the tennis-based shot evaluation system comprises: The information extraction module is used to acquire multiple complete hitting processes of the target user in the target court, and extract the three-dimensional position coordinates of the target user, the ball machine's serving information, and the target user's return information from each complete hitting process; The quality assessment module is used to construct a virtual opponent for the target user within the target court. For each complete shot, the module simulates the virtual opponent's return shot to the target user based on the serve information and the return shot information, and obtains the return shot quality in multiple dimensions. The difficulty assessment module is used to assess the hitting difficulty of the ball-serving machine in multiple dimensions based on the three-dimensional position coordinates and the ball-serving information. The ball return scoring module is used to obtain the ball return score of the target user when hitting the ball by performing multiple weighted fusions of all the said hitting difficulties and all the said ball return quality.

9. A terminal, characterized in that, The terminal includes: a memory, a processor, and a tennis-based hit evaluation program stored in the memory and executable on the processor, wherein the tennis-based hit evaluation program, when executed by the processor, implements the steps of the tennis-based hit evaluation method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a tennis-based hit evaluation program, which, when executed by a processor, implements the steps of the tennis-based hit evaluation method as described in any one of claims 1-7.

Images