KR-102963689-B1 - DATA DRIVEN SYSTEM FOR ANALYZING PLAYER PERFORMANCE AND PREDICTING GROWTH POTENTIAL AND HOW TO DO IT

KR102963689B1KR 102963689 B1KR102963689 B1KR 102963689B1KR-102963689-B1

Abstract

A data-based system and method for analyzing athlete performance and predicting growth potential are disclosed. The present invention utilizes an artificial intelligence model based on the athlete's athletic performance and arbitrary evaluation data to quantify the athlete's performance and predict growth potential.

Inventors

손외태
손이경
이동섭

Assignees

주식회사 갤로핑

Dates

Publication Date: 20260512
Application Date: 20250516

Claims (15)

When measurement data related to the athletic ability of the athlete (400) and one or more external data such as the athlete's (400) age, physical condition, training history, surrounding environment, and injury history are input, the position and movement of the athlete's (400) body joints are analyzed based on the measurement data to generate motion data based on changes in joint position over time, and the motion data is input into an artificial intelligence-based analysis model to calculate a quantified performance score for each arbitrary performance measurement item. The motion data and the external data are processed by a time-series analysis model to learn a growth pattern including the athlete's performance, physical data, and performance score. Based on the score calculated therefrom, the athlete's current ability and future growth potential are evaluated. A Growth Potential Score (GPS) is generated by integrating external factor scores that reflect the impact of environmental changes on the athlete's growth to quantitatively evaluate and predict the athlete's growth potential. Additionally, a risk assessment is generated by analyzing the injury risk based on the athlete's (400) past injury data and the performance score, performance variability based on performance data, and external risks arising from the training environment and game schedule to quantitatively evaluate and predict the possibility of injury. A data-based player performance analysis and growth potential prediction system comprising: a performance analysis device (100) that calculates a score (Risk Score) and outputs the performance score, growth potential score, and risk score.
In paragraph 1, A data-based player performance analysis and growth potential prediction system characterized by the above-mentioned measurement items including one or more of agility, acceleration, balance, two-foot utilization, and offensive/defensive transition ability.
In paragraph 1, A data-based player performance analysis and growth potential prediction system characterized by calculating the growth potential score based on the age, physical condition, and training history of the motion data and the external data, while reflecting arbitrary weights for the player's (400) current ability, growth potential, and external factors.
In paragraph 1, A data-based player performance analysis and growth potential prediction system characterized by calculating the above risk score based on the surrounding environment and injury history of the above motion data and the above external data, while reflecting arbitrary weights for each injury risk, performance variability, and external risk of the above player (400).
In paragraph 1, The above performance analysis device (100) includes an input unit (110) that receives measurement data related to athletic ability of the athlete (400) from a measuring device (200) and an administrator terminal (300), as well as external data related to the athlete's (400) age, physical condition, training history, surrounding environment, and injury history; A processing unit (120) that generates motion data based on changes in joint positions over time by analyzing movement patterns related to the position of the player’s hands, feet, arms, and legs, steps, movement of the center of gravity, and angles of the arms, feet, knees, legs, and upper body based on the above measurement data; and A data-based player performance analysis and growth potential prediction system characterized by including an analysis unit (130) that calculates a performance score by quantifying the motion data according to measurement items for agility, acceleration, balance, two-foot utilization, and offensive/defensive transition ability using the above artificial intelligence-based analysis model to analyze performance ability, calculates a growth potential score based on the motion data and the above age, physical condition, and training history, and calculates and outputs a risk score based on the motion data and the above surrounding environment and injury history.
In paragraph 5, The processing unit (120) comprises a preprocessing unit (121) that performs missing value correction, noise removal, frame normalization, and object tracking using an object detection algorithm for the measurement data; and A data-based athlete performance analysis and growth potential prediction system characterized by including a feature extraction unit (122) that generates motion data by agility, acceleration, balance, two-foot utilization, and offensive/defensive transition ability by analyzing movement patterns related to the position of the athlete's hands, feet, arms, and legs, steps, center of gravity movement, and angles of the arms, feet, knees, legs, and upper body based on the above-mentioned preprocessed measurement data.
In paragraph 5, The above analysis unit (130) calculates a quantified performance score according to measurement items through analysis of the motion data based on transition speed, reaction time, time to reach maximum speed, posture stability, motion accuracy, force balance, agility, acceleration, balance, two-foot utilization, and offensive/defensive transition ability using the above artificial intelligence-based analysis model; and A data-based player performance analysis and growth potential prediction system characterized by including: a data prediction unit (132) that calculates a growth potential score based on the motion data and the age, physical condition, and training history, and a risk score based on the motion data and the surrounding environment and injury history, and calculates a growth potential score and a risk score reflecting different weights for the player's (400) current ability, growth potential, external factors, injury risk, performance variability, and external risk, and outputs them as arbitrary visualized indicators.
In paragraph 1, The data-based athlete performance analysis and growth potential prediction system described above further comprises: a measuring device (200) that guides the athlete (400) to perform training based on arbitrary training content, wherein a plurality of training lines (510, 520, 530, 540, 550) are inscribed and exscribed to each other, and a plurality of training sections (560, 570, 580, 590) formed by at least a portion of the training lines (510, 520, 530, 540, 550) or sections that touch each other, and captures and stores the training appearance of the athlete (400).
a) A performance analysis device (100) collects measurement data related to the athletic ability of a player (400) and one or more external data among the player's (400) age, physical condition, training history, surrounding environment and injury history; b) a step in which the performance analysis device (100) analyzes the body joint positions and movements of the player (400) based on the measurement data and generates motion data based on changes in joint positions over time; and c) a step in which the performance analysis device (100) inputs the motion data into an artificial intelligence-based analysis model to calculate a performance score quantified for each arbitrary performance measurement item, processes the motion data and the external data into a time-series analysis model to learn a growth pattern including the athlete's performance, physical data, and performance score, evaluates the athlete's current ability and future growth potential based on the score calculated therefrom, and calculates a Growth Potential Score (GPS) that quantitatively evaluates and predicts the athlete's growth potential by integrating an external factor score that reflects the impact of environmental changes on the athlete's growth, and calculates a Risk Score that quantitatively evaluates and predicts the possibility of injury by analyzing the injury risk based on the athlete's (400) past injury data and the performance score, performance variability based on performance data, and external risks occurring in the training environment and game schedule, and outputs the performance score, growth potential score, and risk score; comprising a data-based athlete performance analysis and growth potential prediction method.
In Paragraph 9, A data-based method for analyzing player performance and predicting growth potential, characterized in that the above measurement items include one or more of agility, acceleration, balance, two-foot utilization, and offensive/defensive transition ability.
In Paragraph 9, A data-based player performance analysis and growth potential prediction method characterized by calculating the growth potential score based on the age, physical condition, and training history of the motion data and the external data, while reflecting arbitrary weights for the player's (400) current ability, growth potential, and external factors.
In Paragraph 9, A data-based method for analyzing player performance ability and predicting growth potential, characterized in that the above risk score is calculated based on the surrounding environment and injury history of the above motion data and the above external data, while reflecting arbitrary weights for each injury risk, performance variability, and external risk of the player (400).
In Paragraph 9, The above step b) includes b-1) a step in which the performance analysis device (100) preprocesses the measurement data by supplementing missing values, removing noise, normalizing frames, and tracking objects using an object detection algorithm; and b-2) A step of generating motion data for agility, acceleration, balance, two-foot utilization, and offensive/defensive transition ability by analyzing movement patterns related to the position of the player’s hands, feet, arms, and legs, steps, center of gravity movement, and angles of the arms, feet, knees, legs, and upper body based on the preprocessed measurement data above; characterized by including the data-based player performance analysis and growth potential prediction method.
In Paragraph 9, The above step c) is c-1) a step in which the performance analysis device (100) calculates a quantified performance score according to measurement items through analysis of transition speed, reaction time, time to reach maximum speed, posture stability, motion accuracy, force balance, agility, acceleration, balance, two-foot utilization, and offensive/defensive transition ability based on the motion data, in order to analyze performance using the artificial intelligence-based analysis model; and c-2) A step of calculating a performance score including a growth potential score based on the motion data and the age, physical condition, and training history, and a risk score based on the motion data and the surrounding environment and injury history, wherein the calculation is performed by reflecting different weights for the current ability, growth potential, external factors, injury risk, performance variability, and external risk of the player (400); and c-3) A step of outputting the calculated performance score as an arbitrary visualized indicator; characterized by a data-based player performance analysis and growth potential prediction method.
In Paragraph 9, A data-based method for analyzing athlete performance and predicting growth potential, characterized in that the measurement data related to the athletic ability of the athlete (400) is image data of the athlete (400) performing arbitrary training along a training pattern (500) in which a plurality of training lines (510, 520, 530, 540, 550) are in contact with and externally contact each other based on arbitrary training content output from a measuring device (200), and a plurality of training sections (560, 570, 580, 590) are formed by at least a portion of the training lines (510, 520, 530, 540, 550) or sections that contact each other.

Description

Data-Driven System for Analyzing Player Performance and Predicting Growth Potential and Method The present invention relates to a data-based athlete performance analysis and growth potential prediction system and method, and more specifically, to a data-based athlete performance analysis and growth potential prediction system and method that quantifies an athlete's performance ability and predicts growth potential using an artificial intelligence model based on the athlete's athletic performance ability and arbitrary evaluation data. Generally, training was planned and implemented based on the personal judgment of coaches or trainers; however, it was difficult to assess the effectiveness of training tailored to individual characteristics, particularly because it was challenging to quantify training outcomes. For example, while it is crucial to apply the appropriate exercise load in endurance training, traditionally, setting the load relied on the subjective judgment of coaches or trainers. Consequently, it was difficult to objectively assess the effectiveness of endurance improvement, resulting in extremely minimal results. Furthermore, although the degree of endurance improvement was also evaluated based on maximum oxygen uptake values, simple maximum oxygen uptake cannot be said to accurately reflect an athlete's endurance during a game, even in situations where athletes must run and rest to momentarily reach maximum activity levels and recover quickly. Furthermore, although agility training is a factor that significantly influences game results, especially for athletes, conventional methods can only measure it through indirect means such as the side step. In addition, conventional athlete evaluation methods have mainly utilized fragmentary data such as straight-line running speed (e.g., 40m run) or competition records, but this has the problem of not sufficiently reflecting the complex movements and transition abilities required in actual competition situations. Furthermore, conventional player evaluation methods mostly focus on game analysis, and there is a lack of technology to systematically evaluate a player's potential in a standardized testing environment. Additionally, there is a problem in that it is difficult to quantitatively assess and predict a player's injury risk and growth potential. FIG. 1 is a block diagram schematically illustrating the configuration of a data-based player performance analysis and growth potential prediction system according to one embodiment of the present invention. FIG. 2 is a block diagram illustrating the configuration of a performance analysis device of a data-based player performance analysis and growth potential prediction system according to an embodiment of FIG. 1. FIG. 3 is a block diagram illustrating the configuration of the input section of a performance analysis device according to an embodiment of FIG. 2. FIG. 4 is a block diagram illustrating the configuration of the processing unit of a performance analysis device according to the embodiment of FIG. 2. FIG. 5 is a block diagram illustrating the configuration of the analysis unit of a performance analysis device according to an embodiment of FIG. 2. FIG. 6 is a block diagram illustrating the configuration of a measurement device for a data-based player performance analysis and growth potential prediction system according to an embodiment of FIG. 1. FIG. 7 is an example diagram illustrating a measurement operation using a measuring device according to an embodiment of FIG. 7. FIG. 8 is an example diagram illustrating measurement data matching in the data-based player performance analysis process according to the embodiment of FIG. 2. FIG. 9 is a flowchart illustrating a data-based player performance analysis and growth potential prediction method according to an embodiment of the present invention. FIG. 10 is a flowchart illustrating the motion data generation process of a data-based player performance analysis and growth potential prediction method according to the embodiment of FIG. 9. FIG. 11 is a flowchart illustrating the score generation process of a data-based player performance analysis and growth potential prediction method according to the embodiment of FIG. 9. FIG. 12 is an example diagram shown to explain the visualization indicator images generated in the data-based player performance analysis and growth potential prediction method according to the embodiment of FIG. 9. Hereinafter, the present invention will be described in detail with reference to preferred embodiments of the invention and the accompanying drawings, under the premise that identical reference numerals in the drawings refer to identical components. Before describing specific details for the implementation of the present invention, it should be noted that configurations not directly related to the technical essence of the present invention have been omitted to the extent that the technical essence of the present invention is