CN118356624A - Football point analysis method, device, equipment, system and storage medium - Google Patents

Abstract

The present disclosure provides a football goal analysis method, device, equipment, system and storage medium. The method comprises the following steps: responding to the received analysis request, acquiring historical ballpoint record information of the target attack party corresponding to the identification of the target attack party, and then extracting body action parameters corresponding to the set action key indexes from body action information corresponding to each ballpoint in the historical ballpoint record information based on the set action key indexes; and then analyzing and processing the position information of the initial standing of the target attack party, the motion track information of the position of the football, which are corresponding to each goal, and the extracted body action parameters, determining a reference index group for identifying the shot position information of the football in the goal when the target attack party is in the goal, and then generating a defending training scheme. The method solves the problem of insufficient accuracy of ball-point analysis in the prior art, and improves the ball-point defending capability and defending effect of goalkeepers.

Description

Football point analysis method, device, equipment, system and storage medium

Technical Field

The disclosure relates to the technical field of motion analysis, in particular to a football point analysis method, a football point analysis device, football point analysis equipment, football point analysis system and a football point analysis storage medium.

Background

With the development of data analysis technology, strategies in sports are optimized through data analysis, and further, the sports level is improved and widely applied. In various fields such as basketball, football, volleyball, puck and the like, the performance of athletes is evaluated through data analysis, and a game strategy is determined, so that the game is becoming a mainstream trend gradually. In the football field, because the ball-pointing links involve relatively few variables, the data difficulty is low, and the ball-pointing strategy and the ball-pointing performance of a player are evaluated through data analysis, much attention is paid.

In the prior art, the strategy analysis of the shot is mainly based on video analysis, and still depends on experience to judge, so that the reliability and practicality of the analysis result are insufficient, and the shot is difficult to convert into an actual scene for application.

Disclosure of Invention

The disclosure provides a football goal analysis method, device, equipment, system and storage medium, so as to solve the problem of insufficient goal analysis accuracy.

In a first aspect, the present disclosure provides a football goal analysis system, comprising:

The attack input module is used for inputting the goal strategy information of the target attack party or selecting the pre-stored history goal record information of the target attack party, wherein the history goal record information is obtained based on the goal strategy information of the target attack party, and the goal strategy information comprises the actual shooting position information and shooting mode of the football in the goal;

the analysis module is used for determining estimated injection position information of the target attack party according to the set action key index and the ball-point strategy information or the historical ball-point record information.

In a second aspect, the present disclosure provides a football goal analysis method, including:

Responding to the received analysis request, and acquiring historical dribbling record information of the target attack party corresponding to the identification of the target attack party; the analysis request comprises an identification of a target attack party to be analyzed, the historical ball striking record information is obtained based on ball striking strategy information of the target attack party, and the ball striking strategy information at least comprises: each time a ball is spotted, the initial standing position information of a target attack party, the motion track information and the body action information of the position of the football, and the actual shooting position information of the football in the goal;

Extracting body action parameters corresponding to the set action key indexes from body action information corresponding to each time of ball striking in the history ball striking record information based on the set action key indexes;

Analyzing and processing the position information of the initial standing of the target attack party, the motion track information of the position of the football, which are corresponding to each ball point, and the extracted body action parameters, and determining a reference index set for identifying the shot position information of the football in the goal when the target attack party points the ball, so as to determine the estimated shot position information of the target attack party based on the reference index set;

and generating a defending training scheme based on the reference index set and the pre-estimated injection position information.

In a third aspect, the present disclosure provides a football goal analysis device, the football goal analysis device comprising:

The acquisition module is used for responding to the received analysis request and acquiring the history ball record information of the target attack party corresponding to the identification of the target attack party; the analysis request comprises an identification of a target attack party to be analyzed, the historical ball striking record information is obtained based on ball striking strategy information of the target attack party, and the ball striking strategy information at least comprises: each time a ball is spotted, the initial standing position information of a target attack party, the motion track information and the body action information of the position of the football, and the actual shooting position information of the football in the goal;

The extraction module is used for extracting body action parameters corresponding to the set action key indexes from body action information corresponding to each time of ball striking in the history ball striking record information based on the set action key indexes;

the analysis module is used for analyzing and processing the corresponding initial standing position information of the target attack party, the motion track information of the position of the football, which is moved to the position of the football, and the extracted body action parameters when each ball is shot, and determining a reference index group for identifying the shot position information of the football in the goal when the target attack party is shot, so as to determine the estimated shot position information of the target attack party based on the reference index group;

and the processing module is used for generating a defending training scheme based on the reference index group and the estimated injection position information.

In a fourth aspect, the present disclosure also provides an electronic device, including:

at least one processor;

and a memory communicatively coupled to the at least one processor;

The memory stores instructions executable by the at least one processor to cause the electronic device to perform a football goal analysis method as in accordance with any embodiment of the first aspect of the present disclosure.

In a fifth aspect, the present disclosure also provides a computer-readable storage medium having stored therein computer-executable instructions which, when executed by a processor, are adapted to carry out a football goal analysis method as in any one of the first aspects of the present disclosure.

In a sixth aspect, the present disclosure also provides a computer program product containing computer-executable instructions for implementing a football goal analysis method as in any one of the first aspects of the present disclosure when executed by a processor.

According to the football goal analysis method, the device, the equipment, the system and the storage medium, the historical goal record information of the target attack party corresponding to the identification of the target attack party is obtained by responding to the received analysis request, and then the body action parameters corresponding to the set action key indexes in the body action information corresponding to each goal in the historical goal record information are extracted based on the set action key indexes; and then analyzing and processing the position information of the initial standing of the target attack party, the motion track information of the position of the football, which are corresponding to each ball point, and the extracted body action parameters, determining a reference index group for identifying the shot position information of the football in the goal when the target attack party points, and then generating a defending training scheme based on the reference index group and the estimated shot position information. Therefore, the relation between the action, the initial standing position and the action track of the target attack party and the corresponding football shooting position can be analyzed and proposed through the history ball shooting record information of the ball shooting of the target attack party, so that the goalkeeper can rapidly identify the action and the position characteristics of the target attack party according to the relation, and further rapidly and accurately judge the football shooting position of the target attack party, further the football shooting analysis accuracy and the training efficiency are improved, and the ball shooting defending capability and the defending effect of the goalkeeper are improved.

Drawings

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

Fig. 1 is an application scenario diagram of the football point ball analysis method provided in the embodiment of the disclosure in fig. 1;

FIG. 2a is a schematic diagram of a football goal analysis system according to an embodiment of the present disclosure;

FIG. 2b is a schematic diagram of an interface for inputting the ball-in-ball strategy information in the football ball-in-ball analysis system according to the embodiment shown in FIG. 2 a;

FIG. 2c is a schematic diagram of an interface for inputting ball-in-ball strategy information in another football ball-in-ball analysis system according to the embodiment shown in FIG. 2 a;

FIG. 2d is a schematic diagram of an interface for outputting a training scenario in the football goal analysis system according to the embodiment shown in FIG. 2 a;

FIG. 2e is a schematic diagram of an interface for inputting defense strategy information in the football goal analysis system according to the embodiment shown in FIG. 2 a;

FIG. 3 is a flow chart of a method for analyzing football at a football goal according to another embodiment of the present disclosure;

FIG. 4a is a flow chart of a method for analyzing a football goal according to yet another embodiment of the present disclosure;

FIG. 4b is a flowchart of a method for determining a reference index set according to the embodiment shown in FIG. 4 a;

Fig. 5 is a schematic structural diagram of a football spot ball analysis device according to another embodiment of the present disclosure;

fig. 6 is a schematic structural diagram of an electronic device according to another embodiment of the present disclosure.

Specific embodiments of the present application have been shown by way of the above drawings and will be described in more detail below. The drawings and the written description are not intended to limit the scope of the inventive concepts in any way, but rather to illustrate the inventive concepts to those skilled in the art by reference to the specific embodiments.

Detailed Description

Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with embodiments of the application. Rather, they are merely examples of apparatus and methods consistent with aspects of embodiments of the application as detailed in the accompanying claims.

The following describes in detail the technical solutions of the embodiments of the present application and how the technical solutions of the embodiments of the present application solve the above technical problems with specific embodiments. The following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.

With the development of data analysis technology, strategies in sports are optimized through data analysis, and further, the sports level is improved and widely applied. In various fields such as basketball, football, volleyball, puck and the like, the performance of athletes is evaluated through data analysis, and a game strategy is determined, so that the game is becoming a mainstream trend gradually. In the football field, training athletes through data analysis, providing the technical and tactical level and competition ability of athletes, is also becoming a popular research direction. The technical and tactical characteristics of the opponents are researched through data analysis, so that a training scheme and a competition strategy are better formulated, and the method becomes a main research direction.

Because the ball-pointing links involve relatively few variables, the ball-pointing strategy and the ball-pointing performance of the player are evaluated through data analysis, and much attention is paid. However, the analysis of the existing goal is mainly based on the assessment of the overall defending ability of the goalkeeper (goalkeeper) and the overall goal level of the offender, and related researches on how the goalkeeper performs targeted training on different offenders are lacking.

In addition, the determination of the football goal strategy is also limited by how to select goalkeepers to ensure the goal defense efficiency. The goal of pertinently improving the goal-guard ability of the goalkeeper and the defense effect of specific offenders cannot be achieved, so that the practicability of the goal-guard analysis is seriously insufficient.

In the prior art, the strategy analysis of the ball point is mainly based on video analysis, and still depends on experience to judge, so that the reliability and the practicability of the analysis result are insufficient, and the method cannot be practically applied to the ball point field.

In order to solve the above problems, the embodiment of the application provides a football goal analysis method, which specifically subdivides the goal strategy information of an attacker into a goal shooting mode and a position expected to be shot, predicts the shot position according to the goal shooting mode, and determines a training scheme according to the determined relation between the goal shooting mode and the expected shot position, so that a goalkeeper can predict a position needing to be defended according to the goal shooting mode and the expected shot position of the attacker quickly through training, thereby providing a goal defending success rate and improving the goal training effect.

Fig. 1 is an application scenario diagram of a football goal analysis method according to an embodiment of the present application. As shown in fig. 1, in the process of performing the football goal analysis, the server 100 acquires the input historical goal record information 110 of the target attack party and/or the historical defending record information 120 of the target goalkeeper, determines the training scheme and tactical strategy scheme 130 of the goal, and completes the football goal analysis.

It should be noted that, in the scenario shown in fig. 1, the server, the historical goal record information of the target attack party, the historical defense record information of the target goalkeeper, and the training scheme and tactical scheme of the goal are only illustrated by taking one example as an example, but the embodiment of the present application is not limited thereto, that is, the number of the server, the historical goal record information of the target attack party, the historical defense record information of the target goalkeeper, the training scheme and tactical scheme of the goal may be arbitrary.

The football point ball analysis method provided by the application is described in detail by specific examples. It should be noted that the following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

Fig. 2a is a schematic structural diagram of a football point ball analysis system according to an embodiment of the application. As shown in fig. 2a, the soccer ball analysis system 200 includes:

the attack input module 210 is configured to input goal strategy information of a target attack party or select pre-stored history goal record information of the target attack party, where the history goal record information is obtained based on the goal strategy information of the target attack party, and the goal strategy information includes actual shooting position information and shooting mode of a football in a goal;

The analysis module 220 is configured to determine estimated injection position information of the target attack party according to the set action key index and the ball-point strategy information or the historical ball-point record information.

Specifically, the target attack party, i.e. the person executing the goal, may be a specific player (e.g. for making a goal strategy for a specific opponent in advance) according to the analysis purpose, or may be a virtual person generated by adjusting on the basis of the historical goal record information of the player (e.g. for evaluating possible technical changes after the specific player is injured, obtained by adjusting the data in the historical goal record information), or may be a non-actual player fitted by data (e.g. when the defending effect or judging accuracy of a goalkeeper for a certain goal mode, the data corresponding to the goal mode may be fitted, so as to obtain the non-actual player fitted by data).

The attack input module 210 may be configured to enter the goal attack party's shot strategy information, a set of goal attack party's shot strategy information over a certain period of time (e.g., the past year), i.e., historical shot record information.

The goal strategy information is information obtained through collected goal data of a target attack party, and the goal data is usually video records or picture data of each goal so as to specifically identify information such as a shooting mode, an actual shooting position and the like of the target attack party during the goal. Since the habit of a player in playing a ball is usually kept constant for a long time, the history of the ball may include a long time span (e.g., five years of data) to improve accuracy of identification when analyzed for a particular player.

In actual competition, the personnel who play the ball, usually the personnel who are good at the ball in the team, the history ball record information of such personnel is usually more, the data is more, the sample size is larger, therefore, the reliability and the accuracy of the analysis result can be better ensured by analyzing the history ball record information of such personnel.

The goal shooting mode of the goal comprises standing positions, kicking actions and the like of the target attack party, and generally, when the same target attack party points the goal (namely, the actual shooting position of the football) at different positions in the goal, the goal shooting mode has certain difference, the goal shooting mode is split into different actions and positions (namely, different action key indexes and body action parameters corresponding to the action key indexes or summarized into different goal shooting modes), the relevance of the actions and the positions to the actual shooting positions is identified, and the actions and the positions with high relevance to the corresponding actual shooting positions are extracted, so that a goalkeeper can quickly judge the actual shooting positions in the goal which the football can reach when identifying the corresponding actions and positions, and further, accurate and effective defense is performed.

The actual shooting position does not refer to a position entering the interior of the goal, but is a position intersecting the vertical plane of the goal during the shooting of the football into the goal, or a position where the football is caught (or a position projected onto the vertical plane of the football, because the football is caught at this time, and cannot enter the goal). By determining the actual injection position corresponding to each shooting mode, the relation between the positions (which can be called as the estimated injection positions when no actual shooting is performed) of the goal is expected to be shot corresponding to the shooting mode can be conveniently established, so that the subsequent analysis and the application of training the goalkeeper and the like can be facilitated.

The analysis module 220 is connected to the attack input module 210, and the analysis module 220 is configured to perform a function of identifying a shot mode and an actual shot position from shot strategy information or historical shot record information, and establishing a correlation between the shot mode and the actual shot position. The analysis module 220 may further extract a reference index set corresponding to the corresponding estimated injection position from the goal shooting mode, so that in the training of the goalkeeper, by learning the estimated injection position of the target attack party and the corresponding reference index set (or the goal shooting mode), the estimated injection position of the target attack party is rapidly judged and the effect of the corresponding defense mode is adopted by identifying the reference index set, thereby improving the pertinence and the effectiveness of the goal analysis and the training.

Optionally, the shooting mode includes: each time a ball is spotted, the target attack party initially stands on the position information and/or the motion trail information and the body action information of the position of the football;

The analysis module 220 is specifically configured to determine, based on the set action key indicator, a reference indicator set for identifying the injection position information of the soccer ball in the goal when the target hits the goal according to the goal policy information or the history goal record information, so as to determine the estimated injection position information of the target hit party based on the reference indicator set.

Specifically, in the scene of training the goalkeeper, the shooting mode may be specifically three parts of an initial standing position, a movement track and a body action of the target attacking party, where each part includes a plurality of corresponding action key indexes, so as to analyze, according to historical goal record information (or goal strategy information), which action key indexes are related to the actual shooting positions of the target attacking party (because the action habits of the target attacking party corresponding to different target attacking parties are different, the action key indexes related to the actual shooting positions of the target attacking party are not necessarily the same), so as to find out the body action parameters corresponding to the relevant action key indexes and the associated corresponding shooting positions (estimated shooting positions), and further train the goalkeeper.

Optionally, the football goal analysis system 200 further comprises:

the generating module 230 is configured to generate a defensive training scheme based on the reference index set and the estimated injection position information.

Specifically, the generating module 230 is connected to the analyzing module 220, and in a scenario of training the goalkeeper, the generating module 230 may generate a defensive training scheme to train the goalkeeper based on the reference index set and the estimated injection position obtained by the analyzing module 220.

The defending training scheme generally comprises a learning part and a training part, wherein the learning part mainly informs a goalkeeper of the estimated injection position in advance and enables the goalkeeper to watch actions, postures, physical states and the like corresponding to the reference index groups (corresponding to physical action parameters) so as to establish the recognition capability of the goalkeeper on the reference index groups and the memory of the estimated injection position; the training part mainly generates virtual athletes through VR, AR and the like, simulates the corresponding body action states of the reference index groups, and transmits football to the estimated injection positions corresponding to the reference index groups through the existing training equipment such as an automatic service robot and the like, so that the goalkeeper is trained to timely defend the estimated injection positions when observing the corresponding reference index groups.

In practical situations, the estimated shooting positions corresponding to the goal are usually concentrated in a few areas (if the goal is divided into a plurality of areas), the corresponding reference index groups are also only a few groups (usually only the reference index groups with highest occurrence frequency are used for training the goal for the goal shooting side, and the training of the goal shooting side is not more than five reference index groups), so that the goal defending training of a specific team only needs to memorize and train a small number of reference index groups and combinations of estimated shooting positions (the main goal personnel need to memorize two to three groups, and for the rest goal personnel, only the reference index groups with highest occurrence frequency need to be trained, and the total memory and the training number are usually within ten groups), the goal defending capability can be remarkably improved, and the goal training effect (such training is usually the training of the professional athlete for the professional event, and the athletes have long training time, so that the goal training effect can be guaranteed by using the system as a training tool for memorizing and the estimated shooting positions).

In some embodiments, the soccer ball analysis system 200 further includes:

the defending input module 240 is configured to input defending policy information of a target defending party or select pre-stored historical defending record information of the target defending party.

The history defending record information is obtained based on defending strategy information of the target defending party, wherein the defending strategy information comprises position information and defending modes of the defending party when football is actually shot in the goal.

Specifically, the defending input module 240 is connected to the analyzing module 220, and besides the training scene of the gatekeeper, the system can also be used for making a goal tactic strategy during the competition, at this time, the system further comprises the defending input module 240 for analyzing the defending strategy trend (including the position of the defending party with higher occurrence frequency and the defending mode, such as defending, kicking and enclosing, etc.) by inputting the defending strategy information of the target defending party, thereby finding out the position of the defending party which is not good at defending, so as to improve the goal success rate.

In some embodiments, the analysis module 220 is further configured to:

And determining the goal hitting result of the target attack party based on the input defense strategy information of the target defense party, the goal hitting strategy information of the target attack party and the set comparison rule.

Specifically, in the scene of making and analyzing the goal tactics, the goal tactics information can be input, and meanwhile, the possible goal tactics information (which can be the planned shooting mode and the expected shooting position at the moment) can be input, so that the goal tactics information and the goal tactics information are compared through the analysis module 220, and the probability of successful execution of the goal tactics strategy (namely, the goal result of the target attack party) is determined according to the probability of the occurrence of the goal tactics information in different positions and the probability of the selection of the goal tactics mode in the goal tactics information.

In practical application, the information of the planned defending strategy (comprising the position of the target defending party expected by single defending and the defending mode) and the planned attack strategy (comprising the planned shooting mode of the target defending party in single shot and the expected shooting position) can be respectively input, and compared with each other, whether the goal of the target attacking party is successfully defended by the target defending party or not can be judged (for example, two tactical coaches can explore the attack and defending effects of different tactical strategies in the simulated goal fight of the system, and the proper goal tactical strategy or defending strategy is selected, so that the efficiency can be improved and the configuration of the tactical strategy is convenient compared with the simulated training and the attempt of the actual tactical strategy. The method can be compared with the actual training result (such as the automatic service robot is used for sending balls according to the ball-point strategy information and informing the gatekeeper to defend according to the defending strategy information so as to check and train the defending strategy execution capacity of the gatekeeper, or conversely, the player is used for carrying out ball-point according to the ball-point strategy information to judge and train the ball-point strategy execution capacity of the player so as to improve the strategy execution capacity of the player in the ball-point aspect.

In some embodiments, the comparison rule includes: if the actual shooting position information is the same as the goal area corresponding to the position information of the defender, and the shooting mode and the defending mode meet the set matching relation, the ball-shooting result is that the defending is successful and the ball-shooting fails; the goal area is used for representing a set number of areas into which the goals are divided based on a set mode, and comprises an edge area and a non-edge area; the matching relationship is used for representing the corresponding relationship between the shooting mode obtained by classifying the shooting mode based on the set parameter range and the defending mode obtained by classifying the defending mode based on the set parameter range.

Specifically, in the comparison scene of the goal strategy and the defending strategy, in order to simplify analysis, the actual injection position information and the position information of the defending party can be generally classified into one of a plurality of goal regions (i.e. the vertical plane of the goal is divided into a plurality of goal regions), and then whether the corresponding goal regions are the same or not and whether the goal shooting mode and the defending mode are corresponding or not are compared, so as to judge whether the defending is successful or not.

Because the area close to the goal edge in the goal area is more difficult to defend and the goal is more difficult to shoot (because the goal shot on the edge area can easily hit the goal post or kick the area outside the goal), the goal can be evaluated separately from the non-edge area so as to be more in line with the actual situation. According to the analysis precision requirement, the goal area can be 3*2 (namely six areas in three rows and two columns), 4*2, 5*2, 5*3 and other different situations, and the method is not limited herein.

The matching relation is to simplify the comparison of the shooting mode and the defending mode, and the parameter range classification is carried out based on the action key indexes, so that whether the shooting mode and the defending mode can be regarded as the relation of successful defending or not is obtained.

Further, at least three shooting modes and at least three defending modes are provided, and the corresponding relation comprises: the first type of shooting mode is used for the non-edge area, and the corresponding relation of all the shooting modes to the non-edge area is the same as that of the first type of shooting mode; wherein, all the shooting modes except the first type shooting mode are used for the edge area; the second type shooting mode of the edge area corresponds to the first type defending mode; the third type of shooting mode corresponds to a defending mode except the first type of defending mode; the second type defending mode is used for defending the edge area and the adjacent non-edge area below the edge area, or the second type defending mode is also used for a single edge area, and the third type defending mode is used for defending any two adjacent areas.

Specifically, in practical application, there is often less history record information related to the goal hitting the goal, so that the relevance between the accurate parameter index in the goal strategy information and the actual shooting position of the football cannot be accurately judged, in the comparison scene of the goal strategy and the defending strategy, the accuracy requirement is high, and the practical application value is lacking (for example, when the goal is hit by a training player, the goal is not required to keep the same physical action each time, and the goal is not realized in the practical scene), therefore, the goal hitting method is summarized into a plurality of shooting modes and a plurality of defending modes, and the practical training and evaluation requirements can be better met.

In order to avoid excessive simplification, the shooting mode and the defending mode generally comprise at least three modes, so as to reflect the characteristics of different types of shooting actions, speed and track change when the football reaches the goal (because the defending difficulty of the football under different characteristics is different, such as higher speed when the football is shot under high force, the defending difficulty is higher, and shooting with lower speed usually only has the opportunity to successfully enter the ball after the goalkeeper is completely out of position), and modes such as defending, holding the ball, directly kicking out and the like (the defending success rate for the football is different under different defending modes, the defending success rate is higher when the speed of the football is lower and the position judgment is correct, and the success rate of trying to hold the ball is lower when the speed of the football is higher).

The number of shooting modes and defending modes can also be different, such as five modes, according to different accuracy of actual training requirements and evaluation requirements and different training sample sizes.

The following are examples of several common shot and defending modes:

The first type of goal shooting mode is used to represent a goal shooting for a non-edge area (also the most common normal goal shooting), and the goal shooting mode is usually closer to a goalkeeper station, so that the goalkeeper has lower defending difficulty, and the goal shooting mode can be reset to the same type even though the football speed is different.

The second type of shot mode is used to represent a shot that is directed at a high speed and toward the edge area (i.e., the ratio of the difference between the high speed and the average speed to the average speed is greater than a set value), and can be used alone because such a shot mode is generally more difficult to defend, but has a lower actual hit rate (the most frequent kicking is such a shot mode).

The defending of the second-class shooting mode is mainly through frightening interference type defending, because when the second-class shooting mode is successfully shot to the goal along the corresponding estimated shooting position, the gatekeeper cannot normally conduct effective defending, at the moment, the hit success rate can be further reduced through the interference mode (because the shooting mode is large in general action amplitude, slight deformation of actions is easy to occur due to slight interference, and the hit rate is reduced). In general, there are different interference modes on a practical court, such as sudden shouting (frightening interference), making exaggerated anti-action (action interference), making anti-false action, etc., where the interference modes can be regarded as first-class defending modes, that is, frightening interference defending, and not defending attempting to actually control football.

The third type of shooting mode is used to represent shooting at a lower ball speed for an edge area (i.e., a case where a ratio of a difference between an average shooting ball speed and a ball speed to the average shooting ball speed is greater than a set value, and the ball speed is smaller than the average shooting ball speed) used to ensure a hit success rate, and for such shooting modes, the first type of defending mode is generally not interfered, and effective defending can be achieved only when a goalkeeper actually controls a football.

The defending of the third type of shooting mode can be generally called as a second type of defending mode (the defending can usually only defend a single goal area) as long as the defending personnel can judge the correct corresponding estimated shooting position, for example, effective defending can be generally realized, such as defending, kicking with feet and blocking with bodies (namely, opening the arms and blocking balls with bodies), the defending area of the defending personnel can be larger at the moment, and the defending mode using four limbs can be generally called as a third type of defending mode (the defending can usually defend a single goal area) as a simple classification, and the defending mode using the whole body can be generally called as a third type of defending mode (the defending can usually defend a plurality of adjacent areas continuously).

Based on the defending mode, a plurality of different defending modes can be simultaneously used, such as a first defending mode and a second defending mode (such as first making frightening interference action and then selecting one goal area to defend or first making false action of defending one goal area and then selecting another goal area to defend) are combined, and a first defending mode and a third defending mode (such as first making exaggeration defending action of a certain goal area and then selecting another or a plurality of areas to defend with a body) are combined.

At this time, the first type defending mode can be combined with the second type defending mode or the third type defending mode at the same time, but usually the second type defending mode cannot be combined with the third type defending mode, and the first type defending mode cannot be implemented independently (namely, when deterring interference type defending is made, a gatekeeper should also be in the goal, so that the gatekeeper should also defend certain area through the body or the limbs, namely, at this time, an area is also needed to be selected for defending or defending with the body). The goal area corresponding to the first type defending mode and the goal area corresponding to the second type defending mode and the third type defending mode can be the same or different.

According to different actual demands, more types of defending modes, such as a curve ball and a falling ball, can be added, and the defending modes of the shooting modes can be correspondingly adjusted according to actual conditions. Those skilled in the art can continuously add different types of shooting modes and defending modes according to the needs, and configure the corresponding relation between the added shooting modes and defending modes, and the corresponding relation is not limited herein.

Further, the soccer ball analysis system 200 further includes:

A mode input module 250 for determining a play mode of the dribbling to determine the number of the input dribbling strategy information and the number of the defending strategy information based on the play mode;

The storage module 260 is configured to record the input ball-pointing policy and defending policy, and generate corresponding historical ball-pointing record information and historical defending record information;

The voice prompt module 270 is configured to output voice prompt information corresponding to the attack input module, the defending input module, and the defending result;

And/or, the defending mode comprises: each time the goal is secured, the position information of the goal defender in the initial standing, and/or the physical action information of the goal defender when the goal defender contacts the football or the football enters the goal.

Specifically, the mode input module 250 is connected to the analysis module 220, and in the scene of making and analyzing the goal tactics, since there may be multiple goal playing modes in the actual game, such as the number of goals in the same time, a single goal, and not limited total times until the difference in the number of goals in both sides (i.e. the "sudden death phase" frequently occurring in the goals), the goal playing modes may be selected in advance in the mode input module 250, so that the number of corresponding number or times of goal tactics information and the number of defending tactics information are input according to the goal playing modes, so that the goal performance in the actual game may be better simulated (according to the actual requirement, a mode of multi-person fight may be also included, such as the presence of multiple target goals and multiple gates, and the corresponding gates will coach and fine coach, so as to select the goal order of different target goals and the goal order of different gates, and better simulate the scene in the actual game.

In some embodiments, according to different selected dribbling modes, different people may input dribbling policy information and defending policy information on the same device (for example, different input devices connected to the same terminal are used to input dribbling policy information and defending policy information respectively, the terminal includes an analysis module to output a competition result obtained by analysis), or different people may input dribbling policy information and defending policy information through different devices (for example, a server obtains dribbling policy information and defending policy information input by different terminals, and then an analysis module in the server determines a competition result and distributes the competition result to each device).

In order to record the history ball record information and the history defending record information of the player conveniently, the recorded related information can be recorded through the storage module 260, and the result of the simulation analysis can be recorded. The storage module 260 is connected to the analysis module 220.

In the field of special football such as football for the blind, different kicking modes, shooting positions and the like are corresponding, the football is represented by different sounds, so that the player for the blind can recognize the football. Therefore, when the ball-setting training or the tactical strategy corresponding to the football for the blind person is formulated (in the ball-setting training of the football for the blind person, the reference index set for identification generally only contains fewer indexes, because the goalkeeper in the football for the blind person can be a person with sound eyesight, and can also be a person with poor eyesight, and the problem that the identification of the position, the action and the like of the person with the poor eyesight cannot be effectively performed can be solved when the index in the reference index set is too many), the prompt of the sound prompt module 270 (generally, the information in the reference index set corresponds to the sound prompt and the prompt corresponding injection position) is convenient for the football player for the blind person to train. The voice prompt module 270 is coupled to the analysis module 220.

By way of example, the following description is based on the present football point ball analysis system 200, and provides some application interface references, and the following description and corresponding figures are merely exemplary and are not intended to be limiting of interfaces and elements therein.

Fig. 2b is a schematic diagram of an interface for inputting the ball-in-ball strategy information in the football ball-in-ball analysis system. The interface may be obtained by a combination of the analysis module 220 and the attack input module 210. Selecting an injection position and a shooting mode to realize the effect of manually inputting the shot strategy information, wherein a column of the target attack party is the input name of the target attack party; the goal position selection is used for selecting a goal region corresponding to the actual injection position (wherein the region below the middle is a non-edge region, and other positions belong to edge regions, and if a circle in the figure is the actual injection position, the region where the circle is located is the selected goal region); the shooting mode is selected as a simplified shooting mode (such as the three shooting modes). In practical applications, the division of goal regions may be adjusted, or the division of goal shooting modes may be adjusted. Entry and storage of the ball-point strategy information (corresponding to the function of the storage module 260) is accomplished by triggering the button below.

As shown in fig. 2c, an interface diagram for inputting the ball-in-ball strategy information in another football ball-in-ball analysis system is shown. The method comprises the steps of automatically extracting corresponding balling strategy information by uploading historical balling information corresponding data marked with a target attack party, wherein a column of the target attack party is used for inputting the name of the target attack party, and a balling strategy extraction part is used for inputting the number of balling times required to be extracted and a corresponding time period (because a balling corresponding picture required to be in a part of time period generally under the condition of inputting video data); meanwhile, there is a game mode selection menu below to select a corresponding game mode (corresponding to the function of the mode input module 250). Therefore, the corresponding ball-pointing strategy information can be automatically extracted by inputting the historical ball-pointing information.

FIG. 2d is a schematic diagram of an interface for outputting a training regimen in a football goal analysis system. The interface may be obtained by a combination of the analysis module 220 and the generation module 230. The uppermost area 1 is different goal areas corresponding to the high-frequency goal modes based on analysis, usually each goal area corresponds to only one goal mode, different area labels are selected, different high-frequency goal modes are switched, a reference index group corresponding to a specific goal mode may correspond to different stages, such as a standby batting action stage and a batting action stage in the figure, wherein the standby batting action stage corresponds to two indexes (reference characteristic 1 and reference characteristic 2 in the figure correspond to right-hand action and left-hand action respectively), and the reference index group is intuitively represented through imaging, so that the visual recognition and understanding of a goalkeeper are facilitated. By selecting the triangular buttons in the figure, voice prompt can be performed to realize the functions corresponding to the voice prompt module 270. By selecting the output button, the obtained reference index set and the corresponding estimated injection position can be output so as to be combined with AR, VR equipment and the like for training.

Fig. 2e is a schematic diagram of an interface for inputting defense strategy information in the football goal analysis system. The interface may be obtained through a combination of the analysis module 220 and the defensive input module 240. Selecting a defending position and a defending mode to realize the effect of manually inputting defending strategy information, wherein the number of selectable defending areas is different corresponding to different defending modes, for example, a third defending mode, two or three continuous goal areas (areas filled by three oblique lines at the left side in the figure) can be selected, one of the second defending mode and the third defending mode can be selected, and whether a first defending mode is adopted at the same time (namely, the third defending mode and the first defending mode are selected at the same time in the figure), wherein the goal area corresponding to the first defending mode is an independent oblique line filling area at the right side in the figure) can be selected; the defending position is selected for selecting the defending position selected after the defending operator prejudges (different defending modes can correspond to different defending positions). And finishing the input and storage of the ball-pointing strategy information by triggering the button below. By combining the comparison rules, whether the input defending strategy information and the input ball-pointing strategy information in the figure 2b meet the corresponding matching relation can be judged, and whether defending is successful or whether ball-pointing is successful is further judged.

According to the football goal analysis system provided by the embodiment of the application, through the combination of the modules, the history goal record information of the goal of the target attack party is realized, and the relation between the body, the action and the position characteristics and the corresponding football shooting position is analyzed and refined, so that a goalkeeper can rapidly identify the action and the position characteristics of the target attack party according to the relation, and further rapidly and accurately judge the football shooting position, further the football goal analysis accuracy and the training efficiency are improved, the goal defending capability and the defending effect of the goalkeeper are improved, the system is applicable to different scenes such as goalkeeper training, goal tactic formulation and analysis, and blind person football player training, and the like, and the goal tactic level of a player is remarkably improved.

Fig. 3 is a flowchart of a football goal analysis method according to an embodiment of the present application. As shown in fig. 3, including but not limited to the following steps:

Step S301, responding to the received analysis request, and acquiring historical dribble record information of a target attack party corresponding to the identification of the target attack party.

The analysis request comprises an identification of a target attack party to be analyzed, the historical ball striking record information is obtained based on ball striking strategy information of the target attack party, and the ball striking strategy information at least comprises: each time a ball is spotted, the target offender's initial standing position information, movement trajectory information and body motion information at the position of the soccer ball, and the actual shot position information of the soccer ball in the goal.

Specifically, the analysis request is a request for performing a goal-related analysis received by the soccer goal analysis system. The information contained in the analysis request includes an identification of the target offender to be analyzed for identification.

The mark of the target attack party is a mark manually added in the history ball striking record information (in a specific form of video data or picture data), and the target attack party in the ball striking strategy information can be identified through an image identification algorithm by adding the mark, so that information such as an action key index of the target attack party can be extracted.

The initial standing position, the movement track and the body action are three main indexes for analyzing the predicted point ball injection position, each index possibly corresponds to a plurality of parameters, corresponds to different athletes (target attack parties), and has the advantages of little relevance with the actual injection position and larger relevance. Therefore, it is necessary to combine the history spot ball record information to analyze the parameters associated with the actual injection position of each target attack party.

The initial standing position mainly relates to habits of players, for example, according to different standing positions when a ball is clicked, running distances from the player to the football are different, and shooting trends are also different; the motion trail has influence on the state of the football when the football is shot, and the shot football state (such as whether the football rotates) is different according to whether acceleration action and body rotation action exist in the moving process; physical actions have a great influence on the state of football in flight, such as the action gestures of footsteps, arms and the like in shooting, and are generally associated with the final shooting position of football, so specific analysis is required.

Step S302, based on the set action key index, extracting body action parameters corresponding to the set action key index from body action information corresponding to each time of ball striking in the history ball striking record information.

Specifically, the motion key indexes are quantitative index parameters established based on the initial standing position, the motion track and the body motion, and motion characteristics (namely body motion information) of a person containing a target attack mark in a specified video frame or picture in the history ball record information are generally extracted, so that the body motion parameters corresponding to the motion key indexes can be obtained.

The physical action parameters are specific values corresponding to the action key indexes, and are usually represented as information which is convenient for goalkeeper to quickly identify, such as specific actions, positions and the like, when the action key indexes are hand state indexes, the corresponding physical action parameters can be fist making, opening, specific gestures and the like, and when the action key indexes are hand position indexes, the corresponding physical action parameters can be chest, body two sides, high lifting, body back and the like.

Determining body action parameters, such as initial standing positions, and identifying the positions of side lines (or forbidden zone lines and goal lines) in an image through an image identification model to determine the positions of target attack parties so as to obtain specific body action parameters; if the left arm corresponds to the body motion parameter in the body motion, the left arm can be identified through an image identification algorithm and then is determined according to the relative position of the left arm and the body trunk.

Step S303, analyzing and processing the initial standing position information of the target attack party, the moving track information of the position of the football, and the extracted body action parameters, which correspond to each ball point, and determining a reference index set for identifying the shot position information of the football in the goal when the target attack party is in the ball point, so as to determine the estimated shot position information of the target attack party based on the reference index set.

Specifically, by comparing the relevance between the physical action parameters in the key indexes of the actions and the actual injection position of the football at each ball-hitting time, the relevance between the physical action parameters (different parameters associated with different target attacks) and the actual injection position of the football can be determined through the changed significance or relevance, and the physical action parameters are used as reference index groups, so that when the goalkeeper recognizes the reference index groups, the corresponding injection position (namely the estimated injection position) can be quickly determined.

The reference index group is not a common array or a digital combination (stored in a digital mode in the system), but is a combination of action characteristics which can be identified by a goalkeeper, such as the reference index group [ left hand is on the inner side of the body, head and left side are offset, feet are lifted to be as high as knees ], and the corresponding injection position is a ground contact position below the left side of the goal. The reference index group can be conveniently and quickly identified by a goalkeeper, and the estimated injection position can be judged through training for a limited number of times.

The specific determination of the reference index group can be determined by determining the change condition of the body action parameters corresponding to each action key index under the condition of the actual shooting position (shooting towards the same goal area) and/or shooting mode (determining the speed, angle and the like) of the same football in the history goal record information, and if the same action key index corresponding to the actual shooting position of the same football is unchanged (or the proportion of the same body action parameters to the total number exceeds a set value), the same action key index can be regarded as an alternative reference index.

If the alternative reference indexes corresponding to the actual injection positions of different football are the same, filtering the same alternative reference indexes (because the alternative reference indexes cannot be distinguished), taking the filtered alternative reference indexes as a reference index group (if the filtered alternative reference indexes are too many, the alternative reference indexes with the set quantity can be selected as the reference index group based on the ranking of the proportion of the same body action parameters to the total number).

Step S304, a defending training scheme is generated based on the reference index set and the estimated injection position information.

Specifically, the training can be performed by using VR glasses, AR imaging and other modes, based on a reference index set, corresponding actions are simulated and generated, and football (such as a rotating ball, a falling ball and the like) in a corresponding state is transmitted to a corresponding injection position through existing automatic training equipment such as an automatic football transmitter and the like, and a goalkeeper is informed of a predicted injection position corresponding to a certain reference index set in advance, so that the goalkeeper can quickly generate muscle memory or response by identifying the reference index set and combining with the pre-known injection position, and quickly judge the predicted injection position when the reference index set is seen. By repeating the process, the effect of targeted training on the goalkeeper is achieved.

According to the football goal analysis method provided by the embodiment of the application, the historical goal record information of the target attack party corresponding to the identification of the target attack party is obtained by responding to the received analysis request, and then the body action parameters corresponding to the set action key indexes in the body action information corresponding to each goal in the historical goal record information are extracted based on the set action key indexes; and then analyzing and processing the position information of the initial standing of the target attack party, the motion track information of the position of the football, which are corresponding to each ball point, and the extracted body action parameters, determining a reference index group for identifying the shot position information of the football in the goal when the target attack party points, and then generating a defending training scheme based on the reference index group and the estimated shot position information. Therefore, the relation between the action, the initial standing position and the action track of the target attack party and the corresponding football shooting position can be analyzed and proposed through the history ball shooting record information of the ball shooting of the target attack party, so that the goalkeeper can rapidly identify the action and the position characteristics of the target attack party according to the relation, and further rapidly and accurately judge the football shooting position of the target attack party, further the football shooting analysis accuracy and the training efficiency are improved, and the ball shooting defending capability and the defending effect of the goalkeeper are improved.

Fig. 4a is a flowchart of a football goal analysis method according to another embodiment of the present application. As shown in fig. 4a, the football point ball analysis method includes:

step S401, responding to the received analysis request, and acquiring historical dribble record information of the target attack party corresponding to the identification of the target attack party.

Specifically, the content of this step is the same as that of step S301 in the embodiment shown in fig. 3, and the details are not repeated here.

Step S402, determining the previous frame of the video frame of which the football position starts to change in the history ball striking record information as a batting frame.

Specifically, for the case where the history ball record information is video data, a specific frame needs to be selected to extract the physical motion parameters corresponding to the motion key indicators.

Further, the action key indicators include a preparatory action key indicator and a batting action key indicator; the key indexes of the preparation actions comprise a standing position direction index, a standing position distance index and a body orientation index; the key indexes of the batting action comprise a hand position index, an elbow position index, a shoulder action key index, a leg highest lifting position index and a body inclination index when batting.

Specifically, according to different actual conditions, different action key indexes can be selected and adjusted to reflect action characteristics in the football goal process.

Step S403, determining a previous frame of the video frame in which the target offensive leg starts descending last time before the batting frame as a batting action frame for determining motion trajectory information and body action information.

Specifically, the human body motion before kicking, including running in the football direction and lifting the kicking leg before kicking, ensures enough momentum to the football, and can recognize the movement trace (corresponding movement trace information) and the preparatory motion (corresponding body motion information) of the leg according to the state (i.e., the batting motion frame) when the kicking leg is lifted to the highest point.

Step S404, determining a video frame of the last two legs of the target offender on the ground before the ball striking action frame as a preparation action frame for determining the initial standing position information of the target offender.

Specifically, similarly, since the target attack party is in a motion state after running in the football direction, a state that both feet stand on the ground does not occur at this time, therefore, the position of the target attack party in the video frame of the ground, where the last two legs are located, can be determined as the initial standing position before the attack is initiated, and the video frame is taken as the corresponding preparation action frame.

Step S405, extracting initial standing position information based on the position of the target attack party relative to the football in the preparation action frame.

Specifically, after the initial standing position is determined, the relative positions of the target attack party, the football and the boundary (or the forbidden zone line) are combined, so that initial standing position information corresponding to the target attack party, such as a standing direction (corresponding to a standing direction index), a distance between the target attack party and the football (corresponding to a standing distance index) and a body orientation (corresponding to a body orientation index), can be determined.

Step S406, extracting motion trail information based on the video frames from the preparation action frame to the batting action frame.

Specifically, by preparing different video frames from the action frame to the batting action frame, the motion track, acceleration, and other conditions of the target attack party can be determined by changing the target attack party relative to different positions of the football, so as to extract the physical action parameters corresponding to the action key indexes (such as the physical movement acceleration before reaching the football, the greater the acceleration, the higher the ball speed in general).

Step S407, extracting body action parameters corresponding to the key indexes of the preparation action based on the relative positions of the target attack party and the football in the batting action frame and the body action information.

Specifically, in the video frame, which is the batting action frame, the relative position of the target attack side and the football (including the combination boundary line and the forbidden zone line, determining the angle of the target attack side relative to the football, the distance of the footstep of the target attack side relative to the football, and the like) and the body action information (such as hand action, head action, and the like) can be extracted, and the body action parameters corresponding to the set key indexes of the preparation action are determined.

Step S408, based on the body motion information of the target attack party in the batting frame, extracting the body motion parameters corresponding to the batting motion key indexes.

Specifically, similar to the foregoing steps, the body motion parameters corresponding to the set key indexes of the batting motion may be determined by the body motion information (mainly, the hand motion and the head motion) of the target attacking party in the batting frame.

Step S409, based on the value interval division of the body motion parameters corresponding to each motion key index, determining the value interval of the body motion parameters corresponding to the target attack party when each ball is spotted.

Specifically, the physical action parameters extracted through the foregoing steps are usually actual values, for example, the distance between the foot of the target attacking party and the football is 700 units (the unit length may be mm), and such values are not recognized by naked eyes for goalkeepers, and have low reference value. Therefore, the value interval corresponding to the body motion parameter needs to be determined according to the relation between the value of the body motion parameter and the preset value interval, and the value interval is represented by the characteristic corresponding to the value interval, if the target attacking side's foot is 700 units of length from the football and belongs to the value interval [0,1000], the corresponding characteristic is "very close", the body motion parameter of the motion key index can be determined to be "very close" (because the characteristic is rapidly recognized by the goalkeeper through naked eyes).

Meanwhile, the value interval is divided, so that the phenomenon that the obtained body action parameters are too discrete, so that the goal attack party is difficult to analyze is avoided, and the effectiveness and the practicability of analysis are further ensured.

Step S410, determining the preparation action key index corresponding to at least two physical action parameters with the largest number of times in the same value interval as a first alternative action key index when the ball is clicked, and determining the same value interval corresponding to the first alternative action key index as a corresponding target value interval.

Specifically, the preparation action key index of the corresponding action key indexes can select the index with the highest probability of the same corresponding body action parameter (located in the value interval) as a group of candidate indexes, namely the first candidate action key index, and takes the same value interval as a target value interval, wherein the corresponding characteristic (or the corresponding body action parameter) of the target value interval is taken as a part of a candidate reference index group.

Step S411, determining the batting action key index corresponding to at least two physical action parameters with the largest number of times in the same value interval as the second alternative action key index when the batting is performed, and determining the same value interval corresponding to the second alternative action key index as the corresponding target value interval.

Specifically, the determination of the candidate motion key indicator (i.e., the second candidate motion key indicator) corresponding to the batting motion key indicator is the same as the first candidate motion key indicator, and will not be repeated here.

Step S412, selecting at least one action key indicator from the first candidate action key indicator and the second candidate action key indicator, respectively, as a potential reference indicator group.

Specifically, since the first candidate action key index and the second candidate action key index are action key indexes obtained from different links (a hitting preparation link and a hitting link) in the hitting process, respectively, at least one action key index needs to be selected from the first candidate action key index and the second candidate action key index as a potential reference index group for more comprehensively reflecting action characteristics of the target attacking party in the hitting process.

The determination of the potential reference index set may be performed by optionally selecting the first candidate action key index and the second candidate action key index, and then determining the finally obtained reference index set by combining the actual injection position information.

Step S413, determining a reference index set for identifying the injection position information based on the potential reference index set, the initial standing position information of the target attack party, the motion trail information, and the actual injection position information in the history ball record information.

Specifically, in the potential reference index group, at least one index is selected from the first candidate action key index and the second candidate action key index to reflect the characteristics of the preparation action and the striking action of the striking ball, and because the initial standing position and the movement track of the target attack party are generally larger to the ball speed and the ball direction of the striking ball, the action key index and the body action parameter corresponding to the two information are generally selected at least one from each other, so that the reference index group is determined according to the corresponding relation between the selected indexes and the actual shooting position information.

Further, as shown in fig. 4b, which is a flowchart of a method for determining a reference index set, the step of selecting the reference index set from a plurality of indexes specifically includes:

Step S4131, traversing the history shot record information corresponding to each shot of the target attack party, and determining all the actual injection position information corresponding to the shot of which the potential reference index group is positioned in the corresponding target value interval as the estimated injection position.

Specifically, when determining the reference index group, the actual injection position information in the history ball record information needs to be combined, if the selected index combination (i.e., the potential reference index group) can correspond to the actual injection position information (i.e., the actual injection position determined by the index combination is unique or the goal area corresponding to the actual injection position is unique), the index combination can be determined as the reference index group corresponding to the actual injection position information.

For convenience of description, all actual injection positions (which may be the same or different) corresponding to each potential reference index set are collectively referred to as estimated injection positions.

In step S4132, if the potential reference index set is located in the corresponding target value interval, the estimated injection position can be uniquely determined, and the potential reference index set located in the corresponding target value interval is determined as the reference index set corresponding to the estimated injection position.

Specifically, the corresponding estimated injection positions of each potential reference index group are traversed in sequence, and when two or more (including a plurality of corresponding goal regions) corresponding to the estimated injection positions of the potential reference index groups exist, the potential reference index groups can be directly determined to be incapable of serving as the reference index groups.

If the estimated injection position corresponding to the potential reference index set is unique, the potential reference index set can be used as the reference index set corresponding to the estimated injection position.

The same estimated injection position may correspond to a plurality of reference index sets, and then a reference index set with a smaller number of indexes and convenient for goalkeeper to identify is preferentially selected as the reference index set corresponding to the estimated injection position.

For example, if the body motion parameter [ closer to the ball distance ] corresponding to the key target action index and the key target action index [ right-hand deviation from the head ] corresponding to the key target action index can be uniquely determined to an estimated injection position [ corresponding to the goal region in the upper right corner of the goal ], the estimated injection position [ closer to the ball distance, right-hand deviation from the head ] corresponding to the upper right corner of the goal ] can be used as the reference index set corresponding to the estimated injection position [ corresponding to the goal region in the upper right corner of the goal ].

In step S4133, if the estimated injection position can be uniquely determined by combining with one of the position information or the motion trail information of the target attack party, one of the potential reference index set, the position information or the motion trail information is determined as the reference index set corresponding to the estimated injection position.

Specifically, similar to the foregoing steps, if some estimated injection positions cannot be uniquely determined only by the potential reference index set (here, the goal area corresponding to the estimated injection position is usually determined), one of the position information or the motion trail information of the target attack party that initially stands may be combined, and after one of the two indexes is combined, whether the estimated injection position can be uniquely determined is determined, and if the estimated injection position can be uniquely determined, the corresponding potential reference index set and the combined index may be used as the reference index set corresponding to the estimated injection position.

For example, if the body action parameter [ right body side facing goal ] corresponding to the key action index of the target attack party is prepared, the key action index [ leg elevation angle is lower, and the left hand is located in front of the body ] cannot uniquely determine the estimated injection position (a plurality of estimated injection positions are simultaneously corresponding), but the body action parameter [ speed when moving towards the ball is faster ] corresponding to the motion track information is combined, one estimated injection position [ goal region at the lowest position in the middle of the goal ] can be uniquely determined, then the reference index group corresponding to the estimated injection position [ right body side facing goal, leg elevation angle is lower, speed when the left hand is located in front of the body and moves towards the ball ] can be used as the reference index group corresponding to the estimated injection position [ goal region at the lowest position in the middle of the goal ].

In step S4134, if the estimated injection position can be uniquely determined by combining the position information and the motion trail information of the target attack party, the potential reference index set, the position information and the motion trail information are determined as the reference index set corresponding to the estimated injection position.

Specifically, if the estimated injection position cannot be determined uniquely by only using the potential reference index set, the initial standing position information, or the motion trail information, the potential reference index set, the initial standing position information, and the motion trail information may be combined to determine whether the estimated injection position can be determined. The principle is similar to the previous steps and will not be repeated here.

In step S4135, if the potential reference index set is located in the corresponding target value interval, there is no corresponding estimated injection position, and it is determined that the potential reference index set does not belong to the reference index set corresponding to the estimated injection position.

Specifically, if the estimated injection position cannot be uniquely determined whether the potential reference index set is merely used or combined with indexes such as initial standing position information, motion track information and the like, it is determined that the potential reference index set does not belong to the reference index set corresponding to the estimated injection position; if the predicted injection position does not have the corresponding reference index group, the goal can preferentially pay attention to other goal areas to improve the defending effect, and the goal cannot kick the goal to the predicted injection position most likely due to the goal-hitting habit of the target.

Step S414, generating a defending training scheme based on the reference index set and the estimated injection position information.

Specifically, the content of this step is the same as that of step S304 in the embodiment shown in fig. 3, and will not be described here again.

According to the football goal analysis method provided by the embodiment of the application, after an analysis request is received, image data in a designated video frame is automatically extracted based on an image analysis algorithm, initial standing position information, movement track information, preparation action key indexes, batting action key indexes and other indexes of a target attack party and actual injection positions are identified, and a reference index group corresponding to the injection positions is determined according to the corresponding relation between the indexes and the actual injection positions, so that a defensive training scheme is generated based on the corresponding relation between the reference index group and the estimated injection positions, and the corresponding goalkeeper trains, thereby improving the pertinence of goalkeeper goal training and further improving the goalkeeper goal defense effect.

Fig. 5 is a schematic structural diagram of a football spot analyzing device according to an embodiment of the application. As shown in fig. 5, the soccer ball analysis device 500 includes: an acquisition module 510, an extraction module 520, an analysis module 530, and a processing module 540, wherein:

The obtaining module 510 is configured to obtain, in response to the received analysis request, historical dribbling record information of a target attack party corresponding to the identifier of the target attack party; the analysis request comprises an identification of a target attack party to be analyzed, the historical ball striking record information is obtained based on ball striking strategy information of the target attack party, and the ball striking strategy information at least comprises: each time a ball is spotted, the initial standing position information of a target attack party, the motion track information and the body action information of the position of the football, and the actual shooting position information of the football in the goal;

the extracting module 520 is configured to extract, based on the set motion key indicator, a body motion parameter corresponding to the set motion key indicator from body motion information corresponding to each time of ball striking in the history ball striking record information;

The analysis module 530 is configured to perform analysis processing on the position information of the initial standing of the target attack party, the motion track information of the position of the soccer where the target attack party travels, and the extracted body motion parameters, and determine a reference index set for identifying the shot position information of the soccer in the goal when the target attack party is in the goal, so as to determine the estimated shot position information of the target attack party based on the reference index set;

the processing module 540 is configured to generate a defensive training scheme based on the reference index set and the estimated injection position information.

Further, the extraction module 520 includes, the action key indicators include a preparation action key indicator and a batting action key indicator, where the preparation action key indicator includes a standing direction indicator, a standing distance indicator, and a body orientation indicator; the key indexes of the batting action comprise a hand position index, an elbow position index, a shoulder action key index, a leg highest lifting position index and a body inclination index when batting.

Further, the extracting module 520 is specifically configured to determine, as the batting frame, a previous frame of the video frame in which the football position starts to change in the history ball-in-ball record information if the history ball-in-ball record information is video data; determining a previous frame of a video frame in which the target offender's leg last begins to descend before the batting frame as a batting action frame for determining motion trajectory information and body action information; determining a video frame of the last two legs of the target offender on the ground before the ball striking action frame as a preparation action frame for determining the initial standing position information of the target offender; extracting initial standing position information based on the position of a target attack party relative to the football in the preparation action frame; extracting motion trail information based on video frames between the preparation action frame and the batting action frame; extracting body action parameters corresponding to key indexes of the preparation action based on the relative positions of the target attack party and the football and the body action information in the batting action frame; and extracting body action parameters corresponding to key indexes of the batting action based on the body action information of the target attack party in the batting frame.

Further, the analysis module 530 is specifically configured to determine, based on the value interval division of the body motion parameter corresponding to each motion key indicator, a value interval where the body motion parameter corresponding to the target attack party is located when each ball is spotted; determining preparation action key indexes corresponding to at least two body action parameters with the largest frequency in the same value interval as first alternative action key indexes when the ball is clicked, and determining the same value interval corresponding to the first alternative action key indexes as a corresponding target value interval; determining the batting action key indexes corresponding to at least two physical action parameters with the largest times in the same value interval as a second alternative action key index when the batting is carried out, and determining the same value interval corresponding to the second alternative action key index as a corresponding target value interval; selecting at least one action key index from the first candidate action key index and the second candidate action key index respectively as a potential reference index group; the reference index set for identifying the injection position information is determined based on the potential reference index set, the initial standing position information of the target attack party, the movement track information and the actual injection position information in the history point ball record information.

Further, the analysis module 530 is specifically configured to traverse the history shot record information corresponding to each shot of the target attack party, and determine all the actual injection position information corresponding to the shot of the potential reference index group located in the corresponding target value interval as the estimated injection position; if the potential reference index group is positioned in the corresponding target value interval, the estimated injection position can be uniquely determined, and the potential reference index group positioned in the corresponding target value interval is determined as the reference index group corresponding to the estimated injection position; or if the estimated injection position can be uniquely determined by combining with one of the position information or the motion trail information of the initial standing of the target attack party, determining one of the potential reference index group, the position information or the motion trail information as the reference index group corresponding to the estimated injection position; or if the estimated injection position can be uniquely determined by combining the position information and the motion trail information of the initial standing of the target attack party, determining the potential reference index group, the position information and the motion trail information as the reference index group corresponding to the estimated injection position; or if the potential reference index group is located in the corresponding target value interval and the corresponding estimated injection position does not exist, determining that the potential reference index group does not belong to the reference index group corresponding to the estimated injection position.

In the embodiment, the football goal analyzing device can record information through the history goal of the target attack party through the combination of the modules, analyzing and refining the relation between the action, the initial standing position and the action track of the football and the corresponding football shooting position when the football is shot, so that the goalkeeper can quickly identify the action and the position characteristics of the target attack party, and further quickly and accurately judge the football shooting position, thereby improving the accuracy of football goal analysis and training efficiency, and improving the goal defending capability and defending effect of the goalkeeper.

Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application, as shown in fig. 6, the electronic device 600 includes: a memory 610 and a processor 620.

Wherein the memory 610 stores a computer program executable by the at least one processor 620. The computer program is executed by the at least one processor 620 to cause the electronic device to implement the football spot analysis method as provided in any one of the embodiments above.

Wherein the memory 610 and the processor 620 may be connected by a bus 630.

According to the actual application scene, the electronic equipment can be an intelligent terminal, such as a smart phone, a tablet computer, a personal computer or other terminal equipment capable of realizing the football point analysis method; the above-mentioned football goal analysis method may also be implemented jointly by a device system including a server and a terminal device in communication with each other, for example, by combining the terminal (as an input device) and the server.

The relevant descriptions and effects corresponding to the relevant description and effects corresponding to the method embodiments may be understood, and are not repeated herein.

An embodiment of the present application provides a computer-readable storage medium having stored thereon a computer program that is executed by a processor to implement the football goal analysis method of any of the embodiments described above.

The computer readable storage medium may be, among other things, ROM, random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

One embodiment of the present application provides a computer program product containing computer-executable instructions for implementing the football goal analysis method of any of the embodiments as corresponds to fig. 2a to 4a when executed by a processor.

In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of modules is merely a logical function division, and there may be additional divisions of actual implementation, e.g., multiple modules or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or modules, which may be in electrical, mechanical, or other forms.

Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope of the application being indicated by the following claims.

It is to be understood that the application is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof.

Claims (10)

1. A football goal analysis system, the system comprising:

The goal shooting system comprises a goal shooting input module, a goal shooting input module and a goal shooting control module, wherein the goal shooting input module is used for inputting goal shooting strategy information of a target attacker or selecting pre-stored history goal shooting record information of the target attacker, the history goal shooting record information is obtained based on the goal shooting strategy information of the target attacker, and the goal shooting strategy information comprises actual shooting position information and shooting mode of football in a goal;

And the analysis module is used for determining the estimated injection position information of the target attack party according to the set action key index and the ball-pointing strategy information or the historical ball-pointing record information.

2. The football goal analysis system of claim 1, wherein the goal shooting method comprises: each time a ball is spotted, the target attack party initially stands on the position information and/or the motion trail information and the body action information of the position of the football;

The analysis module is specifically configured to determine, based on a set action key indicator, a reference indicator set for identifying information of an injection position of a soccer in a goal when the target hits the goal according to the goal strategy information or the history goal record information, so as to determine, based on the reference indicator set, estimated injection position information of the target hit;

The system further comprises:

and the generating module is used for generating a defending training scheme based on the reference index group and the estimated injection position information.

3. The football spot ball analysis system of claim 1, further comprising:

The defending input module is used for inputting defending strategy information of a target defending party or selecting pre-stored historical defending record information of the target defending party, wherein the historical defending record information is obtained based on defending strategy information of the target defending party, and the defending strategy information comprises position information and defending modes of the defending party when football is actually shot in a goal;

The analysis module is also configured to:

Determining a goal attack goal based on the input defense strategy information of the goal defense party, the goal attack goal strategy information and the set comparison rule;

The comparison rule includes:

If the actual shooting position information is the same as the goal area corresponding to the position information of the defender, and the shooting mode and the defending mode meet the set matching relation, the ball-point result is that the defending is successful and the ball-point fails;

the goal areas are used for representing a set number of areas into which the goals are divided based on a set mode, and the goal areas comprise edge areas and non-edge areas;

The matching relationship is used for representing the corresponding relationship between the shooting mode obtained by classifying the shooting mode based on the set parameter range and the defending mode obtained by classifying the defending mode based on the set parameter range,

The shooting modes are at least three, the defending modes are at least three, and the corresponding relation comprises:

The first type of shooting mode is used for the non-edge area, and the corresponding relation of all the shooting modes to the non-edge area is the same as that of the first type of shooting mode; wherein, all the shooting modes except the first type shooting mode are used for the edge area;

The second type shooting mode of the edge area corresponds to the first type defending mode;

The third type of shooting mode corresponds to a defending mode except the first type of defending mode; the second type defending mode is used for defending the edge area and the adjacent non-edge area below the edge area, or the second type defending mode is also used for a single edge area, and the third type defending mode is used for defending any two adjacent areas.

4. A football spot ball analysis system according to claim 3, wherein the system further comprises:

the mode input module is used for determining the playing mode of the dribbling so as to determine the quantity of the input dribbling strategy information and the quantity of the defending strategy information based on the playing mode;

the storage module is used for recording the input ball-pointing strategy and defending strategy and generating corresponding historical ball-pointing record information and historical defending record information;

the sound prompt module is used for outputting sound prompt information corresponding to the attack input module, the defending input module and the defending result;

And/or, the defending mode comprises: and each time the goal is defended, the initial standing position information of the target defender and/or the body action information of the target defender when the target defender contacts the football or the football enters the goal.

5. A football goal analysis method, comprising:

Responding to the received analysis request, and acquiring historical dribbling record information of the target attack party corresponding to the identification of the target attack party; the analysis request comprises an identifier of the target attack party to be analyzed, the historical ball striking record information is obtained based on ball striking strategy information of the target attack party, and the ball striking strategy information at least comprises: each time a ball is clicked, the target attack party initially stands on the position information, the motion track information and the body action information of the position of the football, and the actual shooting position information of the football in the goal;

extracting body action parameters corresponding to the set action key indexes from body action information corresponding to each time of ball striking in the history ball striking record information based on the set action key indexes;

Analyzing and processing the position information of the initial standing of the target attack party, the motion track information of the position of the football, which are corresponding to each goal, and the extracted body action parameters, and determining a reference index set for identifying the shot position information of the football in the goal when the target attack party is in the goal, so as to determine the estimated shot position information of the target attack party based on the reference index set;

and generating a defending training scheme based on the reference index set and the estimated injection position information.

6. The method of claim 5, wherein the motion key indicators include a preliminary motion key indicator and a striking motion key indicator,

The key indexes of the preparation action comprise a standing position direction index, a standing position distance index and a body orientation index;

The key indexes of the batting action comprise a hand position index, an elbow position index, a shoulder action key index, a leg highest lifting position index and a body inclination index when batting.

7. The method of claim 6, wherein the history ball record information is video data,

The step of extracting body action parameters corresponding to the set action key indexes from body action information corresponding to each time of ball striking in the history ball striking record information based on the set action key indexes comprises the following steps:

determining a previous frame of a video frame of which the football position starts to change in the history ball striking record information as a ball striking frame;

Determining a previous frame of a video frame in which the target offender's leg last starts descending before the batting frame as a batting action frame for determining the motion trail information and the body action information;

Determining a video frame of the last two legs of the target offender on the ground before the batting action frame as a preparation action frame for determining the initial standing position information of the target offender;

extracting position information of the initial standing based on the position of the target offender relative to the football in the preparation action frame;

extracting the motion trail information based on video frames from the preparation action frame to the batting action frame;

Extracting body action parameters corresponding to the preparation action key indexes based on the relative positions of the target attack party and the football and body action information in the batting action frame;

and extracting body action parameters corresponding to the key indexes of the batting action based on the body action information of the target attacking party in the batting frame.

8. The method according to claim 7, wherein the analyzing the position information of the initial standing of the target offender, the movement locus information of the soccer ball at the position of the soccer ball, and the extracted body motion parameters for each ball point, determining the reference index set for identifying the shot position information of the soccer ball in the goal when the target offender hits the ball, includes:

Determining a value interval of the body action parameter corresponding to the target attack party when each ball is clicked based on the value interval division of the body action parameter corresponding to each action key index;

Determining a preparation action key index corresponding to at least two body action parameters with the largest frequency in the same value interval as a first alternative action key index when the ball is clicked, and determining the same value interval corresponding to the first alternative action key index as a corresponding target value interval;

Determining a batting action key index corresponding to at least two physical action parameters with the largest frequency in the same value interval as a second alternative action key index when the batting is performed, and determining the same value interval corresponding to the second alternative action key index as a corresponding target value interval;

Selecting at least one action key index from the first candidate action key index and the second candidate action key index respectively as a potential reference index group;

And determining a reference index group for identifying the injection position information based on the potential reference index group, the initial standing position information of the target attack party, the motion trail information and the actual injection position information in the history ball record information.

9. The method according to claim 8, wherein the determining the reference index set for identifying the shot position information based on the actual shot position information in the potential reference index set, the position information of the initial standing of the target offender, the movement track information, the history shot record information, comprises:

traversing historical shot record information corresponding to each shot of the target attack party, and determining all actual shot position information corresponding to the shot of which the potential reference index group is positioned in the corresponding target value interval as a predicted shot position;

If the potential reference index group is located in the corresponding target value interval, the estimated injection position can be uniquely determined, and the potential reference index group located in the corresponding target value interval is determined as the reference index group corresponding to the estimated injection position;

Or if the estimated injection position can be uniquely determined by combining with one item of position information or motion trail information of the target attack party, determining one item of the potential reference index set, the position information or the motion trail information as a reference index set corresponding to the estimated injection position;

Or if the estimated injection position can be uniquely determined by combining the position information of the initial standing of the target attack party and the movement track information, determining the potential reference index group, the position information and the movement track information as a reference index group corresponding to the estimated injection position;

Or if the potential reference index group is located in the corresponding target value interval and the corresponding estimated injection position does not exist, determining that the potential reference index group does not belong to the reference index group corresponding to the estimated injection position.

10. An electronic device, comprising:

at least one processor;

and a memory communicatively coupled to the at least one processor;

Wherein the memory stores instructions executable by the at least one processor to cause the electronic device to perform the football goal analysis method of any one of claims 5 to 9.