Smart Billiards Training System

Abstract

In the field of image processing technology, we provide a smart billiards training system. [Solution] This smart billiards training system uses multi-sensor fusion technology, including depth cameras, laser radar, and high-speed cameras, to accurately construct a three-dimensional model of the billiard table and ball, creating a more realistic training environment. The path simulation module automatically calculates the optimal hitting angle and hitting strength, helping the player find the optimal attack path. The system measures the friction coefficient of the tablecloth in real time and performs trajectory correction based on different tablecloth materials to ensure the calculated cue ball speed matches the actual situation and prevent trajectory deviation. If the player's hitting angle or strength deviates from the optimal calculated value, the system automatically corrects and provides optimization advice. The system also monitors the player's stance, grip, and body balance.

Description

[Technical Field]

[0001] The present application relates to the field of image processing technology, and more particularly to a smart billiards training system. [Background technology]

[0002] Billiards training is an exercise that requires extremely high precision, and its training effect is influenced by various factors, including the practitioner's hitting strength, hitting angle, the impact trajectory between the cue ball and the target ball, and the friction coefficient of the tablecloth. In traditional billiards training, practitioners generally practice based on personal experience and lack quantitative indicators to evaluate hitting accuracy. Because practitioners are unable to accurately grasp the appropriate hitting strength and angle, training efficiency is constantly reduced and it is difficult to achieve accurate hitting. Furthermore, traditional training methods mainly rely on visual judgment and repeated trials, lacking scientific error analysis and path optimization, making it difficult for practitioners to systematically improve their skill level.

[0003] Although existing partial billiards training devices can provide some trajectory analysis or video playback functions, they still have many shortcomings. For example, most devices cannot accurately measure the hit strength of the practitioner, and they lack the ability to dynamically adjust the friction coefficient of the tablecloth, which causes discrepancies between the simulated cue ball movement trajectory and the actual situation. The friction coefficient of the tablecloth may vary slightly depending on the material. Ignoring these details will result in errors between the simulation results and the actual hitting situation, which will affect the practitioner's judgment and reduce the accuracy and effectiveness of training.

[0004] Furthermore, if the friction coefficient of the tablecloth is not accurately taken into account during the impact trajectory simulation, the cue ball's movement trajectory will deviate significantly from the actual situation, causing the practicer to make adjustments based on inaccurate data and thereby increasing training errors. For example, in high-friction areas, the cue ball's velocity attenuation is rapid, while in low-friction areas, the cue ball may move farther. If the training system cannot accurately simulate these details, the practicer may mistakenly attribute the problem to their own technical issues rather than field factors, which may lead to incorrect adjustments of the impact strength or angle, further affecting the training effect. Summary of the Invention [Problem to be solved by the invention]

[0005] In response to the shortcomings of the prior art, the present application provides a smart billiard training system to solve the problems mentioned in the background art. [Means for solving the problem]

[0006] In order to achieve the above object, the present application provides a smart billiards training system, a three-dimensional billiard scene modeling module, which is used to acquire image data and three-dimensional data of the billiard table and balls using a depth camera and a laser radar, construct a three-dimensional model of the billiard table, and project the trajectories of the cue ball and target balls that a player hits each time onto the three-dimensional model; The path simulation module performs simulation analysis calculations to determine the optimal cue ball hitting angle θ when a trainee sets a target ball during training. hit and the initial velocity of the first cue ball V c and the friction coefficient μ of the tablecloth is collected, and the initial velocity V of the first cue ball is obtained. c Correction and conversion are performed to obtain the initial velocity of the second cue ball.

number

[0007] Preferably, the path simulation module includes an image acquisition unit and a simulation calculation unit; The image collection unit collects the coordinates of the ith pocket on the billiard table, (x p,i ,y p,i ) and the current cue ball coordinates (x m ,y m ), the coordinates of the target sphere (x t ,y t ) and the radius R of the target sphere, collecting a current spin state and a motion state of the cue ball, the spin state including no spin, top spin, and back spin; the motion states include stationary, rolling, and sliding states; When a practicer sets the target ball during training, the simulation calculation unit projects the cue ball and the target ball onto a three-dimensional model and calculates the optimal cue ball hitting angle θ by a collision physics algorithm. hit , the first cue ball initial velocity V c , the second cue ball initial velocity

number

number

number

number

number

[0008] Preferably, the first cue ball initial velocity V c and the second cue ball initial velocity

number

number

number

number

number

number

number

number

number

number

number

number

[0009] Preferably, the method for acquiring the impact strength F is as follows: the second cue ball initial velocity

number

number

[0010] Preferably, the error analysis module collects the training shot strength F1 and the training shot angle θ1 of the practice player through an angle sensor and a force sensor attached to the cue ball, and respectively calculates the optimal cue ball hitting angle θ1. hit and compared with the impact strength F, and used to calculate the impact strength error ΔF and the impact angle error Δθ using the following formula: ΔF = F1 - F, Δθ=θ1-θ hit , If ΔF≠0, the first warning signal is sent to notify the practitioner that the hitting strength is unacceptable; if ΔF>0, the first warning signal is sent to notify the practitioner that the hitting strength is excessive and the practitioner is advised to reduce the hitting strength; if ΔF<0, the first warning signal is sent to notify the practitioner that the hitting strength is insufficient and the practitioner is advised to increase the hitting strength; If Δθ≠0, the second warning signal is sent to notify the practicer that the impact angle is unacceptable; if Δθ>0, the practicer is notified that the tip is deviated to the left when impacting; and if Δθ<0, the practicer is notified that the tip is deviated to the right when impacting.

[0011] Preferably, the trajectory smart monitoring module includes: a second collecting unit; and a second calculating unit; a second collecting unit for collecting the actual movement trajectories and the predicted movement trajectories of the cue ball and the target ball after each swing by a practicer, and establishing the trajectory movement data set; The second calculation unit performs advanced calculation analysis based on the trajectory motion data set to calculate the cue ball trajectory deviation coefficient T cue_dev , the target ball motion deviation factor T starget_dev and the cue ball abnormal pop-out factor T out is used to calculate T cue_dev =a1×|θ cur_dev |+a2×|d cur_dev |+a3×|v cur_dev |+a4×|w cur_dev |, In the formula, θ cur_dev represents the trajectory deviation angle difference value of the cue ball, and d cur_dev represents the lateral displacement that the cue ball experiences during linear motion, and v cur_dev represents the deviation between the actual velocity of the cue ball and the theoretically calculated velocity, and w cur_dev represents the cue ball forward spin or back spin state deviation value, and a1, a2, a3, and a4 represent weighting coefficients; T starget_dev =a5×|θ target_dev |+a6×|d target_dev |+a7×|v strget_dev |+a8×|w target_dev |, In the formula, θ target_dev represents the angle between the actual driving direction of the target ball and the desired direction, and d target_dev represents the distance error between the final position of the target ball and the target point, and v strget_devrepresents the error between the actual velocity of the target ball and the desired velocity, and w target_dev represents the deviation between the rotation direction of the target ball and the rotation direction of the target ball in the predicted motion trajectory, and a5, a6, a7, and a8 represent weighting coefficients; T out =a9×|d cue_min |+a 10 ×|h cue |+a 11 ×|w cue |, In the formula, d cue_min represents the closest distance between the cue ball and the boundary of the billiard table, and h cue represents the jump height of the cue ball, and |w cue | represents the deviation value of the non-spin state of the cue ball, and a9, a 10 and a 11 represents the weighting coefficient.

[0012] Preferably, the trajectory smart monitoring module further includes: a first association unit; and a first alarm unit; The first association unit calculates the cue ball trajectory deviation coefficient T cue_dev , the target ball motion deviation factor T starget_dev and the cue ball abnormal pop-out factor T out After non-dimensionalizing the impact trajectory accuracy coefficient T accuracy is used to calculate T accuracy =1-(β1×T cue_dev +γ1×T starget_dev +δ1×T out ), In the formula, β1, γ1, and δ1 are the cue ball trajectory deviation coefficients T cue_dev , the target ball motion deviation factor T starget_dev and the cue ball abnormal pop-out factor T out is the weighting factor of The first alarm unit presets a passing threshold X and the impact trajectory accuracy coefficient T accuracy is compared with the predetermined pass threshold X, and the impact trajectory accuracy coefficient T accuracyIf the passing threshold X is less than the passing threshold X, it indicates that the practitioner's billiards training is unsuccessful, triggering the third warning signal, and generating a first policy that adjusts the current training plan, including increasing the tip angle training task by 10% to 20%, including backspin training, topspin training, and sidespin training; The impact trajectory accuracy coefficient T accuracy If the passing threshold X is greater than or equal to the passing threshold X, it indicates that the trainee's billiards training is passing, and is used to continue the current training plan.

[0013] Preferably, the smart billiard training system analyzes the stance, grip method, and center of gravity distribution of the player, and calculates a stance stability factor T stance , grip stability factor T hold and body balance factor T balance and a posture correction module for analytically calculating the attitude correction module includes a pressure collection unit, a grip collection unit, and an inertial collection unit; The pressure collection unit measures the left foot pressure Fσ of the exerciser using a pressure sensor attached to the ground pad. left and right foot pressure Fσ right is used to detect The grip collection unit measures the actual grip pressure P by a grip pressure sensor attached to the grip position of the cue. geip is used to detect The inertial collection unit measures the center of gravity (X) using a posture sensor attached to the waist of the trainee. cg ,Y cg ), longitudinal inclination angle A qhqx and the body sway velocity V boby It is used to monitor the following in real time.

[0014] Preferably, the posture correction module further includes a first analysis unit, a second analysis unit and a third analysis unit; The first analysis unit calculates the center of gravity position (X cg ,Y cg ), the left foot pressure Fσ of the exerciser left and right foot pressure Fσ right After extracting and non-dimensionalizing the stance stability factor T stance is used to calculate T stance =1-|Fσ left -Fσ right | / F total -(|X cg -X mid |+|Y cg -Y mid |) / D max , In the formula, F total represents the total body force, and F total =Fσ left +Fσ right and D max represents the maximum allowable deviation distance of the stance center of gravity, and 0 <T stance ≦1 and X cg represents the horizontal coordinate of the current center of gravity of the trainee, and Y cg represents the anterior-posterior coordinate of the current center of gravity of the trainee, and X mid represents the horizontal coordinate of the center of gravity in the standard stance, and Y mid represents the front-to-back coordinate of the center of gravity in the standard stance, X cg >X mid If it is, it indicates that the practitioner's current center of gravity is biased to the right, X cg <X mid If it is, it indicates that the practitioner's current center of gravity is biased to the left, Y cg >Y mid If this is the case, it indicates that the practitioner's current center of gravity is biased forward, causing forward leaning. Y cg <Y mid If this is the case, it indicates that the practitioner's center of gravity is currently biased backward, causing backbend. The second analysis unit calculates the actual grip pressure P geip The grip stability factor T is calculated using the following formula: hold is used to calculate T hold =1-|P geip -P opt | / P max -θ dev / θ max , In the formula, P geip is the actual grip pressure, and P opt represents the optimal grip pressure, and P max represents the maximum allowable grip pressure, and θ dev represents the cue body angle deviation value when gripping, and θ max represents the maximum allowable grip angle deviation value, The third analysis unit calculates the longitudinal inclination angle A qhqx and the body swing speed V boby After extracting and non-dimensionalizing the body balance factor T balance is used to calculate T balance =1-A qhqx / A max -V boby / V max , In the formula, A max represents the maximum allowable tilt angle, and V max represents the maximum allowable swing speed.

[0015] Preferably, the attitude correction module further includes a second association unit and a second alarm unit; The second association unit calculates the stance stability factor T stance , the grip stability factor T hold and the body balance factor T balance After extracting and non-dimensionalizing the standard posture index T pose is used to obtain T accuracy =β1×T stance +γ2×T hold +δ2×T balance , In the formula, β2, γ2, and δ2 are the stance stability factors T stance , the grip stability factor T hold and the body balance factor T balanceis the weighting factor of The second alarm unit presets a posture normalization threshold Z, and the posture normalization index T pose is compared with the posture standardized threshold Z to determine whether the posture of the trainee is acceptable, and includes: The posture standard index T pose If the posture standardization threshold Z is less than the posture standardization threshold Z, it indicates that the practitioner's posture during billiards training is unacceptable, and a fourth alarm signal is triggered. A second policy is generated to adjust the current training plan, adjust the stance, and add 30 seconds of stance balance training before each training, where the stance balance training includes one-legged standing training and core muscle training, and adjust the stance so that the center of gravity distribution is X. mid and Y mid Approach and increase grip strength by 5% with stability training The posture standard index T pose If Z is greater than or equal to the posture standardized threshold value Z, it indicates that the posture of the practitioner during billiard training is acceptable, and the current training plan is continued. [Effects of the Invention]

[0016] The present application provides a smart billiards training system, which has the following beneficial effects: (1) Using multi-sensor fusion technology including depth cameras, laser radar, and high-speed cameras, an accurate 3D model of the billiard table and ball is constructed, creating a more realistic training environment. The path simulation module automatically calculates the optimal hitting angle and hitting strength (F), helping the practitioner find the optimal attack path and reducing the cost of trial and error. The friction coefficient of the tablecloth is measured in real time, and trajectory correction is performed based on different tablecloth materials, ensuring that the calculated cue ball speed matches the actual situation and avoiding trajectory deviation. If the practitioner's hitting angle or strength deviates from the optimal calculated value, the system automatically corrects and provides optimization advice. (2) The smart billiards training system is integrated with a high-frame-rate motion capture device, which collects the actual and predicted motion trajectories of the player's swings and establishes a trajectory motion dataset for subsequent calculation and analysis. Based on the trajectory motion dataset, it intelligently calculates the cue ball trajectory deviation coefficient, target ball motion deviation factor, and cue ball abnormal launch factor, performs real-time analysis of the flight trajectory, and recognizes abnormal situations in jump shots and cue shots. The system can automatically match the optimal hitting trajectory and provide adjustment advice, effectively reducing errors caused by angle, speed, or rotation errors, allowing players to more accurately control the ball's landing point and improving their game performance. (3) The pressure collection unit, grip collection unit, and inertia collection unit are combined with a multi-factor calculation model to realize accurate quantification and real-time feedback of the practitioner's stance stability, grip stability, and body balance. The pressure collection unit detects the pressure of the practitioner's left and right feet, and combines it with the center of gravity position to calculate the stance stability factor T stance The system calculates the posture and analyzes the practitioner's center of gravity shift. If the stance stability is insufficient, the system automatically provides posture optimization advice, such as adjusting steps and conducting core stability training, to help the practitioner maintain an optimal stance. (4) The grip collection unit acquires the actual grip pressure of the practitioner, and calculates the grip stability factor T hold It calculates the grip stability factor T and analyzes in real time whether the grip force and cue body angle meet the optimal standard. hold Too small or too large indicates a large deviation in grip force or cue body angle, which may affect the hitting trajectory and require directional training (e.g., increasing grip force and adjusting practice). (5) The inertial collection unit monitors the forward and backward tilt angle and body swing speed of the exerciser, and calculates the body balance factor T balance The system calculates the stance stability factor Tstance , grip stability factor T hold and body balance factor T balance The weighted calculation is performed to obtain the posture standard index T pose This is used to comprehensively evaluate the posture passability of the trainee, and the posture standard index T pose If the preset threshold Z is reached, the system will trigger a fourth alarm signal and generate a second training policy to correct the practitioner's posture. Compared with traditional training modes, this system can accurately monitor stance, grip, and body balance, and quantitatively evaluate posture stability, allowing the practitioner to correct poor posture in a short time and improve hitting stability, thereby avoiding posture deviations caused by differences in the instructor's observation angle or the practitioner's own lack of perception. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a schematic flow chart of the smart billiard training system provided in the present application. DETAILED DESCRIPTION OF THE INVENTION

[0018] The following clearly and completely describes the technical solutions in the embodiments of this application with reference to the drawings in the embodiments of this application. The described embodiments are only some of the embodiments of this application, not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application without any creative work fall within the scope of protection of this application.

[0019] Example 1 Referring to FIG. 1, the present application provides a smart billiards training system, which includes a three-dimensional billiards scene modeling module, a path simulation module, an error analysis module, and a trajectory smart monitoring module.

[0020] The 3D billiard scene modeling module uses a depth camera and laser radar to acquire image data and 3D data of the billiard table and balls, constructs a 3D model of the billiard table, and projects the trajectories of the cue ball and target balls that the player hits each time onto the 3D model.

[0021] Specifically, a depth camera is used to obtain spatial coordinate information of the billiard table and surrounding environment in real time, forming point cloud data, combining RGB information, and converting the point cloud data into a color 3D model, making the training environment more intuitive. A self-adaptive depth learning algorithm is used to optimize edge detection and improve the accuracy of 3D modeling. A high-precision laser radar is used to perform detailed scans of the billiard table surface, and combined with an optical recognition algorithm, the position of the billiard table cushions, pockets, and ball distribution are detected in real time, and subsequent trajectory simulation provides accurate data support. A high-frame-rate (1000 fps or more) camera is used to record the movement trajectories of the cue ball and target ball in real time, ensuring accurate data capture at the moment of impact. The trajectories of the cue ball and target ball hit by the practitioner are projected, and by combining with visual recognition technology and image processing technology, their movement paths are reconstructed in a 3D model in real time. Spatial point cloud reconstruction technology is used to generate a high-precision billiard table model, ensuring that the movement paths of all balls in the virtual environment are based on their actual proportions.

[0022] The path simulation module performs simulation analysis calculations when the trainee sets the target ball during training to determine the optimal hitting angle of the cue ball θ hit and the initial velocity of the first cue ball V c and obtain the initial velocity of the first cue ball V c is the instantaneous velocity of the cue ball before it hits the target ball, and the friction coefficient μ of the tablecloth is taken into account. The initial velocity V of the first cue ball is c Correction and conversion are performed to obtain the initial velocity of the second cue ball.

number

[0023] In this embodiment, multi-sensor fusion technology, including depth cameras, laser radar, and high-speed cameras, is used to accurately construct a three-dimensional model of the billiard table and ball, creating a more realistic training environment. The path simulation module automatically calculates the optimal hitting angle and initial hitting speed, helping the practitioner find the optimal attack path and reducing the cost of trial and error. The friction coefficient of the tablecloth is measured in real time, and trajectory correction is performed based on different areas of the tablecloth to ensure that the calculated cue ball speed matches the actual situation and avoid trajectory deviation. If the practitioner's hitting angle or strength deviates from the optimal calculated value, the system automatically corrects and provides optimization advice.

[0024] The system detects the player's actual hitting strength and angle, compares them with the calculated optimal hitting strength F and optimal hitting angle, and calculates the error. If the hitting strength error ≠ 0, it triggers a first alarm signal, prompting the player to adjust the strength. If the hitting angle error ≠ 0, it triggers a second alarm signal, prompting the player to adjust the angle. The trajectory movement data set constructed by the trajectory smart monitoring module also includes a cue ball trajectory deviation coefficient (detects whether the cue ball moves along the desired path), a target ball movement deviation factor (calculates whether the target ball rolls along the desired direction), and a cue ball abnormal pop-up factor (detects whether the cue ball pops out abnormally). The hitting trajectory accuracy coefficient is calculated, and if it is lower than the set threshold X, it triggers a third alarm signal, prompting the player to adjust their hitting method. The module can accumulate training data over the long term, analyze the player's technical stability, and provide directionality improvement advice.

[0025] The system records the hit data of the practice user and supports slow-motion playback and keyframe analysis, helping the practice user review the details of their training and providing personalized training solutions for their shortcomings. For example, if the practice user consistently experiences a large intensity error, the system can recommend a precise intensity control training mode. If the practice user's angle control is unstable, the system can recommend an optimal hitting angle training mode.

[0026] Example 2 This embodiment is an interpretation and explanation of the first embodiment. Referring to FIG. 1, the path simulation module specifically includes an image collection unit and a simulation calculation unit.

[0027] The image collection unit collects the coordinates of the ith pocket on the billiard table, (x p,i ,y p,i ) and the current cue ball coordinates (x m ,y m ), the coordinates of the target sphere (x t ,yt ) and the radius R of the target ball, and collects the current rotation state and motion state of the cue ball. The rotation state includes no rotation, topspin, and backspin states, and the motion state includes stationary, rolling, and sliding states. When the trainee sets the target ball during training, the simulation calculation unit projects the cue ball and the target ball onto a three-dimensional model and calculates the optimal hitting angle θ of the cue ball using a collision physics algorithm. hit , the initial velocity of the first cue ball V c , second cue ball initial velocity

number

[0028] Optimal cue ball hitting angle θ hit Acquisition methods include the following:

[0029] In S11, if the target ball meets the requirement to enter the i-th pocket, the pocket entry vector of the target ball

number

number

number

[0030] In S12, the i-th pocket approach angle of the target ball is θ pocket,i Calculate θ pocket,i =arctan(y p,i -y t ) / (x p,i -x t ), i-th pocket approach direction angle θ pocket,i indicates how the target ball must move to land accurately in the pocket.

[0031] In S13, in order for the target ball to move towards the i-th pocket, the cue ball needs to contact the contact point of the target ball. First, the collision point coordinates (x c ,y c ) and x c =x t -R×cosθ pocket,i , y c =y t -R×sinθ pocket,i , The coordinates of the impact point of the target ball indicate the specific position where the cue ball must impact the target ball.

[0032] In S14, the cue ball hitting direction

number

number

number

[0033] In S15, the optimal hitting angle θ of the cue ball is determined to ensure accurate impact with the target ball. hit is calculated, and the expression is as follows: θ hit =arctan(y c -y m ) / (x c -x m ), In the formula, x m represents the x-axis coordinate of the cue ball along the long side of the billiard table in the Cartesian coordinate system, y m represents the y-axis coordinate of the cue ball along the short side of the billiard table in the Cartesian coordinate system.

[0034] In this embodiment, the optimal hitting angle θ of the cue ball hit Based on this, the practice user can learn how to aim the cue ball so that the target ball enters the pocket on an accurate trajectory. That is, by calculating the optimal hitting angle for the cue ball based on an actual physical model, it avoids traditional empirical judgment errors, reduces unnecessary trial and error, and improves the practice user's success rate in complex hitting situations. The system can automatically adapt to different target ball and pocket positions, improving the applicability of training. Furthermore, combined with real-time trajectory feedback, it can adjust training policies and allow the practice user to more quickly grasp hitting techniques.

[0035] Example 3 This embodiment is an interpretation and explanation of the first embodiment. Referring to FIG. 1, specifically, the first cue ball initial velocity V c and the initial velocity of the second cue ball

number

[0036] In S21, in order for the target ball to smoothly enter the pocket, it is necessary to calculate the hitting strength F of the cue ball. The hitting strength F of the cue ball is affected by the friction coefficient μ of the tablecloth. First, based on the law of conservation of momentum, the mass m of the cue ball is calculated. c and target ball mass m t Collect the cue ball mass m c and the target sphere mass m t If these are equal, the initial velocity of the first cue ball V c Calculate the initial velocity of the first cue ball, V c teeth, m c ×V c =m t ×V t The formula should be satisfied, In the formula, V t is the target ball velocity and the cue ball mass m c and the target sphere mass m t If the velocity of the cue ball before the collision is equal to the velocity of the target ball after the collision, and the target ball moves along the collision direction, then Cue ball mass m c ≠ Target sphere mass m t If so, the initial velocity of the first cue ball is V c teeth,

number

number

number

[0037] In S22, the target ball speed V t is attenuated by the friction coefficient μ of the tablecloth and the moving distance, and the target ball velocity V is calculated by the following formula: t Correct the target ball speed

number

number

[0038] An example graph of the coefficient of friction μ of a tablecloth is shown below. [Table 1]

[0039] If a tablecloth with a high wool content is used (for example, 91% to 100% fine wool), the friction is relatively small, and the target ball rolls farther. When a relatively coarse wool tablecloth (such as a 50% to 80% American eight-ball tablecloth) is used, the friction force is relatively large and the target ball speed decays more quickly.

[0040] In the actual hitting process, the target ball speed is affected by the friction of the tablecloth and decreases as the distance traveled increases.

number

[0041] In S23, the corrected target ball speed

number

number

number

number

number

number

[0042] In this embodiment, the billiards training system must accurately calculate the cue ball's impact force F and its initial velocity to ensure the target ball smoothly enters the pocket, and then correct the target ball's velocity by taking into account the effect of tablecloth friction. The system optimizes the cue ball's motion trajectory based on the law of conservation of momentum and the effect of friction, improving hitting accuracy. Based on momentum conservation, the system ensures that the cue ball's impact force F is appropriate, preventing hits that are too strong or too weak. The system corrects the target ball's velocity to better fit the actual playing environment, ensuring that the calculation results match the actual hitting trajectory. The calculation formula can be automatically adjusted based on different tablecloth materials, improving applicability. This helps practitioners find the optimal impact force and improves the success rate of hitting into the pocket. The system calculates the cue ball's velocity after impact, helping to create a continuous attack strategy and achieving advanced billiards tactical training.

[0043] Example 4 This embodiment is an interpretation and explanation of the first embodiment, and referring to FIG. 1, the method for acquiring the impact strength F is specifically as follows.

[0044] Second cue ball initial speed

number

number

[0045] The mass of a standard billiard cue ball is approximately 0.17kg to 0.18kg (American 8-ball, snooker, 9-ball, etc. vary slightly).

[0046] Depends on the hitting target and tactical requirements, for example, hits that require relatively strong force (such as hitting a long-distance ball with strong force) and hits that require controlled force (such as lightly pushing with small angle positioning). Short contact time (0.001s~0.002s): Generates stronger hitting strength and is suitable for powerful attacks, K balls, and long-distance hitting. Long contact time (0.003s~0.005s): Generates relatively weak hitting strength and is suitable for soft positioning, cushion balls, etc.

[0047] In this embodiment, the initial velocity of the cue ball is converted into the force applied by the tip, helping the practicer to determine the optimal force. The hitting force F can be measured in real time using a camera tracking system or a ball table sensor, and compared with the theoretically calculated value, improving the accuracy of training.

[0048] Example 5 This embodiment is an interpretation and explanation of the fourth embodiment. Referring to FIG. 1, specifically, the error analysis module collects the training hitting strength F1 and the training hitting angle θ1 of the practice player through the angle sensor and the force sensor attached to the cue ball, and calculates the optimal hitting angle θ1 of the cue ball. hit and impact strength F, and is used to calculate and obtain the impact strength error ΔF and impact angle error Δθ using the following formula: ΔF = F1 - F, Δθ=θ1-θ hit , If ΔF≠0, a first warning signal is sent to notify the trainee that the hitting strength is unacceptable; if ΔF>0, it is indicated that the trainee's hitting strength is excessive and the trainee is advised to reduce the hitting strength; if ΔF<0, it is indicated that the trainee's hitting strength is insufficient and the trainee is advised to increase the hitting strength; If Δθ≠0, a second warning signal is sent to notify the practicer that the impact angle is unacceptable; if Δθ>0, the practicer is notified that the tip is deviated to the left when hitting; and if Δθ<0, the practicer is notified that the tip is deviated to the right when hitting.

[0049] In this embodiment, the system's error analysis module uses angle and force sensors attached to the cue to collect the practitioner's hitting strength and training hitting angle in real time. It compares these data with the theoretically calculated optimal hitting angle and hitting strength F of the cue ball to calculate hitting strength and hitting angle errors and provide directional feedback adjustment. Real-time data comparison allows the practitioner to intuitively sense hitting errors and gradually adjust hitting strength and angle to improve accuracy. First and second warning signals allow the practitioner to immediately correct errors and prevent long-term impacts on performance due to incorrect posture. The system records the practitioner's hitting error data, analyzes common problems (e.g., excessive force, angle deviation), and provides personalized training advice.

[0050] Example 6 If the initial velocity of the cue ball is too high, it may bounce abnormally after hitting the cushion, and may even fly off the billiard table. If the tip is tilted too much, the cue ball may jump excessively, increasing the risk of it flying off the table. This is usually considered a fatal mistake in billiard training.

[0051] This embodiment is an interpretation and explanation of the first embodiment. Referring to FIG. 1, specifically, the trajectory smart monitoring module includes a second collecting unit and a second calculating unit. The second collecting unit is used to collect the actual movement trajectories and the predicted movement trajectories of the cue ball and the target ball after each swing by the practicer, and establish a trajectory movement data set. The second calculating unit performs advanced calculation analysis based on the trajectory movement data set, and calculates the cue ball trajectory deviation coefficient T according to the following formula: cue_dev , target ball motion deviation factor T starget_dev and Cue ball abnormal pop-out factor T out is used to calculate and obtain T cue_dev =a1×|θ cur_dev |+a2×|d cur_dev |+a3×|vcur_dev |+a4×|w cur_dev |, In the formula, θ cur_dev represents the deviation angle difference value of the cue ball trajectory, and d cur_dev represents the lateral displacement that occurs during the linear motion of the cue ball, and v cur_dev represents the deviation between the actual speed of the cue ball and the theoretical calculated speed, and w cur_dev represents the cue ball forward spin or back spin state deviation value, a1, a2, a3 and a4 represent weighting coefficients, and the sum of the weights is 1; T starget_dev =a5×|θ target_dev |+a6×|d target_dev |+a7×|v strget_dev |+a8×|w target_dev |, In the formula, θ target_dev represents the angle between the actual driving direction of the target ball and the desired direction, and d target_dev represents the distance error between the final position of the target ball and the target point, and v strget_dev represents the error between the actual velocity of the target ball and the desired velocity, and w target_dev represents the deviation between the rotation direction of the target ball and the rotation direction of the target ball in the predicted motion trajectory, a5, a6, a7 and a8 represent weighting coefficients, and the sum of the weights is 1; T out =a9×|d cue_min |+a 10 ×|h cue |+a 11 ×|w cue |, In the formula, d cue_min represents the closest distance between the cue ball and the billiard table boundary, and h cue represents the jump height of the cue ball, and |w cue | represents the deviation value of the cue ball in the no-spin state, and a9, a 10 and a 11 represents a weighting coefficient, and the sum of the weights is 1.

[0052] Cue ball trajectory deviation coefficient T cue_dev represents the deviation of the cue ball trajectory, and the target ball motion deviation factor T starget_dev represents the accuracy of the target ball's motion trajectory, and the cue ball abnormal jump factor T outindicates whether the cue ball will fly out abnormally.

[0053] In this embodiment, the actual and predicted motion trajectories of the cue ball and the target ball after each swing by the practicer are collected, and a trajectory motion data set is established for subsequent calculation analysis. Based on the trajectory motion data set, the cue ball trajectory deviation coefficient, the target ball motion deviation factor, and the cue ball abnormal pop-out factor are calculated in an advanced manner to evaluate the hitting quality and accuracy.

[0054] Example 7 This embodiment is an interpretation and description of the first embodiment, and with reference to FIG. 1, specifically, the trajectory smart monitoring module further includes a first association unit and a first alarm unit.

[0055] The first association unit calculates the cue ball trajectory deviation coefficient T cue_dev , target ball motion deviation factor T starget_dev and Cue ball abnormal pop-out factor T out After non-dimensionalizing the impact trajectory accuracy coefficient T accuracy is used to calculate T accuracy =1-(β1×T cue_dev +γ1×T starget_dev +δ1×T out ), In the formula, β1, γ1, and δ1 are the cue ball trajectory deviation coefficients T cue_dev , target ball motion deviation factor T starget_dev and Cue ball abnormal pop-out factor T out are weighting coefficients, and the sum of the weights is 1.

[0056] The first alarm unit presets a passing threshold X and a hitting trajectory accuracy coefficient T accuracy is compared with the preset pass threshold X, and the impact trajectory accuracy coefficient T accuracyIf X<the passing threshold, it indicates that the practitioner's billiards training is unsuccessful, and a third warning signal is triggered. A first policy is generated, which includes adjusting the current training plan to increase the tip angle training task by 10% to 20%, including backspin training, topspin training, and side spin training, and increasing the hitting trajectory accuracy coefficient T accuracy If the passing threshold X is satisfied, it indicates that the trainee's billiards training is passing, and the current training plan is continued.

[0057] In this embodiment, by calculating the cue ball trajectory deviation coefficient, target ball motion deviation factor, and cue ball abnormal launch factor, the system can comprehensively measure the practicer's hitting trajectory deviation and accurately analyze hitting quality. Practicers can intuitively grasp the deviation between the actual cue ball trajectory and the ideal trajectory, thereby optimizing their hitting strategy and improving hitting accuracy. A non-dimensional processing method is used to standardize multiple deviation data, eliminating the influence of different physical quantities and improving the scientific nature of data analysis. Calculating the hitting trajectory accuracy coefficient establishes a quantifiable evaluation standard, making the training process more objective and precise. By setting an alarm threshold X, if the hitting trajectory accuracy coefficient is lower than this threshold, the system automatically triggers a third alarm signal, prompting the practicer to adjust their hitting method and prevent the development of bad habits. This allows the practitioner to receive timely directional training advice, reducing poor hitting habits caused by long-term incorrect training. After the third alarm signal is triggered, the system automatically adjusts the training plan to increase tip angle training tasks by 10% to 20%, including specialized training such as backspin, topspin, and sidespin, thereby optimizing the practitioner's hitting control ability in a directional manner.

[0058] Example 8 In billiards training, accurate stance, gripping technique, and body center of gravity control are crucial for hitting accuracy and stability. However, traditional training methods often make it difficult for practitioners to intuitively detect their own posture deviations, leading to the following problems: The uneven distribution of the practitioner's center of gravity between both feet can lead to loss of balance, which can affect hitting stability. A too-tight grip limits tip control, while a too-loose grip can easily cause tip deviation, affecting the hitting trajectory and spin control. When aiming or hitting, the practitioner's body may sway, causing deviations in the cue ball trajectory and affecting accuracy, especially when hitting topspin, backspin, or sidespin. Traditional training primarily relies on instructor observation or the practitioner's subjective perception, which lacks data support, making it difficult to accurately identify posture errors and preventing the creation of directional correction training solutions.

[0059] This embodiment is an interpretation and explanation of the first embodiment. Referring to FIG. 1, the smart billiard training system analyzes the stance, gripping method, and distribution of the center of gravity of the player, and calculates the stance stability factor T stance , grip stability factor T hold and body balance factor T balance Further comprising an attitude correction module for analytically calculating and obtaining The posture correction module includes a pressure collection unit, a grip collection unit, and an inertial collection unit. The pressure collection unit collects the left foot pressure Fσ of the exerciser by a pressure sensor attached to the ground pad. left and right foot pressure Fσ right The grip collection unit detects the actual grip pressure P by a grip pressure sensor attached to the grip position of the cue. geip The inertial collection unit detects the center of gravity (X) using a posture sensor attached to the waist of the exerciser. cg ,Y cg ), longitudinal inclination angle A qhqxand the body sway velocity V boby It is used to monitor the following in real time.

[0060] The posture correction module further includes a first analysis unit, a second analysis unit and a third analysis unit.

[0061] The first analysis unit calculates the center of gravity (X cg ,Y cg ), the left foot pressure of the trainee Fσ left and right foot pressure Fσ right After extracting and non-dimensionalizing the stance stability factor T stance is used to calculate and obtain T stance =1-|Fσ left -Fσ right | / F total -(|X cg -X mid |+|Y cg -Y mid |) / D max , In the formula, F total represents the total body force, and F total =Fσ left +Fσ right and D max represents the maximum allowable deviation distance of the stance center of gravity, and 0 <T stance ≦1, and T stance The closer to 1, the more stable the stance is, and X cg represents the horizontal coordinate of the current center of gravity of the trainee, and Y cg represents the anterior-posterior coordinate of the current center of gravity of the trainee, and X mid represents the horizontal coordinate of the center of gravity in the standard stance, and Y mid represents the longitudinal coordinate of the center of gravity in the standard stance, and X cg >X mid If it is, it indicates that the practitioner's current center of gravity is biased to the right, and X cg <X mid If Y is the center of gravity of the practitioner, it indicates that the center of gravity is biased to the left. cg >Y midIf Y is the same, it indicates that the center of gravity of the exerciser is biased forward, causing forward leaning. cg <Y mid If this is the case, it indicates that the practitioner's center of gravity is currently biased backward, causing backbend.

[0062] The second analysis unit calculates the actual grip pressure P geip The grip stability factor T is calculated using the following formula: hold is used to calculate and obtain T hold =1-|P geip -P opt | / P max -θ dev / θ max , In the formula, P geip is the actual grip pressure, and P opt represents the optimal grip pressure, and P max represents the maximum allowable grip pressure, and θ dev represents the cue body angle deviation value when gripping, and θ max represents the maximum allowable grip angle deviation value, and T hold The closer to 1, the more stable the practitioner's grip is.

[0063] The third analysis unit is the longitudinal tilt angle A qhqx and body sway speed V boby After extracting and non-dimensionalizing the body balance factor T balance is used to calculate T balance =1-A qhqx / A max -V boby / V max , In the formula, A max represents the maximum allowable tilt angle, and V max represents the maximum allowable swing speed, and T balance The closer to 1, the better the physical balance of the practitioner.

[0064] The attitude correction module further includes a second associating unit and a second alarming unit.

[0065] The second association unit is the stance stability factor T stance , grip stability factor T hold and body balance factor T balance After extracting and non-dimensionalizing the standard posture index T pose is used to obtain T accuracy =β1×T stance +γ2×T hold +δ2×T balance , where β2, γ2, and δ2 are the stance stability factors T stance , grip stability factor T hold and body balance factor T balance are weighting coefficients, and the sum of the weights is 1.

[0066] The second alarm unit presets a posture normalization threshold Z and a posture normalization index T pose is compared with the posture standardization threshold Z to determine whether the trainee's posture is acceptable or not, and the posture standard index T pose If the posture standardization threshold Z is less than the posture standardization threshold Z, it indicates that the posture of the practitioner during billiards training is unacceptable, and a fourth alarm signal is triggered. A second policy is generated to adjust the current training plan, adjust the stance, and increase the stance balance training for 30 seconds before each training, where the stance balance training includes one-legged standing training and core muscle training, and adjust the stance so that the center of gravity distribution is X. mid and Y mid Increase your grip strength by 5% with stability training and improve your posture standard index (T) pose If the posture standardization threshold Z is greater than or equal to the posture standardization threshold Z, it indicates that the posture of the practitioner is acceptable during billiard training, and is used to continue the current training plan.

[0067] In this embodiment, the pressure collection unit, grip collection unit, and inertia collection unit are combined with a multi-factor calculation model to realize accurate quantification and real-time feedback of the practitioner's stance stability, grip stability, and body balance. The pressure collection unit detects the pressure of the practitioner's left and right feet, and combines it with the center of gravity position to calculate the stance stability factor T stance The system calculates the posture and analyzes the practitioner's center of gravity shift. If the stance stability is insufficient, the system will automatically provide posture optimization advice, such as adjusting the stance and conducting core stability training, to help the practitioner maintain an optimal stance.

[0068] The grip collection unit acquires the actual grip pressure of the exerciser and calculates the grip stability factor T hold It calculates the grip stability factor T and analyzes in real time whether the grip force and cue body angle meet the optimal standard. hold Too small or too large indicates a large deviation in grip force or cue body angle, which may affect the hitting trajectory and require directional training (e.g., increasing grip force and adjusting practice).

[0069] The inertial collection unit monitors the forward and backward tilt angle and body swing speed of the exerciser, and calculates the body balance factor T balance The system calculates the stance stability factor T stance , grip stability factor T hold and body balance factor T balance The weighted calculation is performed to obtain the posture standard index T pose This is used to comprehensively evaluate the correctness of the practitioner's posture. If Z<the preset threshold Z, the system will trigger a fourth alarm signal and generate a second training policy, which will increase 30s stance balance training (single-leg standing training, core muscle training), adjust the stance, ensure a more uniform center of gravity distribution, and conduct stability training to improve grip strength by 5%, thereby strengthening the ability to control the cue body.

[0070] The posture correction solution of the above embodiment facilitates accurate monitoring of stance, grip and body balance conditions compared to conventional training modes, and can quantitatively evaluate posture stability, allowing the practitioner to correct poor posture in a short time and improve hitting stability, thereby avoiding posture deviations caused by differences in the instructor's observation angle or the practitioner's own lack of perception.

[0071] The threshold value is set to facilitate comparison, and the threshold value depends on the amount of sample data and the number of bases set by a person skilled in the art for each group of sample data, as long as it does not affect the proportional relationship between the parameter and the quantified value.

[0072] The above formulas are all obtained by collecting a large amount of data and performing software simulations, and one formula close to the true value is selected, and the coefficients in the formulas are set by those skilled in the art according to the actual situation. What has been described above is a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Those skilled in the art may make equivalent substitutions or modifications based on the technical solutions and inventive ideas of the present application within the technical scope disclosed in the present application, and all of these shall be included in the protection scope of the present application.

Claims

1. A smart billiards training system, a three-dimensional billiard scene modeling module, which is used to acquire image data and three-dimensional data of the billiard table and balls using a depth camera and a laser radar, construct a three-dimensional model of the billiard table, and project the trajectories of the cue ball and target balls that a player hits each time onto the three-dimensional model; The path simulation module is a module that performs simulation analysis calculations using image processing technology to determine the optimal hitting angle θ of the cue ball when the trainee sets the target ball during training. ｈｉｔ and the initial velocity of the first cue ball V ｃ and collect the friction coefficient μ of the tablecloth, and calculate the initial velocity V of the first cue ball. ｃ Correction and conversion are performed to obtain the initial velocity of the second cue ball. [Equation 1] and used to obtain the impact strength F; The error analysis module is configured to calculate the training hitting strength F of the trainee. １ and training impact angle θ １ and respectively collect the optimal cue ball hitting angle θ ｈｉｔ and compares and calculates the impact strength F to obtain an impact strength error ΔF and an impact angle error Δθ, and if ΔF≠0, issues a first warning signal to notify the trainee that the impact strength is unacceptable, and if Δθ≠0, issues a second warning signal to notify the trainee that the impact angle is unacceptable; A trajectory smart monitoring module monitors the actual and predicted movement trajectories of the cue ball and the target ball after each swing by a practicer, establishes a trajectory movement data set, and performs analysis based on the trajectory movement data set to calculate a cue ball trajectory deviation coefficient T ｃｕｅ＿ｄｅｖ , target ball motion deviation factor T ｓｔａｒｇｅｔ＿ｄｅｖ and Cue ball abnormal jump factor T ｏｕｔ and obtain the hit trajectory accuracy coefficient T ａｃｃｕｒａｃｙ The pass threshold X is set in advance, and the impact trajectory accuracy coefficient T ａｃｃｕｒａｃｙ <used to trigger a third alarm signal if said pass threshold X is met, the path simulation module includes an image acquisition unit and a simulation calculation unit; the image collection unit collects the coordinates of the ith pocket on the billiard table, denoted as (x p,i , y p,i ), the current cue ball coordinates (x m , y m ), the target ball coordinates (x t , y t ) and the target ball radius R; collecting a current spin state and a motion state of the cue ball, the current spin state including no spin, top spin, and back spin; the motion states include stationary, rolling, and sliding states; When a practicer sets the target ball during training, the simulation calculation unit projects the cue ball and the target ball onto the three-dimensional model, and calculates the optimal cue ball hitting angle θ hit , the first cue ball initial velocity V c , the second cue ball initial velocity V c , and the cue ball optimal hitting angle θ hit , the first cue ball initial velocity V c , and the second cue ball initial velocity V c , using a collision physics algorithm. [Equation 2] and used to calculate and obtain the impact strength F, The method for obtaining the optimal cue ball hitting angle θ hit is as follows: In S11, if the target ball meets the requirement to enter the i-th pocket, the pocket entry vector of the target ball is [Equation 3] Calculate [Equation 4] In the formula, x p,i represents the x-axis coordinate of the i-th pocket in the Cartesian coordinate system along the long side of the billiard table; y p,i represents the y-axis coordinate of the i-th pocket in the Cartesian coordinate system along the short side of the billiard table, x t represents the x-axis coordinate of the target ball in the Cartesian coordinate system along the long side of the billiard table; y t represents the y-axis coordinate of the target ball in the Cartesian coordinate system along the short side direction of the billiard table, In S12, the i-th pocket approach direction angle θ (pocket, i) of the target ball is calculated; θ pocket, i = arctan(y p, i - y t ) / (x p, i - x t ), In S13, in order for the target ball to move toward the i-th pocket, the cue ball needs to contact the contact point of the target ball. First, the collision point coordinates (x c , y c ) of the target ball are calculated. x c =x t -R×cosθ pocket,i, y c =y t -R×sinθ pocket,i, In S14, the hitting direction of the cue ball [Equation 5] Calculate [Equation 6] In S15, the optimum cue ball hitting angle θ hit is calculated so that the cue ball hits the target ball accurately. θ hit = arctan(y c - y m ) / (x c - x m ), In the formula, x m represents the x-axis coordinate of the cue ball in the Cartesian coordinate system along the long side of the billiard table; The smart billiard training system is characterized in that y m represents the y-axis coordinate of the cue ball in the Cartesian coordinate system along the direction of the short side of the billiard table.

2. The first cue ball initial velocity V ｃ and the second cue ball initial velocity [Equation 7] The acquisition method is as follows: In S21, in order for the target ball to smoothly enter the pocket, it is necessary to calculate the hitting strength F of the cue ball. The hitting strength F of the cue ball is affected by the friction coefficient μ of the tablecloth. First, based on the law of conservation of momentum, the cue ball mass m ｃ and target ball mass m ｔ and collect the cue ball mass m ｃ and the target ball mass m ｔ are equal, the first cue ball initial velocity V ｃ teeth, m ｃ ×V ｃ = m ｔ ×V ｔ The formula is satisfied, In the formula, V ｔ is the target ball velocity, and the cue ball mass m ｃ and the target ball mass m ｔ is equal, when the cue ball comes to a complete stop after colliding with the target ball, the velocity of the cue ball before the collision is equal to the velocity of the target ball after the collision, and the target ball moves along the collision direction; The cue ball mass m ｃ ≠Target ball mass m ｔ When the first cue ball initial velocity V ｃ teeth, [Equation 8] [Equation 9] The formula is satisfied, During the ceremony, [Equation 10] represents the velocity of the cue ball after impact, In S22, the target ball speed V ｔ is attenuated by the friction coefficient μ of the tablecloth and the movement distance, and the target ball velocity V is calculated by the following formula: ｔ Correct the target ball speed [0011] Get [0012] In the formula, μ represents the coefficient of friction of the tablecloth, and when the tablecloth is made of a nylon blend material, μ is 0.03 to 0.05; when the tablecloth is a sports tablecloth with a wool content of 85 to 90%, μ is 0.13 to 0.025; when the tablecloth contains American 8-pile coarse wool and the wool content is between 50% and 80%, μ is 0.03 to 0.04; when the tablecloth contains 91% to 100% fine wool, μ is 0.008 to 0.015; when the tablecloth is a casino 9-pile wool tablecloth, μ is 0.01 to 0.02; and ｉ is the distance from the target ball to the i-th pocket, and e －μ×ｄｉ represents a velocity attenuation factor, and the longer the movement distance, the greater the attenuation of the target ball velocity; In S23, the corrected target ball velocity [0013] Based on this, the first cue ball initial velocity V ｃ is synchronously corrected, and the second cue ball initial velocity is calculated by the following formula: [0014] and calculate the cue ball mass m ｃ and the target ball mass m ｔ is equal to, the second cue ball initial velocity [Equation 15] teeth, [0016] The formula is satisfied, The cue ball mass m ｃ ≠Target ball mass m ｔ When the second cue ball initial velocity is [Equation 17] teeth, [Equation 18] 2. The smart billiard training system according to claim 1, wherein the above condition is satisfied.

3. The method for acquiring the impact strength F is as follows: the second cue ball initial velocity [Equation 19] Based on this, according to the impulse-momentum theorem and Newton's second law, the impact strength F is obtained by the following formula: [Equation 20] 3. The smart billiard training system of claim 2, wherein t represents the time when the tip of the cue ball contacts the cue ball, and is expressed in seconds (s) and is set to 0.001 s to 0.005 s.

4. The error analysis module calculates the training hitting strength F of the practicer by using an angle sensor and a force sensor attached to the cue. １ and the training impact angle θ １ and respectively collect the optimal cue ball hitting angle θ ｈｉｔ and compared with the impact strength F, and used to calculate the impact strength error ΔF and the impact angle error Δθ using the following formula: ΔF=F １ -F、 Δθ=θ １ -θ ｈｉｔ 、 If ΔF≠0, the first warning signal is transmitted to notify the practicer that the hitting strength is unacceptable; if ΔF>0, the first warning signal is transmitted to notify the practicer that the hitting strength is excessive and the practicer is advised to reduce the hitting strength; if ΔF<0, the first warning signal is transmitted to notify the practicer that the hitting strength is insufficient and the practicer is advised to increase the hitting strength; The smart billiards training system of claim 3, wherein if Δθ≠0, the second warning signal is sent to notify the practicer that the impact angle is unacceptable, if Δθ>0, the practicer is notified that the tip is deviated to the left when impacted, and if Δθ<0, the practicer is notified that the tip is deviated to the right when impacted.

5. The trajectory smart monitoring module includes a second collecting unit and a second calculating unit; the second collecting unit is used to collect the actual movement trajectories and the predicted movement trajectories of the cue ball and the target ball after each swing by a practicer, and establish the trajectory movement data set; The second calculation unit performs advanced calculation analysis based on the trajectory motion data set to calculate the cue ball trajectory deviation coefficient T by the following formula: ｃｕｅ＿ｄｅｖ , the target ball motion deviation factor T ｓｔａｒｇｅｔ＿ｄｅｖ and the cue ball abnormal jump factor T ｏｕｔ is used to calculate T ｃｕｅ＿ｄｅｖ =a １ ×|θ ｃｕｒ＿ｄｅｖ |+a ２ ×|d ｃｕｒ＿ｄｅｖ |+a ３ ×|v ｃｕｒ＿ｄｅｖ |+a ４ ×|w ｃｕｒ＿ｄｅｖ |、 In the formula, θ ｃｕｒ＿ｄｅｖ represents the trajectory deviation angle difference value of the cue ball, and d ｃｕｒ＿ｄｅｖ represents the lateral displacement that the cue ball experiences during linear motion, and v ｃｕｒ＿ｄｅｖ represents the deviation between the actual velocity of the cue ball and the theoretically calculated velocity, and w ｃｕｒ＿ｄｅｖ represents the forward spin or back spin state deviation value of the cue ball, and a １ , a ２ , a ３ and a ４ represents the weighting coefficient, T ｓｔａｒｇｅｔ＿ｄｅｖ =a ５ ×|θ ｔａｒｇｅｔ＿ｄｅｖ |+a ６ ×|d ｔａｒｇｅｔ＿ｄｅｖ |+a ７ ×|v ｓｔｒｇｅｔ＿ｄｅｖ |+a ８ ×|w ｔａｒｇｅｔ＿ｄｅｖ |、 In the formula, θ ｔａｒｇｅｔ＿ｄｅｖ represents the angle between the actual driving direction of the target ball and the desired direction, and d ｔａｒｇｅｔ＿ｄｅｖ represents the distance error between the final stopping position of the target ball and the target point, and v ｓｔｒｇｅｔ＿ｄｅｖ represents the error between the actual velocity of the target ball and the desired velocity, and w ｔａｒｇｅｔ＿ｄｅｖ represents the deviation between the rotation direction of the target ball and the rotation direction of the target ball in the predicted motion trajectory, and a ５ , a ６ , a ７ and a ８ represents the weighting coefficient, T ｏｕｔ =a ９ ×|d ｃｕｅ＿ｍｉｎ |+a １０ ×|h ｃｕｅ |+a １１ ×|w ｃｕｅ |、 In the formula, d ｃｕｅ＿ｍｉｎ represents the closest distance between the cue ball and the boundary of the billiard table, and h ｃｕｅ represents the jump height of the cue ball, and |w ｃｕｅ | represents the deviation value of the non-spin state of the cue ball, and a ９ , a １０ and a １１ 2. The smart billiard training system of claim 1, wherein: represents a weighting coefficient.

6. The trajectory smart monitoring module further includes a first association unit and a first alarm unit; The first association unit calculates the cue ball trajectory deviation coefficient T ｃｕｅ＿ｄｅｖ , the target ball motion deviation factor T ｓｔａｒｇｅｔ＿ｄｅｖ and the cue ball abnormal jump factor T ｏｕｔ After performing non-dimensional processing on the impact trajectory accuracy coefficient T ａｃｃｕｒａｃｙ is used to calculate T ａｃｃｕｒａｃｙ =1-(b) １ ×T ｃｕｅ＿ｄｅｖ +c １ ×T ｓｔａｒｇｅｔ＿ｄｅｖ +d １ ×T ｏｕｔ ) In the formula, β １ , γ １ and δ １ are the cue ball trajectory deviation coefficients T ｃｕｅ＿ｄｅｖ , the target ball motion deviation factor T ｓｔａｒｇｅｔ＿ｄｅｖ and the cue ball abnormal jump factor T ｏｕｔ is the weighting factor of The first alarm unit presets the passing threshold X and the impact trajectory accuracy coefficient T ａｃｃｕｒａｃｙ is compared with the predetermined pass threshold X, and the impact trajectory accuracy coefficient T ａｃｃｕｒａｃｙ If the passing threshold value is X, it indicates that the trainee's billiards training is unacceptable, trigger the third warning signal, and generate a first policy that includes adjusting the current training plan to increase the tip angle training task by 10% to 20%, including backspin training, topspin training, and sidespin training; The impact trajectory accuracy coefficient T ａｃｃｕｒａｃｙ The smart billiards training system of claim 5, wherein if the pass threshold X is greater than or equal to the pass threshold X, it indicates that the practitioner's billiards training has passed, and is used to continue the current training plan.

7. The stance stability factor T is calculated by analyzing the stance, grip method, and center of gravity distribution of the trainee. ｓｔａｎｃｅ , grip stability factor T ｈｏｌｄ and body balance factor T ｂａｌａｎｃｅ and a posture correction module for analytically calculating the attitude correction module includes a pressure collection unit, a grip collection unit, and an inertial collection unit; The pressure collecting unit measures the left foot pressure Fσ of the exerciser by a pressure sensor attached to the ground pad. ｌｅｆｔ and right foot pressure Fσ ｒｉｇｈｔ is used to detect The grip collection unit measures the actual grip pressure P by a grip pressure sensor attached to the grip position of the cue. ｇｅｉｐ is used to detect The inertial collection unit detects the center of gravity (X ｃｇ , Y ｃｇ ), longitudinal inclination angle A ｑｈｑｘ and the body swing velocity V is measured by an acceleration sensor attached to the body of the exerciser. ｂｏｂｙ 4. The smart billiard training system according to claim 3, wherein the smart billiard training system is used to monitor the following in real time:

8. the posture correction module further includes a first analysis unit, a second analysis unit, and a third analysis unit; The first analysis unit calculates the center of gravity position (X ｃｇ , Y ｃｇ ), the left foot pressure Fσ of the exerciser ｌｅｆｔ and right foot pressure Fσ ｒｉｇｈｔ After extracting and non-dimensionalizing the stance stability factor T ｓｔａｎｃｅ is used to calculate T ｓｔａｎｃｅ =1-|Fσ ｌｅｆｔ -Fσ ｒｉｇｈｔ | / F ｔｏｔａｌ -(|X ｃｇ -X ｍｉｄ |+|Y ｃｇ -Y ｍｉｄ |) / D ｍａｘ 、 In the formula, F ｔｏｔａｌ represents the total body force, and F ｔｏｔａｌ = Fσ ｌｅｆｔ +Fσ ｒｉｇｈｔ and D ｍａｘ represents the maximum allowable deviation distance of the stance center of gravity, and 0<T ｓｔａｎｃｅ ≦1, and X ｃｇ represents the horizontal coordinate of the current center of gravity of the trainee, and Y ｃｇ represents the front-back coordinate of the current center of gravity of the trainee, and X ｍｉｄ represents the horizontal coordinate of the center of gravity in the standard stance, and Y ｍｉｄ represents the front-to-back coordinate of the center of gravity in the standard stance, X ｃｇ >X ｍｉｄ If it is, it indicates that the practitioner's current center of gravity is biased to the right, X ｃｇ <X ｍｉｄ If it is, it indicates that the practitioner's current center of gravity is biased to the left, Y ｃｇ >Y ｍｉｄ If this is the case, it indicates that the practitioner's current center of gravity is biased forward, causing forward leaning. Y ｃｇ <Y ｍｉｄ If this is the case, it indicates that the practitioner's center of gravity is currently biased backward, causing backbend. The second analysis unit calculates the actual grip pressure P ｇｅｉｐ The grip stability factor T is extracted using the following formula: ｈｏｌｄ is used to calculate T ｈｏｌｄ =1-|P ｇｅｉｐ -P ｏｐｔ | / P ｍａｘ -θ ｄｅｖ / θ ｍａｘ 、 In the formula, P ｇｅｉｐ is the actual grip pressure, and P ｏｐｔ represents the optimal grip pressure, and P ｍａｘ represents the maximum allowable grip pressure, and θ ｄｅｖ represents the cue body angle deviation value when gripping, and θ ｍａｘ represents the maximum allowable grip angle deviation value, and T ｈｏｌｄ The closer to 1, the more stable the grip of the practitioner is. The third analysis unit calculates the longitudinal inclination angle A ｑｈｑｘ and the body swing speed V ｂｏｂｙ After extracting and non-dimensionalizing the body balance factor T ｂａｌａｎｃｅ is used to calculate T ｂａｌａｎｃｅ =1-A ｑｈｑｘ / A ｍａｘ -V ｂｏｂｙ / V ｍａｘ 、 In the formula, A ｍａｘ represents the maximum allowable tilt angle, and V ｍａｘ 8. The smart billiard training system according to claim 7, wherein: represents the maximum permissible swing speed.

9. the attitude correction module further includes a second association unit and a second alarm unit; The second association unit calculates the stance stability factor T ｓｔａｎｃｅ , the grip stability factor T ｈｏｌｄ and the body balance factor T ｂａｌａｎｃｅ After extracting and non-dimensionalizing the posture standard index T ｐｏｓｅ is used to obtain T ａｃｃｕｒａｃｙ =b １ ×T ｓｔａｎｃｅ +c ２ ×T ｈｏｌｄ +d ２ ×T ｂａｌａｎｃｅ 、 In the formula, β ２ , γ ２ and δ ２ are the stance stability factors T ｓｔａｎｃｅ , the grip stability factor T ｈｏｌｄ and the body balance factor T ｂａｌａｎｃｅ is the weighting factor of The second alarm unit presets a posture normalization threshold Z, and the posture normalization index T ｐｏｓｅ is compared with the posture standardized threshold Z to determine whether the posture of the trainee is acceptable, and includes: The posture standard index T ｐｏｓｅ If the posture standardization threshold value is Z, it indicates that the posture of the practitioner during billiards training is in an unacceptable state, a fourth alarm signal is triggered, and a second policy is generated to adjust the current training plan, adjust the stance, and add 30 seconds of stance balance training before each training, the stance balance training including one-legged standing training and core muscle training, and adjust the stance, so that the center of gravity distribution is X ｍｉｄ and Y ｍｉｄ Approach with 5% increase in grip strength stability training, The posture standard index T ｐｏｓｅ 9. The smart billiards training system of claim 8, wherein if the posture standardization threshold Z is greater than or equal to the posture standardization threshold Z, the posture of the practitioner during billiards training is deemed acceptable, and the current training plan is continued.

Images