CN116611737A

CN116611737A - Intelligent table tennis ball examination grade assessment and lifting method and system

Info

Publication number: CN116611737A
Application number: CN202310601865.2A
Authority: CN
Inventors: 刘文明
Original assignee: Zhejiang University ZJU
Current assignee: Zhejiang University ZJU
Priority date: 2023-05-26
Filing date: 2023-05-26
Publication date: 2023-08-18

Abstract

The invention relates to a table tennis intelligent examination grade evaluation and lifting method and a system, wherein the method comprises the following steps of (1) performing real-time monitoring and capturing actions of a test taker around a service robot; (2) detecting the position of the table tennis ball and the player's hand; (3) Setting a test grade item and corresponding service parameters and scoring standards; (4) performing data analysis; and (5) personalized training recommendation and monitoring. The system comprises an intelligent service robot, a monitoring camera and a sensor. The invention relates to an automatic examination grade assessment system for table tennis fans. By combining a programmable intelligent service robot, a real-time monitoring camera, analysis software and a cloud data analysis platform, accurate evaluation of the table tennis skill level of a user is realized. In addition, the invention can provide personalized training plans for users, and helps students to improve the table tennis skills.

Description

Intelligent table tennis ball examination grade assessment and lifting method and system

Technical Field

The invention relates to the field of artificial intelligence, in particular to a table tennis intelligent examination grade evaluation and lifting method and system.

Background

With the rapid development of technology, intelligent devices are increasingly used in various fields. In the sports field, intelligent device can help improving athlete's training effect, improve the sight of match etc.. In the existing table tennis teaching, no perfect technology and product capable of accurately acquiring the training condition of a student in real time by using modern scientific and technical means and guiding the basic learning and training of the student in table tennis scientifically and effectively exist. At present, a complete table tennis examination grade system of an intelligent service robot is urgently needed, and a more scientific, accurate and intelligent level assessment scheme is provided for table tennis fans.

Disclosure of Invention

The embodiment of the invention provides a method and a system for intelligent table tennis ball grade assessment and lifting. The following presents a simplified summary in order to provide a basic understanding of some aspects of the disclosed embodiments. This summary is not an extensive overview and is intended to neither identify key/critical elements nor delineate the scope of such embodiments. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is presented later.

The embodiment of the invention provides an intelligent table tennis ball examination grade assessment and lifting method, which is characterized by comprising the following steps:

(1) Performing real-time monitoring and capturing actions of a test taker around the service robot;

(2) Detecting the positions of the table tennis ball and hands of a player;

(3) Setting a test grade item and corresponding service parameters and scoring standards;

(4) Performing data analysis;

(5) And (5) personalized training recommendation and monitoring.

Preferably, the step (1) includes time stamping shot video and sensor data captured by the high speed camera, and training the model to evaluate table tennis speed and spin using the vision and measurement data as the true value.

Preferably, the step (2) judges whether the player hits the ball successfully or not by analyzing the contact between the ball track and the racket, and the detection includes accuracy detection and batting continuity detection.

Further, the accuracy detection includes: setting a target area, extracting a track of a ball, determining a falling point of the ball, judging whether the point is positioned in the target area according to the falling point coordinates (x, y) of the ball, and calculating an accuracy score;

setting a target area: setting one or more target areas on the table tennis table according to the examination grade requirements;

trajectory of extraction sphere: detecting the position of the table tennis ball in continuous frames in real time by using an object detection model;

determining the ball drop point: finding out the moment of contact between the ball and the table tennis table by calculating the height change deltay of the ball in the continuous frames and the distance d from the table tennis table;

judging whether the point is positioned in the target area according to the falling point coordinates (x, y) of the ball: judging whether the falling point is in the rectangular target area by adopting the following formula:

is_accurate=x1 < =x2 and y1< =y < = y2, and if is_accurate is True, the shot is considered to be accurate; counting the accurate ball hitting times of a player in the whole test item process;

calculating an accuracy score: dividing the number of accurate hits by the total number of successful hits by the player to obtain an accuracy score: accuracy=accurateshits/success_hits.

Further, the shot continuity detection includes: extracting the batting moment, calculating the continuous batting time interval and judging the continuous batting;

extracting the batting moment: detecting the position of the table tennis ball in real time by using an object detection model;

calculating a continuous batting time interval: in the list success_hit_time, calculating the time interval deltat of every two consecutive shots;

judging continuous batting: traversing the time_intervals list to find out the time interval of the continuous shot.

Preferably, the step (3) includes a forward pushing and stopping service parameter, a backward pushing and stopping service parameter, a forward tapping service parameter, a backward tapping service parameter, a forward quick tapping service parameter, a backward quick tapping service parameter, a receiving service parameter, a pulling service parameter and corresponding scoring criteria;

forward push stop serve parameters: low speed, downspin, lower camber line; scoring criteria: scoring according to shot stability, accuracy and spin control;

backhand push ball serving parameters: low speed, downspin, lower camber line; scoring criteria: scoring according to shot stability, accuracy and spin control;

hand tapping service parameters: medium speed, upward rotation, medium camber line; scoring criteria: scoring according to shot stability, accuracy, speed and spin control;

backhand tapping ball serving parameters: medium speed, upward rotation, medium camber line; scoring criteria: scoring according to shot stability, accuracy, speed and spin control;

parameters of the normal hand quick attack service: high speed, light upward rotation or no rotation, low camber line; scoring criteria: scoring according to the speed, accuracy and continuity of the shot;

reverse hand rapid tapping parameters: high speed, light upward rotation or no rotation, low camber line; scoring criteria: scoring according to the speed, accuracy and continuity of the shot;

ball receiving and serving parameters: ball serving combining different speeds, rotations and arcs; scoring criteria: scoring according to stability, accuracy and strain capacity of the ball receiving and delivering;

ball-pulling service parameters: medium speed, lower spin, lower camber line; scoring criteria: scoring is based on shot stability, accuracy, speed, and spin control.

Further, the scoring criteria include hit rate, speed control, spin control, accuracy, and continuity;

setting a total score for each item, distributing weights according to the factors, calculating the score, and recording the score into a corresponding database after each item is finished;

wherein the hit rate is the proportion of successful hit of the ball by the player; speed control is the ability of a player to control the speed of a ball hit; spin control is the ability of a player to control the spin of a ball; accuracy is the ability of a player to hit a ball to a designated area; the continuity is the number of consecutive shots played by the player.

Preferably, the step (4) includes collecting data, storing personal information, class score, scoring details and test data of the player in a database or a file, and performing data analysis according to the following indexes:

average division: average score obtained by player in each grade item;

highest score: the player obtains the highest score in each grade item;

lowest score: the player obtains the lowest score in each grade item;

variance: the player obtains the discrete degree of the score in each grade item;

quasi-difference: the player obtains the degree of dispersion of the score in each grade item.

Preferably, the step (5) includes

(5-1) calculating scores of players in different examination items and making training plans:

the scores of the players in the different class projects were calculated using the following formula:

wherein actual_score is the actual score of the player in the test item, and max_score is the full score of the test item; through calculating the score, the skill level of the player in different examination level projects is known, and the weaknesses and advantages of the player are found out;

according to the weaknesses and advantages of the player, a corresponding training plan is formulated; the training intensity was calculated using the following formula:

the target_score is a target score, the actual_score is an actual score, and the min_score is a lowest pass score of the test grade item; the training difficulty is adjusted by calculating the training intensity;

(5-2) recommending training resources:

providing training resources for the players according to the skill levels of the players; recommending relevant training resources for the player according to the historical training record of the player by using a collaborative filtering algorithm or a content recommendation algorithm;

the recommendation level of the training resources is calculated using the following formula:

where n is the number of training resources, r _i Scoring the ith training resource for the player, s _i Similarity for the ith training resource;

(5-3) training progress monitoring:

the player training was monitored and the training progress of the player was calculated using the following formula:

wherein actual score is the actual score of the player in the test item and initial score is the initial score of the player in the test item at the beginning of training.

The invention also provides a table tennis intelligent examination grade evaluation and lifting system, which is improved in that the system comprises an intelligent service robot, a monitoring camera and a sensor;

the intelligent service robot is used for adjusting the service speed, rotation and angle and is in communication connection with external equipment;

the monitoring camera is used for monitoring and capturing the actions of the examinee in real time;

the sensor comprises a 9-axis gyroscope and is used for acquiring three-dimensional acceleration, angular velocity and angle of a moving object.

The technical scheme provided by the embodiment of the invention can have the following beneficial effects:

the invention discloses an intelligent table tennis examination method and a system thereof, which are an automatic examination evaluation system aiming at table tennis fans. By combining a programmable intelligent service robot, a real-time monitoring camera, analysis software and a cloud data analysis platform, accurate evaluation of the table tennis skill level of a user is realized. In addition, the invention can provide personalized training plans for users, and helps students to improve the table tennis skills.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as claimed.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.

FIG. 1 is a flow chart of a method for intelligent class assessment and promotion of table tennis balls according to an exemplary embodiment.

FIG. 2 is a schematic diagram of a table tennis intelligent class assessment and promotion system according to an exemplary embodiment;

wherein, A-sensor, B-surveillance camera head, C-intelligent service robot.

Detailed Description

The following description and the drawings sufficiently illustrate specific embodiments of the invention to enable those skilled in the art to practice them. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the sequence of operations may vary. Portions and features of some embodiments may be included in, or substituted for, those of others. The scope of embodiments of the invention encompasses the full ambit of the claims, as well as all available equivalents of the claims. Embodiments may be referred to herein, individually or collectively, by the term "invention" merely for convenience and without intending to voluntarily limit the scope of this application to any single invention or inventive concept if more than one is in fact disclosed. Relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed. Various embodiments are described herein in a progressive manner, each embodiment focusing on differences from other embodiments, and identical and similar parts between the various embodiments are sufficient to be seen with each other. The structures, products and the like disclosed in the embodiments correspond to the parts disclosed in the embodiments, so that the description is relatively simple, and the relevant parts refer to the description of the method parts.

The invention is further described below with reference to the accompanying drawings and examples:

as shown in FIG. 1, the invention provides a table tennis intelligent examination grade evaluation and lifting method, which comprises the following steps:

to ensure that valid shots are detected and extracted, shot attributes are identified, shot speed is estimated, and shot spin is estimated based on raw sensor data, the present invention uses additional auxiliary devices for data annotation.

Specifically, 6 marks (i.e., up, down, left, right, back, and front) are drawn on each ping-pong ball, and a sensor is embedded in a special ping-pong ball to minimally affect the weight and performance of the ball, and the data acquired through WiFi transmission using Arduino can reach a maximum frequency of 200Hz. In the implementation, the frequency is set to 20Hz, as 20Hz is sufficient to collect detailed kinematic features of the tester. The shot video and sensor data captured by the high speed camera are time stamped to facilitate synchronization, and a model is trained to estimate the speed and rotation of the ball using a clear visual and measured data as the actual value, such as a sudden acceleration change to manually or automatically align the time stamps of the video and sensor data.

Accelerometer data is extracted from the raw sensor and integrated to calculate ball velocity, as the data consists of multi-dimensional discrete signals, detecting a ball strike is equal to the peak Pi in the detected signal. The present invention uses the acceleration signal of the racket to detect the peak because there are distinct peaks on all axes of the acceleration signal. The detection process comprises two steps: energy calculation and peak detection. First, the energy of the acceleration signal at t is calculated as follows:

wherein r (t) _x ,r(t) _y ,r(t) _z For acceleration signal values on the z-axis at time x, y. To obtain a more pronounced peak, a butt Wo Sigao pass filter is used to further amplify the peak,

for all detected peaks of p= { P1, …, pn }, all hits were noted as s= { S1, …, sn },

and aligning the video frames with the sensor data, and ensuring corresponding speed and angular speed labels of each frame of video. Cleaning the data and removing abnormal values or inaccurate labels. Video frames and marked tags are read according to 7:2:1 is divided into a training set, a validation set and a test set. And constructing an LSTM neural network, adding a full connection layer to an output layer, and returning to corresponding ball speed and angular speed.

In addition to ball speed and angular velocity detection, there is a need to detect the player's hit rate, accuracy and continuity.

(2) Detecting the position of the table tennis ball and the hands of the player, wherein the detection comprises accuracy detection and batting continuity detection;

specifically, an object detection model yolov5 is used for hit rate, and a table tennis detection model is trained. In the video captured by the camera, the position of the table tennis ball is detected in real time. Meanwhile, the hand position of the player is detected by using a key point detection model (openPose). By analyzing the ball trajectory in contact with the racket, it is determined whether the ball was successfully hit by the player. Then, the hit rate is calculated: hit_rate=success_hits/total_hits_attempted;

and (3) accuracy detection:

the accuracy detection includes: setting a target area, extracting a track of a ball, determining a falling point of the ball, judging whether the point is positioned in the target area according to the falling point coordinates (x, y) of the ball, and calculating an accuracy score;

1. setting a target area: on a table tennis table, one or more target areas are set according to the class requirements. For example, a rectangular target area is represented as (x 1, y1, x2, y 2), where (x 1, y 1) is the coordinates of the upper left corner of the rectangle and (x 2, y 2) is the coordinates of the lower right corner of the rectangle.

2. Trajectory of extraction sphere: the position of the table tennis ball in successive frames is detected in real time using an object detection model. The positions (x, y) of the balls in each frame are stored in a list, expressed as ball_positions= [ (x 1, y 1), (x 2, y 2) ].

3. Determining the ball drop point: the moment when the ball contacts the table tennis table is found. By calculating the ball height change deltay and the distance d from the table tennis table in successive frames. When the ball changes in height towards zero (|Δy| < threshold_y) and is very close to the table tennis table (d < threshold_d), the ball may be considered to have been in contact with the playing surface. Appropriate thresholds threshold_y and threshold_d are set, threshold_y=5 pixels, and threshold_d=10 pixels.

4. And judging whether the point is positioned in the target area according to the falling point coordinates (x, y) of the ball. Judging whether the falling point is in the rectangular target area or not by using the following formula:

is_accurate=x1 < =x < =x2 and y1< =y < = y2, and if is_accurate is True, the shot is considered to be accurate. And counting the accurate ball hitting times of the player in the whole test item process.

5. Calculating an accuracy score:

dividing the number of accurate hits by the total number of successful hits by the player to obtain an accuracy score: accuracy=accurateshits/success_hits.

Ball striking continuity detection:

the shot continuity detection includes: extracting the batting moment, calculating the continuous batting time interval and judging the continuous batting;

1. extracting the batting moment: the position of the table tennis ball was detected in real time using an object detection model (yolov 5). The position of the player's hand is detected in conjunction with the keypoint detection model (openPose). By analyzing the ball trajectory in contact with the racket, it can be determined whether the ball was successfully hit by the player. The time stamp t of a successful shot is recorded, expressed as success_hit_times= [ t1, t2, t3 > ].

2. Calculating a continuous batting time interval: in the list success_hit_time samples, the time interval Δt, time_intervals= [ t2-t1, t3-t2, t4-t3, ] for every two consecutive shots is calculated

3. Judging continuous batting:

traversing the time_intervals list to find out the time interval of the continuous shot. A consecutive shot is considered to be a consecutive shot when the consecutive shot time interval is less than a threshold threshold_t (e.g., threshold_t=1 second). The current number of consecutive shots is recorded with a counter and the maximum number of consecutive shots so far is recorded with a variable max_consecutive_hits. Max_continuity_hits are taken as the continuity score.

The trained model is converted into a Tensorrt format, the converted model is loaded and deployed on NVIDIAJetson Xavier NX, and is directly connected with a high-speed camera and a ball dispenser, video frames are directly read from the high-speed camera by using opencv, and the video frames are input into the model through preprocessing (zooming, cutting and the like) to carry out real-time ball speed, angular speed, rotating ball, hit rate, accuracy and batting continuity.

(3) Setting a test grade item and corresponding service parameters and scoring standards

The method comprises a forward-hand push ball-blocking serving parameter, a reverse-hand push ball-blocking serving parameter, a forward-hand ball-tapping serving parameter, a reverse-hand ball-tapping serving parameter, a forward-hand quick ball-tapping serving parameter, a reverse-hand quick ball-tapping serving parameter, a ball-receiving serving parameter, a ball-pulling serving parameter and corresponding scoring standards.

1. Forward Push (forward Push) serve parameters: low speed, downspin, lower camber line. Scoring criteria: scoring is based on shot stability, accuracy and spin control.

2. Backhand Push (Backhand Push) serve parameters: low speed, downspin, lower camber line. Scoring criteria: scoring is based on shot stability, accuracy and spin control.

3. Forward Drive service parameters: medium speed, upward spin, medium camber line. Scoring criteria: scoring is based on shot stability, accuracy, speed, and spin control.

4. Backhand Drive (service) parameters: medium speed, upward spin, medium camber line. Scoring criteria: scoring is based on shot stability, accuracy, speed, and spin control.

5. Forward hand quick attack (Forehand Smash) service parameters: high speed, light upward rotation or no rotation, low camber line. Scoring criteria: scoring is based on the speed, accuracy, and continuity of the shots.

6. Backhand Smash (Backhand Smash) service parameters: high speed, light upward rotation or no rotation, low camber line. Scoring criteria: scoring is based on the speed, accuracy, and continuity of the shots.

7. Service Return (service Return) service parameters: ball serving with different speeds, spin and arcs are combined. Scoring criteria: scoring is performed according to stability, accuracy and strain capacity of the ball receiving and delivering.

8. Ball pull (Loop) service parameters: medium speed, down spin, lower camber line. Scoring criteria: scoring is based on shot stability, accuracy, speed, and spin control.

And integrating the ball speed model and the service robot control, and packaging the service robot control function into a class. In this way, these functions can be called in the main program to control the behavior of the service robot. Specifically, a class named TableTennisLauncher is created and added with the following method: (1) connect (): connecting a service robot, (2) disconnet (): disconnecting from the tee machine, (3) start (): starting serve, (4) stop: stopping serving. (5) set_ball_parameters (speed, spin, angle, frequency): setting a service parameter. (6) adaptive_parameter: the difficulty can be adaptively adjusted according to the performance of the player. (optional)

Setting a main program, instantiating a TableTennisLauncher class, connecting a service robot, and writing a function for each test class item, wherein the function sets the service robot number according to the item requirement.

In the normal mode, the main program sequentially calls each class item function, and simultaneously starts a ball speed and angular speed identification model to detect the performance of a player in real time. At the end of each project, the service is stopped and the player's performance is recorded.

The adaptive_parameter function is called in the adaptive mode, and the adaptive_parameter function firstly reads the performance of the player in the last item and matches corresponding service parameters.

The invention also designs a scoring system to measure the performance of the player in different examination items and realize an automatic scoring function.

A scoring standard is designed for each class item, the scoring taking into account the following factors: hit rate: the proportion of successful ball hits by the player. And (3) speed control: the ability of the player to control the speed of the ball. Rotation control: the player controls the ability of the ball to spin. Accuracy: the ability of a player to hit a ball to a designated area. Continuity: number of consecutive shots by the player. A total score (e.g., 100 points) is set for each item and weights are assigned based on the above factors. For example, the scoring criteria for a positive manual shift item may be: hit rate (40%), speed control (20%), spin control (20%), accuracy (10%), continuity (10%). After the score is calculated, the score is recorded into a corresponding database after each item is finished.

(4) Performing data analysis

Data collection is performed and personal information, class performance, scoring details (e.g., hit rate, accuracy, speed control, spin control, and continuity score) and test data (e.g., ball trajectory, racket position, ball striking time stamp, etc.) of the player are saved to a database or file. Data storage is performed using a relational database (MySQL).

The camera records the video of the whole examination process and stores the video in cloud storage service (AWS S3).

The data analysis was performed according to the following criteria, which can be calculated: average division: average score obtained by player in each class item. Highest score: the player obtains the highest score in each class item. Lowest score: the player obtains the lowest score in each class item. Variance: the player obtains the degree of dispersion of the score in each grade item. Quasi-difference: the player obtains the degree of dispersion of the score in each grade item.

(5) Personalized training recommendation and monitoring and making personalized training plan

1. The personalized training program is formulated to take the weaknesses of the players and the places where the improvement is needed into consideration, and the training time and the training target of the players. The scores of players in different class projects can be calculated using the following formula:

wherein actual_score is the actual score of the player in the test item, and max_score is the full score of the test item. By calculating the score, the skill level of the player in different examination level projects can be known, and the weaknesses and advantages of the player can be found out.

And (5) according to the weaknesses and advantages of the player, making a corresponding training plan. For example, if the player performs poorly in speed control, the exercise of speed control may be enhanced. The training intensity was calculated using the following formula:

the target_score is the target score, the actual_score is the actual score, and the min_score is the lowest pass score of the test item. The training difficulty can be adjusted by calculating the training intensity, so that the player gradually approaches the target score.

2. Recommending training resources requires providing the player with training resources appropriate for the player based on his skill level. Collaborative filtering algorithms or content recommendation algorithms may be used to recommend relevant training resources to a player based on the player's historical training records.

where n is the number of training resources, r _i Scoring the ith training resource for the player, s _i Similarity for the ith training resource. According to the recommendation degree, training resources suitable for the players can be recommended for the players.

3. Training progress monitoring, in order to know the training progress of a player in time, performance and progress of the player in training need to be monitored. The training progress of the player was calculated using the following formula:

As shown in FIG. 2, the invention also provides a table tennis intelligent examination grade evaluation and lifting system, which comprises an intelligent service robot, a monitoring camera and a sensor;

(1) Selecting an applicable intelligent service robot: the intelligent service robot selects ButterflyAmicus Prime with programmable functions, has the functions of adjusting service speed, rotation, angle and the like, and can communicate with external equipment.

(2) Monitoring camera type selection and deployment: the invention selects the high-definition camera, and has higher image processing speed. The cameras need to be deployed around the tee box to achieve omni-directional capture of the movements of the test taker.

(3) And (3) data transmission: the invention uses Wi-Fi or Bluetooth module to realize wireless communication between the intelligent service robot and the camera as well as the data analysis equipment.

Wherein, the surveillance camera head is arranged around the service robot, catches the action of examinee in real time. The camera adopts high definition resolution, and can clearly capture the hand movements, body coordination and the like of the examinee. The cameras work simultaneously, three-dimensional capturing of the actions of the examinee is achieved, and accuracy of the evaluation result is guaranteed.

The sensors used WT901WIFI 3 sensors developed by WitMotion 4 to collect movement data during player testing. Each device was portable enough (51 mm. Times.36 mm. Times.15 mm, weight: 20 g) to minimize the impact of the device on athlete performance. Each sensor comprises a 9-axis gyroscope, acquires the three-dimensional acceleration, angular velocity and angle of a moving object, can be internally provided with a battery, can continuously work for two hours, and ensures the test time.

The foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and their practical application to enable others skilled in the art to make or utilize the invention in various exemplary embodiments and with various alternatives and modifications. It is intended that the scope of the invention be defined by the claims appended hereto and their equivalents.

It is to be understood that the above embodiments are merely illustrative of the application of the principles of the present invention, but not in limitation thereof. Various modifications and improvements may be made by those skilled in the art without departing from the spirit and substance of the invention, and are also considered to be within the scope of the invention.

Claims

1. The intelligent table tennis test grade assessment and lifting method is characterized by comprising the following steps of:

(2) Detecting the positions of the table tennis ball and hands of a player;

(4) Performing data analysis;

(5) And (5) personalized training recommendation and monitoring.

2. The intelligent class assessment and promotion method of table tennis according to claim 1, wherein step (1) comprises adding a time stamp to the shot video and sensor data captured by the high speed camera, and training the model to assess table tennis speed and spin using the vision and measurement data as the true value.

3. The intelligent table tennis class assessment and lifting method according to claim 1, wherein the step (2) is to analyze the contact between the ball track and the racket to determine whether the ball is hit successfully by the player, and the detection includes accuracy detection and batting continuity detection.

4. The method for intelligent class assessment and promotion of table tennis according to claim 3, wherein the accuracy detection comprises: setting a target area, extracting a track of a ball, determining a falling point of the ball, judging whether the point is positioned in the target area according to the falling point coordinates (x, y) of the ball, and calculating an accuracy score;

5. The method for intelligent class assessment and promotion of table tennis according to claim 3, wherein the batting continuity detection comprises: extracting the batting moment, calculating the continuous batting time interval and judging the continuous batting;

6. The intelligent test grade evaluation and promotion method for table tennis according to claim 1, wherein the step (3) comprises a forward pushing and blocking service parameter, a backward pushing and blocking service parameter, a forward tapping service parameter, a backward tapping service parameter, a forward quick tapping service parameter, a backward quick tapping service parameter, a receiving service parameter, a pulling service parameter and corresponding scoring standards;

7. The method of claim 6, wherein the scoring criteria include hit rate, speed control, spin control, accuracy, and continuity;

8. The intelligent table tennis class assessment and promotion method according to claim 1, wherein the step (4) comprises collecting data, storing personal information, class score, various scoring details and test data of a player into a database or a file, and analyzing the data according to the following indexes:

average division: average score obtained by player in each grade item;

highest score: the player obtains the highest score in each grade item;

lowest score: the player obtains the lowest score in each grade item;

9. The method for intelligent table tennis class assessment and promotion according to claim 1, wherein the step (5) comprises

(5-2) recommending training resources:

training resources provided for the players according to the skill levels of the players; recommending relevant training resources for the player according to the historical training record of the player by using a collaborative filtering algorithm or a content recommendation algorithm;

where n is the number of training resources, ri is the score of the player on the ith training resource, si is the similarity of the ith training resource;

(5-3) training progress monitoring:

10. The intelligent table tennis ball examination grade evaluation and lifting system is characterized by comprising an intelligent ball dispenser, a monitoring camera and a sensor;