Billiard hitting data analysis system and method

Abstract

The invention discloses a billiard hitting data analysis system and method, and relates to the technical field of physical training. The system comprises a data acquisition module used for acquiring ball hitting data; the ball hitting recognition module is used for carrying out ball hitting behavior recognition on the ball hitting data, determining a key ball hitting behavior, obtaining single-time ball hitting data according to the key ball hitting behavior and carrying out behavior marking on the single-time ball hitting data to obtain target ball hitting data; the analysis and diagnosis module is used for performing index statistics according to the target ball hitting data to obtain a ball hitting index; and the result feedback module is used for visually displaying the ball hitting process and the corresponding ball hitting indexes. According to the method, the efficiency and the effect of ball hitting training are improved, and coaches and athletes are helped to accurately recognize technical defects, formulate targeted training plans and optimize ball hitting strategies by providing in-depth ball hitting data analysis and visual feedback in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of sports training, and in particular to a billiards hitting data analysis system and method. Background Art

[0002] In the existing technology of billiards, athletes have long relied on personal experience, intuition and long-term practice to guide their aiming and hitting movements during training and competition. Although this traditional method can cultivate high-level athletes, it has obvious deficiencies in the accuracy, efficiency and replicability of technical improvement.

[0003] Specifically, traditional training methods often fail to provide accurate quantitative analysis of key parameters in the batting process, such as batting force, angle, speed, etc. These parameters have a decisive influence on the movement path and final landing point of the target ball. Athletes can only rely on their own feelings and repeated attempts to continuously adjust and improve, a process that is not only time-consuming and labor-intensive, but also often difficult to achieve the best results.

[0004] In addition, although there are some training assistance tools or methods on the market, such as video analysis, batting simulators, etc., most of these tools can only provide a certain degree of assistance and cannot capture and record the detailed data of the batting process in real time and comprehensively. They often focus on the feedback of the batting results, but ignore the dynamic changes in the batting process, so it is difficult to provide athletes with accurate and targeted feedback and improvement guidance. Summary of the invention

[0005] In view of the defects in the prior art, the present invention provides a billiards hitting data analysis system and method, which can collect and analyze various key data in the hitting process in real time and comprehensively, and provide athletes with accurate and targeted feedback and improvement guidance.

[0006] On the one hand, an embodiment of the present invention provides a billiards shot data analysis system, the system comprising:

[0007] A data collection module, used for collecting batting data;

[0008] a shot recognition module, for performing shot behavior recognition on the shot data, determining key shot behaviors, acquiring single shot data according to the key shot behaviors, and marking the single shot data with behaviors to obtain target shot data; wherein the key shot behaviors include preparing to shoot, ending the shot, moving the club forward, moving the club backward, and touching the ball;

[0009] An analysis and diagnosis module, used for performing index statistics according to the target hitting data to obtain hitting indexes;

[0010] The result feedback module is used to visualize the batting process and its corresponding batting indicators.

[0011] Preferably, the data acquisition module includes multiple groups of sensors arranged on the club, the sensors include a six-axis attitude sensor and a vibration sensor, and the hitting data includes one or more of angle data, angular velocity data, speed data, acceleration data and vibration data.

[0012] Preferably, the hitting recognition module is specifically used for: setting the sliding window size; using a scanning program to perform a comprehensive scanning and analysis on the hitting data in the sliding window based on a behavior feature library to determine key hitting behaviors.

[0013] Preferably, a scanning program is used to perform a comprehensive scan and analysis on the batting data in the sliding window based on a behavior feature library to determine key batting behaviors, including: when the batting preparation behavior is scanned, single batting data caching and behavior marking are started until the batting behavior is scanned to end, thereby obtaining target batting data.

[0014] Preferably, the scanning program is configured with an empirical feature recognition model and an AI feature recognition model.

[0015] Preferably, the training process of the AI ​​feature recognition model includes:

[0016] Acquire experimental data corresponding to the batting experiment, and annotate the experimental data to form a training set;

[0017] Construct an initial batting feature recognition model based on sliding windows;

[0018] The initial batting feature recognition model is trained using the training set to obtain an AI feature recognition model.

[0019] Preferably, the method for labeling the experimental data includes one or more of manual labeling, image recognition labeling and large model auxiliary labeling.

[0020] Preferably, it further comprises a data communication module, and the data communication module is a near field communication module.

[0021] On the other hand, an embodiment of the present invention provides a billiards shot data analysis method, the method comprising the following steps:

[0022] Step 1, collecting batting data;

[0023] Step 2, performing batting behavior recognition on the batting data, determining key batting behaviors, acquiring single batting data according to the key batting behaviors, and marking the single batting data with behaviors to obtain target batting data; wherein the key batting behaviors include preparing to bat, ending batting, moving the club forward, moving the club backward, and touching the ball;

[0024] Step 3, performing index statistics according to the target hitting data to obtain the hitting index;

[0025] Step 4: Visually display the batting process and its corresponding batting indicators.

[0026] The beneficial effects of the present invention are: improving the efficiency and effect of batting training, and helping coaches and athletes accurately identify technical shortcomings, develop targeted training plans, and optimize batting strategies by providing in-depth batting data analysis and intuitive feedback in real time. At the same time, it also promotes the selection of personalized club equipment, promotes the development of sports science research, and enhances the user experience and participation. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0028] Figure 1 A system block diagram of a billiards shot data analysis system provided by an embodiment of the present invention;

[0029] Figure 2 A schematic diagram of the structure of a data acquisition module provided in an embodiment of the present invention;

[0030] Figure 3 A schematic diagram of the installation position of the data acquisition module provided in an embodiment of the present invention;

[0031] Figure 4 A flowchart of a ball hitting recognition model provided by an embodiment of the present invention;

[0032] Figure 5 A training flow chart of an AI feature recognition model provided in an embodiment of the present invention;

[0033] Figure 6 The present invention provides a flowchart of a billiards shot data analysis method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0035] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the invention belongs.

[0036] Example 1

[0037] like Figure 1 As shown, an embodiment of the present invention provides a billiards shot data analysis system, the system comprising:

[0038] The data acquisition module is used to collect the batting data.

[0039] Specifically, the data acquisition module includes multiple groups of sensors arranged on the club, the sensors include a six-axis attitude sensor and a vibration sensor, and the hitting data includes one or more of angle data, angular velocity data, speed data, acceleration data and vibration data.

[0040] The data acquisition module is mainly composed of multiple groups of sensors, which are carefully placed at key positions of the billiard cue to ensure that various subtle changes in the process of hitting the ball can be captured. Specifically, the sensors include but are not limited to six-axis attitude sensors and vibration sensors. The six-axis attitude sensor can collect the attitude information of the billiard cue in three-dimensional space in real time, including angle, angular velocity, etc. Through the six-axis attitude sensor, the movement trajectory and attitude changes of the billiard cue during the hitting process can be accurately understood. Vibration sensor: The vibration sensor is used to collect the vibration data generated by the billiard cue when hitting the ball. This data is crucial for analyzing key parameters such as the strength and speed of the hitting, and the instantaneous state of contact between the cue and the ball.

[0041] It should be noted that the data acquisition module also integrates batteries, single-chip microcomputers, switches, charging boards and other devices. These devices are integrated in a housing and can be installed in the front middle and tail end of the billiard cue. The specific installation position is set according to actual needs and is not limited in the embodiments of the present invention.

[0042] The batting recognition module is used to identify the batting behavior of the batting data, determine the key batting behavior, obtain single batting data according to the key batting behavior and perform behavior marking on the single batting data to obtain target batting data; wherein the key batting behavior includes preparing for batting, ending batting, moving the club forward, moving the club backward and touching the ball.

[0043] In an embodiment of the present invention, the batting recognition module is specifically used to: set the size of the sliding window; use a scanning program to comprehensively scan and analyze the batting data in the sliding window based on a behavior feature library to determine key batting behaviors; wherein, using a scanning program to comprehensively scan and analyze the batting data in the sliding window based on a behavior feature library to determine key batting behaviors includes: when scanning the preparation for batting behavior, starting a single batting data cache and behavior marking until scanning the end of the batting behavior to obtain target batting data.

[0044] like Figure 4 As shown, in actual work, the system starts and scans the current batting data through the batting recognition module. During the scanning process, the module will intelligently determine whether there are specific batting behavior patterns that meet the preset requirements in these data. Once the batting data that meets the requirements is identified, the module will immediately start the data recording function to accurately identify and record the key behaviors in this batting. These key behavior data are then sent to the analysis and diagnosis module for more in-depth data processing to support subsequent decision-making and training optimization. If no batting data that meets the specific behavior is found in the scanned batting data, the module will automatically return to the scanning stage and continue to monitor new batting data to ensure that no possible batting behavior is missed.

[0045] Through automated data recording and analysis, the entire process enables accurate identification and efficient processing of batting behavior, providing strong support for improving athletic performance and optimizing training strategies.

[0046] Specifically, the scanning program is configured with an empirical feature recognition model and an AI feature recognition model.

[0047] The empirical feature recognition model is built based on a large amount of batting experimental data and expert experience. It relies on the knowledge of experts in the field and a deep understanding of batting behavior, and identifies batting behavior by manually extracting and defining a series of key features. These features may include angle changes, speed changes, acceleration patterns, etc. of the club, which are all empirical laws derived from long-term observations and experiments.

[0048] The advantage of the empirical feature recognition model is that it is highly interpretable and can clearly reflect the physical characteristics and movement laws of the batting behavior. However, since it relies on manually defined features, it may have certain limitations when facing complex and changeable batting situations. Therefore, the embodiment of the present invention also configures an AI feature recognition model in the behavior feature library.

[0049] In practical applications, the empirical feature recognition model and the AI ​​feature recognition model can complement each other and jointly improve the recognition effect of batting behavior. For example, the empirical feature recognition model can be used for preliminary screening and judgment, and then the AI ​​feature recognition model can be used for refined and personalized recognition. In this way, the advantages of the two models can be fully utilized while avoiding their respective limitations.

[0050] The AI ​​feature recognition model is trained through machine learning algorithms, and it has stronger automatic learning and generalization capabilities. In order to build an AI feature recognition model, it is first necessary to collect a large amount of batting experiment data and annotate these data to form a training set.

[0051] During the training process, the machine learning algorithm automatically extracts useful features from the raw data and learns the relationship between these features and batting behavior. Through continuous iteration and optimization, the AI ​​feature recognition model can gradually learn more complex and detailed batting behavior features, thereby improving the accuracy and robustness of recognition.

[0052] Specifically, Figure 5 As shown, the training process of the AI ​​feature recognition model includes: obtaining experimental data corresponding to the batting experiment, annotating the experimental data to form a training set; constructing an initial batting feature recognition model based on a sliding window; and using the training set to train the initial batting feature recognition model to obtain an AI feature recognition model.

[0053] It should be noted that the selection of the initial batting feature recognition model is diverse, and it can be any one of a random forest classifier model, a support vector machine (SVM) model, a neural network model, etc. These models have their own characteristics and are suitable for different scenarios and data sets. In this embodiment, a random forest classifier model is selected as the initial batting feature recognition model.

[0054] Furthermore, the method for labeling the experimental data includes one or more of manual labeling, image recognition labeling and large model auxiliary labeling.

[0055] It should be noted that the large model auxiliary standard of the embodiment of the present invention refers to inputting the meaning of the observed indicators and a small number of sample data examples into the large model through the prompt word engineering. The large model can sample the GPT-4 large model or other large models. There is no restriction in this embodiment, and the large model is required to automatically label a large amount of data that has not been observed and labeled, and output the labeled results. This can save the time and cost required for labeling, and can provide a large amount of experimental data according to the requirements of subsequent classifiers.

[0056] The analysis and diagnosis module is used to perform index statistics based on the target hitting data to obtain the hitting index.

[0057] In order to achieve efficient data analysis and diagnosis, the module uses mainstream real-time data processing technologies. These technologies enable the module to process data immediately when it is received, without having to wait for a certain amount of data to accumulate before batch processing. In this way, coaches and athletes can get feedback almost immediately after the ball is hit, so as to adjust training strategies or hitting techniques in a timely manner.

[0058] In addition, in order to achieve accurate statistical calculation of indicators, the analysis and diagnosis module also needs to rely on precise data segmentation labels. These labels divide each shot process into different stages, such as preparation stage, club movement stage, and shot stage. By analyzing the data of each stage separately, the module can calculate various shot indicators more accurately and provide more detailed feedback to coaches and athletes.

[0059] The analysis and diagnosis module provided in this embodiment is a core component built on the cloud, which is specifically responsible for processing the batting data transmitted by the data communication module. This module can conduct in-depth analysis of the received data in real time, and complete the statistical calculation of a series of key indicators based on the segmented identification of each batting data. These indicators include but are not limited to up and down shaking, left and right shaking, acceleration time, number of dribbling, dribbling speed and batting time, which are crucial for evaluating the technical level and batting quality of the batter.

[0060] The result feedback module is used to visualize the batting process and its corresponding batting indicators.

[0061] The result feedback module can simulate and display the dynamic picture of the batting process, including the trajectory of the ball, the posture at the moment of batting, etc. The user can observe his batting action from multiple angles through the playback function, so as to understand his batting habits and technical characteristics more intuitively. The module will count and display a series of key batting indicators, including preparation phase indicators, hitting phase indicators and retreat phase indicators. The preparation phase indicators include the number of forward dribbling, the number of backward dribbling, straightness and flatness, etc. The hitting phase indicators include straightness, flatness, amplitude, vibration graph, speed, maximum speed time point and hitting time point, etc. The retreat phase indicators include the time of hitting and retreating. These indicators are presented in the form of charts or numerical values ​​to help users clearly understand their batting performance and identify areas that need improvement. The module can also analyze and display the performance characteristics of the clubs used, such as elasticity, weight distribution, etc. By understanding the characteristics of the clubs, users can better choose the clubs that suit them and optimize their batting strategies.

[0062] The result feedback module provided in this embodiment uses data visualization technology to display complex data analysis results to users in an intuitive and easy-to-understand manner, aiming to help users gain a deeper understanding of their own batting process, identify areas for improvement in batting, and gain insight into the performance characteristics of the clubs used.

[0063] In order to realize communication between acquisition terminals, the present invention further comprises a data communication module, and the data communication module is a near field communication module.

[0064] Specifically, the near-field communication module (Bluetooth) is used to identify and bind the acquisition device; near-field communication uses Bluetooth 4.0 or higher to support the low-power Bluetooth (BLE) function to ensure efficient communication between devices. The standard Bluetooth pairing process is adopted, including PIN code input, device name display, etc. When the communication module is started, the device is first discovered to search for surrounding acquisition devices (data acquisition modules). After the target device is found, a pairing request is initiated and the response from the other party is waited for. The necessary security authentication is performed to ensure the security of the connection. After the pairing is successful, the communication module establishes a binding relationship with the acquisition device to prepare for subsequent data transmission.

[0065] In summary, an embodiment of the present invention provides a billiards hitting data analysis system, which includes a data acquisition module for collecting hitting data; a hitting identification module for identifying hitting behaviors of the hitting data, determining key hitting behaviors, obtaining single hitting data according to the key hitting behaviors and marking the single hitting data with behaviors to obtain target hitting data; wherein the key hitting behaviors include preparing for hitting, ending hitting, moving the rod forward, moving the rod backward and touching the ball; an analysis and diagnosis module for performing index statistics according to the target hitting data to obtain hitting indicators; and a result feedback module for visually displaying the hitting process and its corresponding hitting indicators. The present invention improves the efficiency and effect of hitting training, and helps coaches and athletes accurately identify technical shortcomings, develop targeted training plans, and optimize hitting strategies by providing in-depth hitting data analysis and intuitive feedback in real time. At the same time, it also promotes the selection of personalized club equipment, promotes the development of sports science research, and enhances the user experience and participation.

[0066] Example 2

[0067] like Figure 6As shown, based on the same inventive concept, an embodiment of the present invention provides a billiards hitting data analysis method, the method comprising the following steps: step 1, collecting hitting data; step 2, performing hitting behavior recognition on the hitting data, determining key hitting behaviors, obtaining single hitting data according to the key hitting behaviors and performing behavior marking on the single hitting data to obtain target hitting data; wherein the key hitting behaviors include preparing for hitting, ending hitting, moving the stick forward, moving the stick backward and touching the ball; step 3, performing indicator statistics according to the target hitting data to obtain hitting indicators; step 4, visually displaying the hitting process and its corresponding hitting indicators.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A billiards shot data analysis system, characterized in that: include: A data collection module, used for collecting batting data; a shot recognition module, for performing shot behavior recognition on the shot data, determining key shot behaviors, acquiring single shot data according to the key shot behaviors, and marking the single shot data with behaviors to obtain target shot data; wherein the key shot behaviors include preparing to shoot, ending the shot, moving the club forward, moving the club backward, and touching the ball; An analysis and diagnosis module, used for performing index statistics according to the target hitting data to obtain hitting indexes; The result feedback module is used to visualize the batting process and its corresponding batting indicators.

2. A billiards shot data analysis system according to claim 1, characterized in that: The data acquisition module includes multiple groups of sensors arranged on the club, the sensors include a six-axis attitude sensor and a vibration sensor, and the hitting data includes one or more of angle data, angular velocity data, speed data, acceleration data and vibration data.

3. A billiards shot data analysis system according to claim 1, characterized in that: The batting recognition module is specifically used for: setting the size of the sliding window; using a scanning program to perform a comprehensive scan and analysis on the batting data in the sliding window based on a behavior feature library to determine key batting behaviors.

4. A billiards shot data analysis system according to claim 3, characterized in that: A scanning program is used to perform a comprehensive scan and analysis on the batting data in the sliding window based on a behavior feature library to determine key batting behaviors, including: when a batting preparation behavior is scanned, a single batting data cache and behavior marking are started, until the batting behavior is scanned to end, and target batting data is obtained.

5. A billiards shot data analysis system according to claim 3, characterized in that: The scanning program is configured with an empirical feature recognition model and an AI feature recognition model.

6. A billiards shot data analysis system according to claim 5, characterized in that: The training process of the AI ​​feature recognition model includes: Acquire experimental data corresponding to the batting experiment, and annotate the experimental data to form a training set; Construct an initial batting feature recognition model based on sliding windows; The initial batting feature recognition model is trained using the training set to obtain an AI feature recognition model.

7. A billiards shot data analysis system according to claim 6, characterized in that: The method of labeling the experimental data includes one or more of manual labeling, image recognition labeling and large model auxiliary labeling.

8. A billiards shot data analysis system according to claim 1, characterized in that: It also includes a data communication module, which is a near field communication module.

9. A billiards hitting data analysis method, characterized in that: The following steps are involved: Step 1, collecting batting data; Step 2, performing batting behavior recognition on the batting data, determining key batting behaviors, acquiring single batting data according to the key batting behaviors, and marking the single batting data with behaviors to obtain target batting data; wherein the key batting behaviors include preparing to bat, ending batting, moving the club forward, moving the club backward, and touching the ball; Step 3, performing index statistics according to the target hitting data to obtain the hitting index; Step 4: Visually display the batting process and its corresponding batting indicators.

Images