Football player match analysis method and device

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a football player match analysis method and device.According to the football player match analysis method and device, tracking data of a football match is used for forming analysis summarization on energy metabolism, running speed distribution, running distances in different speed intervals, ball holding time and running hot areas of football players in the whole match; according to the method, the personal physical ability condition of the player and the enthusiasm of the player in a match are evaluated, and a radar map of different-dimension related events of the player is formed on the basis of event data, so that behavior analysis of specific styles of the player is realized. Besides, the event data and the tracking data of the football match are used to realize personal style mining of different events processed by players, scoring of different positions and comprehensive analysis of contribution to teams, and an analysis report based on the target theme football match is generated, so that a comprehensive and more-dimensional evaluation and analysis mode is provided; and therefore, the potential of teenager players can be fully explored, and the players can be cultivated in a more scientific manner.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method and device for analyzing a football player match. Background Art

[0002] Football is very popular among young people in my country. Various youth training teams, campus teams and social training institutions organize football matches of various sizes and forms. The purpose is not only to stimulate young people's enthusiasm for football participation, but also to discover young players with training potential through various types of competitions. Therefore, it is crucial to conduct technical and tactical evaluation of young football players in a scientific way.

[0003] In the related art, a common way to evaluate young football players is the subjective evaluation of coaches and competition judges. By observing and recording the different competitions and daily training of young people, players with different abilities are screened into teams of different levels. This method relies too much on the experience accumulation of coaches and competition judges, and it is difficult to give relatively objective and well-founded evaluation standards and results. Another way is to measure the basic performance of young players by making statistics on the results of each game, such as the number of shots, goals, passes, pass success rate and other indicators of each player. This method reflects the performance of young players in the game to a certain extent. However, this method reflects the ability of the players too single, and the aspects that can be evaluated are relatively one-sided, and it cannot fully take into account the different physical abilities and style characteristics of young players. Summary of the invention

[0004] In view of this, the present invention provides a football player game analysis method and device to solve the problem that it is difficult to provide relatively objective and well-founded evaluation standards and results, and that the player's ability is too single, the aspects that can be evaluated are relatively one-sided, and the different physical abilities and style characteristics of young players cannot be fully considered.

[0005] In a first aspect, the present invention provides a method for analyzing a football player's match, 1. A method for analyzing a football player's match, the method comprising:

[0006] Obtain tracking data, event data, and match data of multiple players participating in a target-themed soccer match; wherein the tracking data records the positions and speeds of multiple players and the ball on the field, as well as the team that possesses the ball; the event data records the entire process of the match in time series; and the match data records match information according to the date and time of the match;

[0007] Pre-processing of tracking data, event data, and game data;

[0008] Based on the pre-processed tracking data and event data, analyze each player's game energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, and radar distribution data;

[0009] Based on the pre-processed event data, analyze each player's individual passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in the game;

[0010] Based on the game energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, radar distribution data, each player's personal passing style based on different types of events, each player's performance behavior score in different activity positions on the court, and each player's collective contribution value in this game, an analysis report on the target theme football game is generated.

[0011] In an optional embodiment, the tracking data, event data and game data are pre-processed, including:

[0012] According to the date and time of the game, the tracking data, event data and game data of the same game of the target theme football game are aligned, filtered and cleaned in turn for preprocessing.

[0013] In an optional implementation, the tracking data, event data and game data of the same game of the target theme football game are aligned, filtered and cleaned in sequence according to the date and time of the game, including:

[0014] Get the player list, jersey numbers, home and away team logos in the game information;

[0015] Align the fields of player rosters, jersey numbers, and home and away team logos in tracking data and event data;

[0016] Filter out-of-order frames from tracking data based on the number of data frames;

[0017] filtering out-of-order frames of the tracking data based on player position coordinates in the tracking data;

[0018] Fix the filtered tracking data according to the preset frame number;

[0019] Clean the tracking data after fixation;

[0020] According to the event time recorded in each frame of the event data, the event data is converted into tracking data of the corresponding number of frames;

[0021] According to the tracking data converted from the event data to the corresponding number of frames, locate the corresponding tracking data segment, and compare the position of the player in the event data with the position of the player in the tracking data to determine whether there is any obvious abnormality in the record of the event data:

[0022] If the position of the player in the event data and the position of the player in the tracking data differ within a preset range, the record in the event data is judged to be normal;

[0023] If the event data exceeds the preset range, the record in the event data is judged to be abnormal and the abnormal data is removed.

[0024] In an optional embodiment, the game energy metabolism data, running speed distribution data, running interval data at different speeds, ball holding time data, running hot zone data, and radar distribution data of each player are analyzed, including:

[0025] Based on the speed of each player at each moment in the preprocessed tracking data, calculate the acceleration of each player at each moment;

[0026] Presenting the energy metabolism data of the players based on the acceleration of each player at each moment, and plotting the values ​​at each moment to form a first data graph;

[0027] Based on the pre-processed tracking data, the pre-processed tracking data is screened and counted according to low-speed, medium-speed, and high-speed running, and the running distances in different speed intervals are calculated to analyze the running abilities of different players;

[0028] Based on the pre-processed tracking data, the ball possession time of the home and away teams in the first and second half or different stages of the game is calculated according to the ball possession field, and a second data graph is drawn; or, the ball possession team is located by the ball possession mark of each frame, and the player currently in possession of the ball is determined by the distance between the player of the team and the ball. If the ball is passed by any player, the possession right still belongs to the passing player, then the ball possession time of each player is calculated based on the time before the next event occurs and the ball possession start time of the passing player;

[0029] Obtain pre-processed tracking data and present each player's running hotspots in the form of a heat map, and then conduct differential analysis on players of different positions and styles;

[0030] Based on different coaches' requirements for comparing players' multi-dimensional abilities, radar distribution data for the same player with different abilities is formed. Radar distribution data includes any player's physical abilities such as maximum acceleration, maximum speed, maximum consumption load, ball possession time, or the number of passes, successful passes, interceptions, and shots.

[0031] In an optional implementation, based on the pre-processed event data, each player's individual passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in the game are analyzed, including:

[0032] Clustering different types of passing sub-events using non-negative matrix factorization;

[0033] Calculate the distribution weight of each player in each passing sub-event to evaluate each player's individual style;

[0034] Generate a visual representation of each player's personal style based on the coach's needs;

[0035] Based on the pre-processed event data, the activity positions of multiple players on the court are classified;

[0036] Give players performance ratings in different positions based on the event types and outcomes of different active positions;

[0037] Based on the pre-processed event data, taking the shot or goal event as the basis, the entire pass chain before the shot is traced back to calculate the expected goal value or expected threat value of each player in the pass chain to evaluate the contribution of each player to the shot or goal event. The expected goal value or expected threat value of each player in the entire game is combined and counted to measure the contribution of each player to the team in this game.

[0038] The collective contribution value of the players to the target-themed football game is evaluated by counting the number of times the expected goal value or expected threat value ranking appears in the top five in multiple games, and a third data graph is generated.

[0039] In an optional implementation, the football player game analysis method in the embodiment of the present disclosure further includes:

[0040] Obtain video data of multiple players participating in a target-themed football match to assist in post-match review of key segments.

[0041] According to a second aspect, an embodiment of the present disclosure provides a soccer player game analysis device, the device comprising:

[0042] A data acquisition module is used to acquire tracking data, event data and match data of multiple players participating in a target theme football match; wherein the tracking data records the positions and speeds of multiple players and the ball on the field, as well as the team that has possession of the ball; the event data records the entire process of the entire match in time series; and the match data records the match information according to the date and time of the match;

[0043] Data processing module, used to pre-process tracking data, event data and game data;

[0044] The first analysis module is used to analyze the game energy metabolism data, running speed distribution data, running interval data at different speeds, ball holding time data, running hot zone data, and radar distribution data of each player based on the pre-processed tracking data and event data;

[0045] The second analysis module is used to analyze each player's individual passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in the game based on the pre-processed event data;

[0046] The report generation module is used to generate an analysis report of a target theme football match based on the match energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, radar distribution data, each player's personal passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in the match.

[0047] According to a third aspect, an embodiment of the present disclosure provides a computer device, including:

[0048] The memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the football player game analysis method in the first aspect or any embodiment of the first aspect by executing the computer instructions.

[0049] According to a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to enable a computer to execute the football player game analysis method in the first aspect or any embodiment of the first aspect.

[0050] According to a fifth aspect, an embodiment of the present disclosure provides a computer program product, including computer instructions, which are used to enable a computer to execute the football player game analysis method in the first aspect or any embodiment of the first aspect.

[0051] The technical solution of the present invention has the following advantages:

[0052] The present invention relates to the field of artificial intelligence technology, and discloses a method and device for analyzing a football player's game. The present invention uses tracking data of football games to form an analysis summary of the player's energy metabolism, running speed distribution, running distance in different speed intervals, ball holding time, and running hot spots throughout the game, so as to evaluate the player's personal physical condition and enthusiasm in the game, and form a radar chart of related events of different dimensions of the player based on event data to achieve behavioral analysis of the player's specific style. In addition, the present invention uses event data and tracking data of football games to realize the mining of the player's personal style in handling different events, scoring of different positions, and comprehensive analysis of the contribution to the team, and generates an analysis report based on the target theme football game to provide a comprehensive and multi-dimensional evaluation and analysis method, thereby being able to fully tap the potential of young players and train players in a more scientific way. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 is a schematic flow chart of a method for analyzing a soccer player's match according to an embodiment of the present invention;

[0055] Figure 2 is a schematic flow chart of a method for analyzing a soccer player's match according to an embodiment of the present invention;

[0056] Figure 3 is a flow chart of another football player game analysis method according to an embodiment of the present invention;

[0057] Figure 4 is a flow chart of another football player game analysis method according to an embodiment of the present invention;

[0058] Figure 5 is a flow chart of another football player game analysis method according to an embodiment of the present invention;

[0059] Figure 6 is a schematic diagram of different types of passing sub-events according to an embodiment of the present invention;

[0060] Figure 7 is a schematic diagram of a division result obtained by clustering all activity positions of players in a cup match according to an embodiment of the present invention;

[0061] Figure 8is a structural block diagram of a soccer player competition analysis device according to an embodiment of the present invention;

[0062] Fig. 9 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0064] According to an embodiment of the present invention, an embodiment of a method for analyzing a soccer player's match is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in an order different from that shown here.

[0065] In this embodiment, a football player game analysis method is provided, which can be used in the above-mentioned mobile terminals, such as mobile phones, tablet computers, etc. Figure 1 is a flow chart of a method for analyzing a soccer player match according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0066] Step S101, obtaining tracking data, event data and match data of multiple players participating in a target theme football match; wherein the tracking data records the positions and speeds of multiple players and the ball on the field, as well as the team that possesses the ball, the event data records the entire process of the entire game in time series, and the match data records the match information according to the date and time of the game.

[0067] Specifically, the multiple players in the disclosed embodiment may be young football players. According to the requirements of football matches, the number of athletes participating in the competition is usually 22, that is, 11 home team players and 11 away team players. The target theme football match is, for example, a "certain cup" match in 2024. Therefore, the participation of multiple players in the target theme football match can be considered as a match or the same type of match participated in by multiple teenagers in the same period, for example, U13 players participate in a "certain cup match" in 2024.

[0068] Furthermore, the tracking data in the embodiments of the present disclosure can be collected by a wearable device with UWB or GPS, which can be specifically extracted from the game video through computer vision technology. The data collection frequency can be 10Hz or 25Hz, that is, 10 frames or 25 frames of data per second. Each frame of data records the position and speed of multiple (22) players and the ball on the court at this moment, as well as the ownership of the ball at this moment. A coordinate system is established with a certain point at the four corners of the court or the center of the court as the origin. The coordinates of the players and the ball can be expressed as two-dimensional and three-dimensional coordinates based on the coordinate system. Among them, the coordinates of the players are expressed as two-dimensional coordinates, and the coordinates of the home team players can be expressed as The player speed can be expressed as Where i = 1, 2, ..., 11 represents the 11 players of the home team; the coordinates of the away team players can be expressed as The player speed can be expressed as i=1,2,…,11 represents the 11 players of the visiting team. The three-dimensional coordinates of the ball can be expressed as pos ball (x,y,z), the ball speed can be expressed as vel ball (x, y, z), where the coordinates of the home and away team players and the ball are scalars, and the velocities are vectors; ball control can be represented by ball_control, where ball_control = 1 if the home team controls the ball, and ball_control = 2 if the away team controls the ball. control =-1, if no one has control of the ball, then ball_Control=0.

[0069] In addition, the event data in the disclosed embodiment records all important events that occurred in the entire game in a time sequence, such as passing, intercepting, dribbling, shooting, scoring, etc. Each piece of data (a frame) is an important event, and contains information such as the type of event, the time and location of occurrence, the name and jersey number of the relevant player, etc. Due to the interval and a certain degree of randomness of events, the recording frequency of event data is not fixed, that is, the number of data frames is not based on a fixed time interval, but the event data still has a time sequence, for example, the event data from the start frame to the end frame must be recorded according to the time process of the game; at the same time, there is continuity between the data, for example, the next frame of the "pass" event must be "interception", "interception", "receiving" and other event types closely related to passing. The event type field is represented by type. If a frame of data is a shooting event, then this event type = 'shot'. The data format, data volume, degree of refinement of the game record, and field representation of the event data collected by different data vendors are different, but overall they all record the key events that occurred in the entire game. The game data records the date and time of the game.

[0070] Step S102, pre-processing the tracking data, event data and game data.

[0071] Specifically, in the disclosed embodiments, the tracking data, event data, and game data of multiple games may be cleaned and aligned.

[0072] Exemplarily, the disclosed embodiments may align and clean the tracking data, event data, and game data of the same game according to the date, time, and match information of the game.

[0073] Step S103, based on the pre-processed tracking data and event data, analyze each player's game energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, and radar distribution data.

[0074] Specifically, the first analysis module (basic analysis module) in the following football player game analysis device is used to analyze the game energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, and radar distribution data of each player based on the pre-processed tracking data and event data. Among them, the game energy metabolism data, running speed distribution data, different speed running interval data, and ball holding time data of each player are used to evaluate the player's personal ability; among them, the running hot zone data and radar distribution data of each player are used to evaluate the player's game enthusiasm.

[0075] The disclosed embodiment uses tracking data of football games to form an analysis summary of the players' energy metabolism throughout the game, running speed distribution, running distance in different speed intervals, ball holding time, and running hot spots, so as to evaluate the players' personal physical condition and enthusiasm in the game, and form a radar chart of related events in different dimensions of the players based on the event data, so as to realize the behavioral analysis of the players' specific styles.

[0076] Step S104, based on the pre-processed event data, analyzing each player's individual passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in the game.

[0077] Specifically, the second analysis module (high-order analysis module) in the following football player game analysis device is used to analyze each player's personal passing style based on different types of events, each player's performance behavior score in different activity positions on the court, and each player's collective contribution value in the game based on the preprocessed event data.

[0078] The high-level analysis module in the disclosed embodiment uses event data and tracking data of football games to realize the mining of players' personal styles in handling different events, scoring of different positions, and comprehensive analysis of their contributions to the team.

[0079] Step S105, based on the game energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, radar distribution data, each player's personal passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in this game, generates an analysis report on the target theme football game.

[0080] Through the basic analysis module and advanced analysis module mentioned above, various player-related data are analyzed to generate analysis reports based on the target theme football games, so as to provide a comprehensive and multi-dimensional evaluation and analysis method, thereby fully tapping the potential of young players and training players in a more scientific way.

[0081] like Figure 2 As shown, it is a schematic diagram of analyzing various data of multiple players through basic analysis modules and advanced analysis modules.

[0082] In an optional embodiment, the tracking data, event data and match data are preprocessed, including: aligning, filtering and cleaning the tracking data, event data and match data of the same match of the target theme football match in sequence according to the date and time of the match.

[0083] By aligning, filtering, cleaning and pre-processing the tracking data, event data and game data to ensure data accuracy, it is helpful to accurately analyze the personal abilities, game enthusiasm, personal physical condition, personal passing style, collective contribution value and performance behavior scores of multiple players.

[0084] In this embodiment, a football player game analysis method is provided, which can be used in mobile terminals, such as mobile phones, tablet computers, etc. Figure 3 is a flow chart of a method for analyzing a soccer player match according to an embodiment of the present invention. Figure 3 As shown, according to the date and time of the game, the tracking data, the event data and the game data of the same game of the target theme football game are aligned, filtered, cleaned and pre-processed in sequence, including the following steps:

[0085] Step S301, obtaining the player list, jersey numbers, and home and away team logos in the game information.

[0086] Specifically, the game information comes from game data, and the game information is specific information recorded according to the date and time of the game, and the specific information includes a player list, jersey numbers, and a home team logo and a visiting team logo.

[0087] Step S302, aligning and unifying the fields of player list, jersey number, home and away team identification in the tracking data and event data.

[0088] Specifically, the player list, jersey number, and home and away team identification information in the game information data are used to align and unify the tracking data with the corresponding fields in the event data.

[0089] Step S303: filtering out-of-order frames from the tracking data based on the number of data frames.

[0090] Specifically, because the tracking data collected by wearable devices with GPS or UWB have problems such as unstable data collection frequency, frame loss, and disorder, the obviously erroneous and disordered frames can be eliminated according to the number of data frames.

[0091] Step S304 , filtering out-of-order frames of the tracking data based on the player position coordinates in the tracking data.

[0092] Specifically, out-of-order frames with obvious jumps are eliminated based on the player position coordinates. For example, if the position of one or more players shows obvious jumps in consecutive frames (not consistent with the players' continuous running), such data will be eliminated; and the required tracking data frequency is determined.

[0093] Step S305, fixing the filtered tracking data according to a preset frame number.

[0094] Specifically, according to the required tracking data frequency, that is, if the number of frames collected in one second is greater than 10 frames, 10 frames with complete information are randomly selected for fixing. Complete information means that this frame contains records of multiple (22) players in the field, the position and speed of the ball, and the ownership of the ball; if the number of frames collected in one second is less than 10 frames, linear difference is used for data enhancement to complete it to 10 frames per second. The preset number of frames in the embodiment of the present disclosure is determined in combination with actual needs. Therefore, the preset number of frames can also be 35 frames, and the fixing method is the same as the above-mentioned 10-frame fixing method.

[0095] Step S306, cleaning the fixed tracking data.

[0096] Specifically, after the tracking data is fixed according to a preset number of frames, if it is fixed at 10 Hz, the data needs to be cleaned at 10 frames per second.

[0097] Step S307 , converting the event data into tracking data of a corresponding number of frames according to the event time recorded in each frame of the event data.

[0098] Step S308, locating the corresponding tracking data segment according to the tracking data converted from the event data to the corresponding frame number, and comparing the position of the player in the event data with the position of the player in the tracking data to determine whether there is any obvious abnormality in the record of the event data.

[0099] Specifically, for the processing of outlier values ​​of event data, we first convert the event time recorded in each frame of the event data into the corresponding number of tracking data frames, then locate the corresponding tracking data segment, and compare the position of the player in the event data with the position of the player in the tracking data to determine whether there are obvious anomalies in the record of event data.

[0100] Step S309: If the position of the player in the event data and the position of the player in the tracking data differ within a preset range, it is determined that the record in the event data is normal.

[0101] Step S310: If the event data exceeds a preset range, the record in the event data is determined to be abnormal, and the abnormal data is removed.

[0102] For example, the preset range may be within plus or minus 10 meters to determine whether an abnormality has occurred. For example, if the player position (horizontal and vertical coordinates) between the event data and the tracking data differ within plus or minus 10 meters, it can be determined that there is no abnormality. If it exceeds plus or minus 10 meters, it is determined that this event data record is abnormal and needs to be removed. In this way, the tracking data and event data of multiple games can be obtained after preprocessing.

[0103] By aligning, filtering, cleaning and pre-processing the tracking data, event data and game data to ensure data accuracy, it is helpful to accurately analyze the personal abilities, game enthusiasm, personal physical condition, personal passing style, collective contribution value and performance behavior scores of multiple players.

[0104] In this embodiment, a football player game analysis method is provided, which can be used in mobile terminals, such as mobile phones, tablet computers, etc. Figure 4 is a flow chart of a method for analyzing a soccer player match according to an embodiment of the present invention. Figure 4 As shown, the above step S103 analyzes the game energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, and radar distribution data of each player based on the pre-processed tracking data and event data, including:

[0105] Step S1031, based on the speed of each player at each moment in the preprocessed tracking data, calculate the acceleration of each player at each moment.

[0106] Specifically, since the tracking data contains the time of each player at each moment and the speed corresponding to that moment, the formula The acceleration of each player at each moment can then be calculated.

[0107] Step S1032: presenting the energy metabolism data of the players based on the acceleration of each player at each moment, and plotting the values ​​at each moment to form a first data graph.

[0108] Specifically, the energy metabolism of the players is presented using acceleration to reflect the physical condition of the players. The presentation form can be a line graph formed by the values ​​at each moment, or a bar graph formed by averaging over a period of time, etc. The line graph and bar graph are the first data graph mentioned above.

[0109] Step S1033, based on the pre-processed tracking data, the pre-processed tracking data is screened and counted according to low-speed, medium-speed, and high-speed running, and the running distances in different speed intervals are calculated to analyze the running abilities of different players.

[0110] Specifically, slow, medium and high speeds can be based on the definitions of different coaches. According to the definitions of low, medium and high speed running by different coaches, the tracking data can be screened and counted, and the running distances in different speed ranges can be calculated to analyze the running abilities of different players.

[0111] Step S1034, based on the pre-processed tracking data, calculate the ball possession time of the home and away teams in the first and second halves or different stages of the game according to the ball possession field, and draw a second data graph; or, locate the ball possession team through the ball possession mark of each frame, and judge the player currently in possession of the ball through the distance between the team's players and the ball. If the ball is passed by any player, the possession right still belongs to the passing player, then calculate the ball possession time of each player based on the time before the next event occurs and the starting time of the passing player's possession.

[0112] Specifically, based on the preprocessed tracking data, the ball_control field can be used to calculate the ball possession time of the home and away teams in the first and second half or different stages of the game, and presented in the form of a pie chart or a bar chart, which is the second data graph.

[0113] Alternatively, the ball-controlling team may be first located through the ball-controlling mark of each frame, and then the player currently controlling the ball may be determined through the distance between the player of this team and the ball. If the ball is passed by a certain player, the ball control right before the next event (interception, reception, and other related events) still belongs to the passing player. That is, the ball control time of each player is calculated based on the ball control start time and the time before the next event occurs, thereby performing a comparative analysis of the ball control abilities of different players. The presentation method may be a pie chart or a bar chart, which is the second data graph mentioned above.

[0114] Step S1035, obtaining the pre-processed tracking data, presenting the running hotspots of each player in the form of a heat map, and then performing differential analysis on players of different positions and styles.

[0115] Based on the pre-processed tracking data, the running hot spots of each player are presented in the form of a heat map, and then differential analysis is performed on players of different positions and styles.

[0116] Step S1036, based on the comparison requirements of different coaches for the multi-dimensional abilities of players, radar distribution data of different abilities of the same player are formed. The radar distribution data includes any player's physical abilities such as maximum acceleration, maximum speed, maximum consumption load, ball possession time, or the number of passes, successful passes, interceptions, and shots.

[0117] Specifically, based on the comparison needs of different coaches for the multi-dimensional abilities of players, a radar distribution calculation of the different abilities of the same player is formed, such as a player's physical abilities such as maximum acceleration, maximum speed, maximum metabolic energy, ball possession time, or his technical abilities such as the number of passes, number of successful passes, number of interceptions, number of shots, etc.; it is also possible to form a comparison of the abilities of different players in one or more dimensions, such as a comparison of the running, passing, interception and other abilities of several players on the home team.

[0118] Among them, the maximum metabolic energy can use the player speed recorded in the tracking data to calculate the energy consumption of each player, reflect the physical fitness differences between different players in the game, and compare the consumption of exercise under different speed changes. For example, the maximum metabolic energy is calculated by the following formula.

[0119] a>0,cost=0.102×((a 2 +96.2) 0.5 )×(4.03×a+3.6×e (-0.408)×a );

[0120] a<0,cost=0.102×((a 2 +96.2) 0.5 )×((-0.85)×a+3.6×e 1.33×a );

[0121] a=0,cost=0. Where a is acceleration and cost is maximum metabolic energy.

[0122] In addition, the maximum acceleration and maximum speed can be calculated through the speed in the tracking data, the ball possession time is calculated by the method mentioned in the above step S1034, and the number of passes, number of successful passes, number of interceptions, and number of shots can be obtained through event data.

[0123] The disclosed embodiment executes the above-mentioned steps S1031 to S1036, and uses the tracking data of the football game to form an analysis summary of the players' energy metabolism, running speed distribution, running distance in different speed intervals, ball holding time, and running hot spots throughout the game, so as to evaluate each player's personal physical condition and enthusiasm in the game, and form a radar chart of related events of different dimensions of the players based on the event data, so as to realize the behavioral analysis of each player's specific style.

[0124] In this embodiment, a football player game analysis method is provided, which can be used in mobile terminals, such as mobile phones, tablet computers, etc. Figure 5 is a flow chart of a method for analyzing a soccer player match according to an embodiment of the present invention. Figure 5 As shown, the above step S104, based on the pre-processed event data, analyzes each player's individual passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in the game, including:

[0125] Step S1041 : clustering different types of passing sub-events using non-negative matrix factorization.

[0126] Specifically, using non-negative matrix factorization, different types of passing sub-events are defined according to the coach's analysis requirements for style types. For example, passing sub-events are divided into dangerous passes, key passes, and defensive line passes, forming a classification of the locations where different events occur. Figure 6 As shown, it is a schematic diagram of the passing sub-event, and the red area indicates the passing position.

[0127] Step S1042, calculating the distribution weight of each player in each passing sub-event to evaluate the personal style of each player.

[0128] Specifically, based on the formed classifications of the locations where different events occurred, the distribution weights of each player in these categories are calculated, thereby evaluating the player's personal style.

[0129] Step S1043, generating a visualization display diagram of each player's personal style based on the coach's needs.

[0130] Specifically, different visualization displays are formed according to the needs of the coach, and personal style visualization displays include heat maps or scatter plots.

[0131] Step S1044, based on the pre-processed event data, classify the activity positions of multiple players on the court.

[0132] Specifically, based on the pre-processed event data, the activity positions of multiple players on the court are classified. Figure 7As shown, this is a schematic diagram of the division results obtained by clustering all activity positions of a certain player in a certain cup game.

[0133] Step S1045 , giving performance scores of players at different positions based on the event types and results of different activity positions.

[0134] Specifically, performance scores of multiple players in different positions are given based on the event types and results in different positions (e.g., whether the pass is successful or not, whether the dribble is intercepted), so as to explore the players' strengths and weaknesses.

[0135] Step S1046, based on the pre-processed event data, taking the shot or goal event as the basis, trace back the entire pass chain before the shot, calculate the expected goal value or expected threat value of each player in the pass chain to evaluate the contribution of each player to the shot or goal event, and combine the expected goal value or expected threat value of each player in the entire game to measure the contribution of each player to the team in this game.

[0136] Specifically, based on the preprocessed event data, taking the shot (missed) or goal event as the basis, the entire pass chain before the shot is traced back, and the expected goal value xG or expected threat value xT of each player in the pass chain is calculated to evaluate the contribution of each player to this shot (missed) or goal event. The xG value or xT value of each player in the entire game is combined and counted to measure the contribution of each player to the team in this game.

[0137] Step S1047, evaluating the collective contribution value of the players to the target theme football game by counting the number of times the expected goal value or expected threat value ranking appears in the top five in multiple games, and generating a third data graph.

[0138] Specifically, the collective contribution of the players to the tournament or cup is evaluated by counting the number of times the xG or xT value ranking appears in the top five in multiple games, and the presentation is in the form of a pie chart or a bar chart, i.e. the third data chart mentioned above.

[0139] The disclosed embodiment executes the above-mentioned steps S1041 to S1047, and uses the event data and tracking data of the football game to realize the mining of the personal style of players in handling different events, the scoring of different positions, and the comprehensive analysis of their contribution to the team, which is conducive to fully tapping the potential of players and training players in a more scientific way.

[0140] In an optional implementation, the football player game analysis method in the disclosed embodiment further includes: acquiring video data of multiple players participating in a target theme football game to assist in the review of key segments after the game.

[0141] Specifically, the game video can be a full-view game video or a tactical-view game video, which is used to assist in the review of key segments after the game and to conduct comparative analysis with the visualization of data segments.

[0142] In the present embodiment, a football player game analysis device is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0143] This embodiment provides a football player game analysis device, such as Figure 8 As shown, including:

[0144] The data acquisition module 81 is used to acquire tracking data, event data and match data of multiple players participating in the target theme football game; wherein the tracking data records the positions and speeds of multiple players and the ball on the field, as well as the team that has possession of the ball; the event data records the entire process of the entire game in time series; and the match data records the match information according to the date and time of the game;

[0145] A data processing module 82 for pre-processing tracking data, event data and game data;

[0146] A first analysis module 83 is used to analyze the game energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, and radar distribution data of each player based on the pre-processed tracking data and event data;

[0147] A second analysis module 84 is used to analyze each player's individual passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in the game based on the pre-processed event data;

[0148] The report generation module 85 is used to generate an analysis report of the target theme football match based on the game energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, radar distribution data, each player's personal passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in this game.

[0149] In an optional implementation, the data processing module 82 includes:

[0150] The data processing submodule is used to align, filter, clean and pre-process the tracking data, event data and game data of the same game of the target theme football game in sequence according to the date and time of the game.

[0151] In an optional implementation, the data processing submodule includes:

[0152] A data acquisition unit, used to acquire the player list, jersey numbers, home and away team logos in the game information;

[0153] A data alignment unit is used to align the fields of player lists, jersey numbers, and home and away team identification in tracking data and event data;

[0154] A first filtering unit, configured to filter out erroneous and out-of-order frames from the tracking data based on the number of data frames;

[0155] a second filtering unit, configured to filter out-of-order frames from the tracking data based on the player position coordinates in the tracking data;

[0156] A data fixing unit, used to fix the filtered tracking data according to a preset frame number;

[0157] A data cleaning unit, used to clean the tracking data after fixation;

[0158] A data tracking unit, used for converting the event data into tracking data of a corresponding number of frames according to the event time recorded in each frame of the event data;

[0159] The first recording unit is used to locate the corresponding tracking data segment according to the tracking data converted from the event data into the corresponding number of frames, and compare the position of the player in the event data with the position of the player in the tracking data to determine whether there is an obvious abnormality in the record of the event data:

[0160] A second recording unit, for abnormality processing, determines that the record in the event data is normal if the difference between the position of the player in the event data and the position of the player in the tracking data is within a preset range;

[0161] The abnormality elimination unit is used to determine that the record in the event data is abnormal if the event data exceeds a preset range, and eliminate the abnormal data.

[0162] In an optional implementation, the first analysis module 83 includes:

[0163] An acceleration data calculation submodule, for calculating the acceleration of each player at each moment based on the speed of each player at each moment in the preprocessed tracking data;

[0164] A first data graph generating submodule, for presenting the energy metabolism data of the players based on the acceleration of each player at each moment, and plotting the values ​​at each moment to form a first data graph;

[0165] The running ability data analysis submodule is used to screen and count the pre-processed tracking data according to low-speed, medium-speed and high-speed running, calculate the running distance in different speed intervals, and analyze the running ability of different players;

[0166] The player possession time calculation submodule is used to calculate the possession time of the home and away teams in the first and second halves or different stages of the game according to the possession field based on the pre-processed tracking data, and draw a second data graph; or, locate the ball-controlling team through the ball-control mark of each frame, and judge the player currently in possession of the ball through the distance between the player of the team and the ball. If the ball is passed by any player, the possession right still belongs to the passing player, then the possession time of each player is calculated based on the time before the next event occurs and the starting time of the passing player's possession of the ball;

[0167] The player data difference analysis submodule is used to obtain the pre-processed tracking data and present each player's running hotspot in the form of a heat map, thereby performing a differential analysis of players of different positions and styles;

[0168] The radar distribution data generation submodule is used to form radar distribution data of different abilities of the same player according to the comparison requirements of different coaches for the multi-dimensional abilities of the players. The radar distribution data includes any player's physical abilities such as maximum acceleration, maximum speed, maximum consumption load, ball possession time, or the number of passes, successful passes, interceptions, and shots.

[0169] In an optional implementation, the second analysis module 84 includes:

[0170] A data clustering submodule is used to cluster different types of passing sub-events using non-negative matrix factorization;

[0171] A style evaluation submodule is used to calculate the distribution weight of each player in each passing sub-event to evaluate the personal style of each player;

[0172] The data display submodule is used to generate a visual display of each player's personal style based on the coach's needs;

[0173] A data classification submodule, for classifying the activity positions of multiple players on the court based on the preprocessed event data;

[0174] The data scoring submodule is used to give the performance scores of players in different positions based on the event types and results of different activity positions;

[0175] The player contribution value calculation submodule is used to, based on the pre-processed event data and the shot or goal event, trace back the entire pass chain before the shot, calculate the expected goal value or expected threat value of each player in the pass chain, so as to evaluate the contribution value of each player to the shot or goal event, and combine the expected goal value or expected threat value of each player in the whole game to measure the contribution value of each player to the team in the game;

[0176] The data graph generation submodule is used to evaluate the collective contribution value of the players to the target theme football game by counting the number of times the expected goal value or expected threat value ranking appears in the top five in multiple games, and generate a third data graph.

[0177] In an optional implementation, the football player game analysis method in the embodiment of the present disclosure further includes:

[0178] The video data acquisition submodule is used to acquire video data of multiple players participating in a target theme football match to assist in the review of key segments after the match.

[0179] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0180] The football player game analysis device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0181] An embodiment of the present invention further provides a computer device having the above-mentioned football player game analysis device.

[0182] See also Fig. 9 , Fig. 9 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Fig. 9As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Fig. 9 A processor 10 is taken as an example.

[0183] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0184] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0185] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0186] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0187] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0188] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0189] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0190] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for analyzing a football player's match, characterized in that: The method comprises: Obtaining tracking data, event data, and match data of multiple players participating in a target theme football match; wherein the tracking data records the positions and speeds of the multiple players and the ball on the field, as well as the team that has possession of the ball; the event data records the entire process of the entire match in time series; and the match data records match information according to the date and time of the match; pre-processing the tracking data, the event data, and the game data; Based on the pre-processed tracking data and event data, analyzing each player's game energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, and radar distribution data; Based on the pre-processed event data, analyzing each player's individual passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in the game; An analysis report of the target theme football match is generated based on the game energy metabolism data, the running speed distribution data, the different speed running interval data, the ball holding time data, the running hot zone data, the radar distribution data, each player's personal passing style based on different types of events, each player's performance behavior score in different activity positions on the court, and each player's collective contribution value in this game.

2. The method according to claim 1, characterized in that Preprocessing the tracking data, the event data, and the game data includes: According to the date and time of the game, the tracking data, the event data and the game data of the same game of the target theme football game are aligned, filtered, cleaned and pre-processed in sequence.

3. The method according to claim 2, characterized in that According to the date and time of the game, the tracking data, the event data and the game data of the same game of the target theme football game are aligned, filtered and cleaned in sequence, including: Obtain the player list, jersey numbers, and home and away team logos in the game information; Aligning and unifying the fields of the player list, the jersey number, and the home and away team identification in the tracking data and the event data; Filtering erroneous and out-of-sequence frames from the tracking data based on the number of data frames; filtering out-of-order frames of the tracking data based on player position coordinates in the tracking data; Fixing the filtered tracking data according to a preset frame number; cleaning and fixing the tracking data; According to the event time recorded in each frame of the event data, converting the event data into tracking data of a corresponding number of frames; According to the tracking data converted from the event data into the corresponding number of frames, the corresponding tracking data segment is located, and the position of the player in the event data is compared with the position of the player in the tracking data to determine whether there is an obvious abnormality in the record of the event data: If the position of the player in the event data and the position of the player in the tracking data differ from each other within a preset range, then determining that the record in the event data is normal; If the event data exceeds a preset range, the record in the event data is determined to be abnormal, and the abnormal data is removed.

4. The method according to claim 1, characterized in that: Analyze each player's game energy metabolism data, running speed distribution data, running interval data at different speeds, ball holding time data, running hot zone data, and radar distribution data, including: Calculate the acceleration of each player at each moment based on the speed of each player at each moment in the preprocessed tracking data; Based on the acceleration of each player at each moment, present the energy metabolism data of the player, and plot the value at each moment to form a first data graph; Based on the pre-processed tracking data, the pre-processed tracking data is screened and counted according to low-speed, medium-speed, and high-speed running, and the running distances in different speed intervals are calculated to analyze the running abilities of different players; Based on the pre-processed tracking data, the ball possession time of the home and away teams in the first and second halves or different stages of the game is calculated according to the ball possession field, and a second data graph is drawn; or, the ball possession team is located by the ball possession mark of each frame, and the player currently in possession of the ball is determined by the distance between the player of the team and the ball. If the ball is passed by any player, the possession right still belongs to the passing player, then the ball possession time of each player is calculated based on the time before the next event occurs and the ball possession start time of the passing player; Obtain the pre-processed tracking data, present each player's running hotspot in the form of a heat map, and then conduct differential analysis on players of different positions and styles; According to different coaches' requirements for comparing players' multi-dimensional abilities, radar distribution data of different abilities of the same player is formed, including any player's physical abilities such as maximum acceleration, maximum speed, maximum consumption load, ball possession time, or the number of passes, successful passes, interceptions, and shots.

5. The method according to claim 1, characterized in that Based on the pre-processed event data, analyze each player's individual passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in the game, including: Clustering different types of passing sub-events using non-negative matrix factorization; Calculate the distribution weight of each player in each passing sub-event to evaluate each player's individual style; Generate a visual representation of each player's personal style based on the coach's needs; Based on the pre-processed event data, classifying the activity positions of the multiple players on the court; Give players performance ratings in different positions based on the event types and outcomes of different active positions; Based on the pre-processed event data, taking the shot or goal event as the basis, the entire pass chain before the shot is traced back to calculate the expected goal value or expected threat value of each player in the pass chain to evaluate the contribution of each player to the shot or goal event. The expected goal value or expected threat value of each player in the entire game is combined and counted to measure the contribution of each player to the team in this game. The collective contribution value of the players to the target-themed football game is evaluated by counting the number of times the expected goal value or expected threat value ranking appears in the top five in multiple games, and a third data graph is generated.

6. The football player game analysis method according to any one of claims 1 to 5, characterized in that: Also includes: Obtain video data of multiple players participating in a target-themed football match to assist in post-match review of key segments.

7. A soccer player game analysis device, characterized in that: The device comprises: A data acquisition module, for acquiring tracking data, event data and match data of multiple players participating in a target theme football match; wherein the tracking data records the positions and speeds of the multiple players and the ball on the field, as well as the team that has possession of the ball; the event data records the entire process of the entire match in time series; and the match data records match information according to the date and time of the match; A data processing module, used for pre-processing the tracking data, the event data and the game data; A first analysis module is used to analyze the game energy metabolism data, running speed distribution data, different speed running interval data, ball holding time data, running hot zone data, and radar distribution data of each player based on the pre-processed tracking data and the event data; A second analysis module is used to analyze, based on the pre-processed event data, each player's individual passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in the game; A report generation module is used to generate an analysis report of the target theme football match based on the game energy metabolism data, the running speed distribution data, the different speed running interval data, the ball holding time data, the running hot zone data, the radar distribution data, each player's personal passing style based on different types of events, each player's performance behavior score at different activity positions on the court, and each player's collective contribution value in this game.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the football player game analysis method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the football player game analysis method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the football player game analysis method according to any one of claims 1 to 6.

Images