An intelligent monitoring method and system for billiards game data

Abstract

The present application discloses an intelligent monitoring method and system for billiards game data, which relates to the technical field of data processing and includes the following steps: obtaining a monitoring video of a target billiards game area; decomposing the monitoring video frame by frame to obtain multiple monitoring images; obtaining the cue stick posture information of a target player and the actual trajectory information of a target billiard ball; obtaining theoretical trajectory information from the historical game database of the target player according to the cue stick posture information; judging whether the actual trajectory information matches the theoretical trajectory information. If so, return to obtaining the monitoring video of the target billiards game area. If not, obtain multiple abnormal factors based on the monitoring images; input the multiple abnormal factors into a preset evaluation model to obtain a cheating evaluation value; judge whether the cheating evaluation value is greater than a preset standard value. If so, output cheating confirmation information. The present application has the advantage of improving the monitoring accuracy of billiards cheating behavior.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to an intelligent monitoring method and system for billiards game data. Background Art

[0002] In traditional billiards games, a human referee is generally used to judge the scores and violations of billiards players. It is easily affected by subjective factors and difficult to capture hidden cheating behaviors in real time, such as subtle hand movements or cheating using equipment. Existing machine vision recognition methods for assisting in billiards game data monitoring can only analyze based on the running trajectory of the billiards. Affected by the player's action habits and technical levels and other self - conditions, relying solely on the analysis of the billiards running trajectory has a large degree of contingency, and it is difficult to accurately identify whether a billiards player is cheating, and the monitoring results have a large error. Summary of the Invention

[0003] The main purpose of this application is to provide an intelligent monitoring method and system for billiards game data, aiming to solve the technical problem of the low accuracy of the existing billiards game data monitoring method in monitoring players' cheating behaviors.

[0004] To achieve the above object, this application provides an intelligent monitoring method for billiards game data, including the following steps:

[0005] Obtain a monitoring video of the target billiards game area;

[0006] Decompose the monitoring video frame by frame to obtain multiple monitoring images;

[0007] According to the multiple monitoring images, obtain the cue stick pose information of the target player and the actual trajectory information of the target billiards; wherein, the cue stick pose information includes the movement amplitude, displacement speed, and hitting angle of the cue stick, and the target billiards includes the cue ball and the ball to be hit;

[0008] According to the cue stick pose information, obtain the theoretical trajectory information from the historical game database of the target player; wherein, the historical game database stores the analysis data of the target player's same - type billiards games;

[0009] Judge whether the actual trajectory information matches the theoretical trajectory information. If so, return to obtain the monitoring video of the target billiards game area. If not, obtain multiple abnormal factors based on the monitoring images; wherein, the abnormal factor is the state parameter when the target player cheats;

[0010] Input the multiple abnormal factors into a preset evaluation model to obtain a cheating evaluation value;

[0011] Judge whether the cheating evaluation value is greater than a preset standard value. If so, output a cheating confirmation message. If not, return to obtain the monitoring video of the target billiards game area.

[0012] Optionally, multiple anomaly factors are obtained based on the monitoring image, including:

[0013] Identify the target player's image information in the monitoring image to obtain the behavior anomaly degree R of the target player; wherein, the behavior anomaly degree R represents the anomaly degree of the target player's behavior actions and micro-expressions;

[0014] According to the monitoring image, obtain the trajectory difference degree G between the actual trajectory information and the theoretical trajectory information; wherein, the trajectory difference degree G represents the difference between the actual trajectory travel and the theoretical trajectory travel of the cue ball and the struck ball;

[0015] According to the monitoring image, obtain the angle deviation degree E of the struck ball; wherein, the angle deviation degree E represents the angular deviation between the actual trajectory of the initial segment of the struck ball and the theoretical trajectory, and the actual trajectory of the initial segment is the actual movement trajectory of the struck ball from the moment of being struck to the next collision moment or the stop moment.

[0016] Optionally, the expression of the evaluation model is:

[0017] Q = (n + 1) (K1 R + K2 G + K3 E);

[0018] In the formula, Q is the cheating evaluation value, n is the historical cheating times of the target player, K1 is the first conversion coefficient, K2 is the second conversion coefficient, and K3 is the third conversion coefficient.

[0019] Optionally, the theoretical trajectory information includes the theoretical trajectory travel of the cue ball and the theoretical trajectory travel of the struck ball, and both the theoretical trajectory travel of the cue ball and the theoretical trajectory travel of the struck ball are range values;

[0020] According to the monitoring image, obtain the trajectory difference degree G between the actual trajectory information and the theoretical trajectory information, including:

[0021] According to the monitoring image, obtain the actual trajectory travel of the cue ball and the actual trajectory travel of the struck ball;

[0022] Compare the actual trajectory travel of the cue ball with the theoretical trajectory travel of the cue ball to obtain the first trajectory difference degree G1;

[0023] Compare the actual trajectory travel of the struck ball with the theoretical trajectory travel of the struck ball to obtain the second trajectory difference degree G2;

[0024] According to the first trajectory difference degree G1 and the second trajectory difference degree G2, obtain the trajectory difference degree G; wherein, G = W1 G1 + W2 G2, W1 is the first weight value, W2 is the second weight value, and W1 < W2.

[0025] Optionally, compare the actual trajectory travel of the cue ball with the theoretical trajectory travel of the cue ball to obtain a first trajectory difference degree G1, including:

[0026] Determine whether the actual trajectory travel of the cue ball is within the range of the theoretical trajectory travel of the cue ball;

[0027] If so, output the first trajectory difference degree G1 as 0;

[0028] If not, obtain a first difference between the maximum or minimum value of the actual trajectory travel of the cue ball and the theoretical trajectory travel of the cue ball, and output the first difference as the first trajectory difference degree G1.

[0029] Optionally, compare the actual trajectory travel of the object ball with the theoretical trajectory travel of the object ball to obtain a second trajectory difference degree G2, including:

[0030] Determine whether the actual trajectory travel of the object ball is within the range of the theoretical trajectory travel of the object ball;

[0031] If so, output the second trajectory difference degree G2 as 0;

[0032] If not, obtain a second difference between the maximum or minimum value of the actual trajectory travel of the object ball and the theoretical trajectory travel of the object ball, and output the second difference as the second trajectory difference degree G2.

[0033] Optionally, obtain the angular deviation degree E of the object ball according to the monitoring image, including:

[0034] According to the monitoring image, obtain the hitting state parameters of the target player to obtain the theoretical initial segment trajectory of the object ball; wherein, the initial segment theoretical trajectory is the theoretical movement trajectory of the object ball from the moment of being hit to the next collision moment or the stop moment, and the initial segment theoretical trajectory has two trajectory boundaries;

[0035] According to the monitoring image, obtain the actual initial segment trajectory of the object ball;

[0036] Determine whether the actual initial segment trajectory is between the two trajectory boundaries;

[0037] If so, output the angular deviation degree E as 0;

[0038] If not, obtain the included angle between the actual initial segment trajectory and the adjacent trajectory boundary, and output the included angle as the angular deviation degree E.

[0039] Optionally, identify the target player image information in the monitoring image to obtain the behavior abnormality degree R of the target player, including:

[0040] Identify the target player image information in the monitoring image;

[0041] Compare the target player's image information with a preset image database to obtain the behavioral action anomaly index R1 and the micro-expression anomaly index R2; among them, the image database stores the image information of the billiards game behavior norms.

[0042] Obtain the behavior anomaly degree R according to the behavioral action anomaly index R1 and the micro-expression anomaly index R2; where R = R1 + R2.

[0043] Optionally, according to the billiard cue posture information, obtain the theoretical trajectory information from the target player's historical game database, including:

[0044] Obtain the current billiards game type of the target player to obtain the historical game database of the same type.

[0045] Input the billiard cue posture information into the historical game database and perform similarity association to determine whether an associated trajectory information matching the input billiard cue posture information can be found.

[0046] If so, output the matching associated trajectory information as the theoretical trajectory information.

[0047] If not, input the billiard cue posture information into a preset trajectory prediction model to obtain the theoretical trajectory information.

[0048] To achieve the above object, the present application also provides an intelligent monitoring system for billiards game data, including:

[0049] A video acquisition module for acquiring a monitoring video of the target billiards game area.

[0050] An image decomposition module for decomposing the monitoring video frame by frame to obtain multiple monitoring images.

[0051] An image recognition module for obtaining the billiard cue posture information of the target player and the actual trajectory information of the target billiard according to multiple monitoring images; among them, the billiard cue posture information includes the action amplitude, displacement speed and hitting angle of the billiard cue, and the target billiard includes the cue ball and the struck ball.

[0052] An information matching module for obtaining the theoretical trajectory information from the target player's historical game database according to the billiard cue posture information; among them, the historical game database stores the analysis data of the target player's billiards games of the same type.

[0053] A first data processing module for determining whether the actual trajectory information matches the theoretical trajectory information. If so, return to acquire the monitoring video of the target billiards game area. If not, obtain multiple abnormal factors based on the monitoring images; where the abnormal factors are the environmental state parameters when the target player cheats.

[0054] A cheating evaluation module, configured to input multiple abnormal factors into a preset evaluation model to obtain a cheating evaluation value;

[0055] A second data processing module, configured to determine whether the cheating evaluation value is greater than a preset standard value. If so, it outputs cheating confirmation information. If not, it returns to obtain the monitoring video of the target billiards game area.

[0056] To achieve the above object, the present application further provides a computer device, which includes a memory and a processor. A computer program is stored in the memory, and the processor executes the computer program to implement the above method.

[0057] To achieve the above object, the present application further provides a computer-readable storage medium, on which a computer program is stored, and the processor executes the computer program to implement the above method.

[0058] The beneficial effects that the present application can achieve are as follows:

[0059] Based on the monitoring video of the target billiards game area, the present application can be decomposed into multiple monitoring images. Using image recognition technology, the billiard cue posture information of the target player and the actual trajectory information of the target billiard can be obtained. Combining the action amplitude, displacement speed, and hitting angle of the billiard cue in the billiard cue posture information, the theoretical trajectory information matching the billiard cue posture information can be obtained from the historical game database of the target player. Then, it is determined whether the actual trajectory information matches the theoretical trajectory information. If so, the monitoring video is obtained to continue monitoring. If not, it indicates that there is an abnormal situation suspected of cheating. To improve the monitoring accuracy, at this time, multiple abnormal factors are further obtained based on the monitoring images, and the multiple abnormal factors are input into a preset evaluation model, so as to obtain a cheating evaluation value. Then, it is determined whether the cheating evaluation value is greater than a preset standard value. If so, it indicates that there is a cheating situation, and at this time, cheating confirmation information can be output to prompt the referee. If not, the monitoring video is continued to be obtained for monitoring. In summary, the present application matches and compares the game state data of the target player with its historical game database, with strong pertinence and more accurate data matching. It can reduce the data comparison error caused by factors such as the different personal billiard skills and habits of different players. At the same time, multiple abnormal factors are comprehensively used to quantitatively calculate a cheating evaluation value, which can realize the quantitative assessment of the cheating probability, thereby improving the accuracy of monitoring the cheating behavior of players and facilitating the referee to make fair and just decisions. Description of the Drawings

[0060] To more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0061] Figure 1 It is a schematic flow chart of an intelligent monitoring method for billiards game data in an embodiment of the present application;

[0062] Figure 2 It is a schematic route diagram of the actual trajectory of the target billiard ball in an embodiment of the present application;

[0063] Figure 3 It is a schematic diagram when the actual trajectory of the initial segment deviates from the theoretical trajectory of the initial segment in an embodiment of the present application.

[0064] Reference numerals:

[0065] 110 - Actual trajectory of the cue ball, 120 - Actual trajectory of the struck ball, 130 - Trajectory boundary, 140 - Actual trajectory of the initial segment.

[0066] The realization, functional features and advantages of the object of the present application will be further described with reference to the embodiments and the drawings. Specific embodiments

[0067] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0068] It should be noted that all the directional indications (such as up, down, left, right, front, back...) in the embodiments of the present application are only used to explain the relative position relationship and movement conditions between components in a specific posture. If the specific posture changes, the directional indications will also change accordingly.

[0069] In the present application, unless otherwise clearly defined and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral one; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal communication of two components or the interaction relationship between two components, unless otherwise clearly limited. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0070] In addition, if the embodiments of the present application involve descriptions such as "first", "second", etc., the descriptions of "first", "second", etc. are only for descriptive purposes and should not be construed as indicating or implying their relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In addition, the meaning of "and / or" appearing throughout the text includes three parallel scenarios. Taking "A and / or B" as an example, it includes Scenario A, or Scenario B, or the scenario where both A and B are satisfied simultaneously. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the ability of those of ordinary skill in the art to implement. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present application. Embodiment 1

[0071] Refer to Figures 1 - 3 , this embodiment provides an intelligent monitoring method for billiards game data, including the following steps:

[0072] Obtain the monitoring video of the target billiards game area;

[0073] Decompose the monitoring video frame by frame to obtain multiple monitoring images;

[0074] According to the multiple monitoring images, obtain the cue stick posture information of the target player and the actual trajectory information of the target billiard ball; wherein, the cue stick posture information includes the action amplitude, displacement speed, and hitting angle of the cue stick, and the target billiard ball includes the cue ball and the ball to be hit;

[0075] According to the cue stick posture information, obtain the theoretical trajectory information from the historical game database of the target player; wherein, the historical game database stores the analysis data of the same type of billiards games of the target player;

[0076] Judge whether the actual trajectory information matches the theoretical trajectory information. If so, return to obtain the monitoring video of the target billiards game area. If not, obtain multiple abnormal factors based on the monitoring images; wherein, the abnormal factors are the state parameters when the target player cheats;

[0077] Input the multiple abnormal factors into a preset evaluation model to obtain a cheating evaluation value;

[0078] Judge whether the cheating evaluation value is greater than a preset standard value. If so, output a cheating confirmation message. If not, return to obtain the monitoring video of the target billiards game area.

[0079] In this embodiment, the monitoring video based on the target billiards game area can be decomposed into multiple monitoring images. The billiard cue posture information of the target player and the actual trajectory information of the target billiard ball can be obtained by using image recognition technology. Combined with the motion amplitude, displacement speed and hitting angle of the billiard cue in the billiard cue posture information, the theoretical trajectory information matching the billiard cue posture information can be obtained from the historical game database of the target player, and then it is determined whether the actual trajectory information matches the theoretical trajectory information. If so, the monitoring video is obtained to continue monitoring. If not, it indicates that there is an abnormal situation suspected of cheating. In order to improve the monitoring accuracy, multiple abnormal factors are further obtained based on the monitoring image, and the multiple abnormal factors are input into a preset evaluation model, so that a cheating evaluation value can be obtained, and then it is determined whether the cheating evaluation value is greater than a preset standard value. If so, it indicates that cheating exists. At this time, cheating confirmation information can be output to prompt the referee. If not, the monitoring video is continued to be obtained for monitoring. In summary, this embodiment is based on the matching and comparison between the target player's game status data and his historical game database, which is highly targeted and has more accurate data matching. It can reduce the data comparison error caused by factors such as the personal billiards technical level and habits of different players. At the same time, it integrates multiple abnormal factors to quantitatively calculate a cheating assessment value, which can realize the quantitative assessment of the cheating probability, thereby improving the accuracy of monitoring the player's cheating behavior, which is conducive to assisting the referee to make fair and just decisions.

[0080] It should be noted that the monitoring image also contains comprehensive characteristic image information such as billiard balls, billiard tables, billiard cues and target players, which is convenient for separately enlarging and identifying the image information required for image recognition, so as to carry out targeted image processing and recognition; the billiard balls, billiard tables, billiard cues and target players can also be separately monitored by video positioning tracking, so as to obtain corresponding image information respectively, and the recognition is more accurate.

[0081] As an optional implementation, multiple abnormal factors are obtained based on the monitoring image, including:

[0082] Identify the target player's image information in the monitoring image to obtain the target player's behavior abnormality R; wherein the behavior abnormality R represents the abnormality of the target player's behavior and micro-expression;

[0083] According to the monitoring image, the trajectory difference G between the actual trajectory information and the theoretical trajectory information is obtained; wherein the trajectory difference G represents the difference between the actual trajectory travel of the cue ball and the hit ball and the theoretical trajectory travel;

[0084] According to the monitoring image, the angle deviation E of the hit ball is obtained; wherein the angle deviation E represents the angle deviation between the actual trajectory 140 of the initial segment of the hit ball and the theoretical trajectory, and the actual trajectory 140 of the initial segment is the actual movement trajectory of the hit ball from the time of being hit to the next collision time or the stop time.

[0085] In this embodiment, based on the monitoring image, the behavioral actions and micro-expression changes of the target player can be recognized simultaneously. If the player has cheating behavior (such as interfering with the opponent's actions), there will be obvious abnormal behavioral characteristics. At the same time, considering that the cheating psychology is easily reflected through micro-expressions, the abnormal micro-expression characteristics can be combined for comprehensive consideration to obtain the behavior abnormality degree R. Since under normal hitting actions, according to the hitting angle, the amplitude of the billiard cue movement, and the number of times the billiard ball is hit, etc., the impact energy and loss situation of the billiard ball can be predicted, and thus the theoretical trajectory travel of the cue ball and the hit ball can be obtained respectively. The theoretical trajectory travel is the total length of the trajectory route that the cue ball or the hit ball passes from the starting point to the stopping point. Then, the actual trajectory travel is compared with the theoretical trajectory travel. If there is equipment cheating (such as using a modified non-compliant billiard cue), the actual trajectory travel will be significantly more or less than the theoretical trajectory travel. Therefore, by calculating the difference between the actual trajectory travel and the theoretical trajectory travel of the cue ball and the hit ball, the trajectory difference degree G can be obtained. At the same time, it is also considered that the player may use small hand movements to affect the running trajectory of the hit ball during hitting. At this time, the actual trajectory 140 of the initial segment of the hit ball can be visually reflected. Therefore, the angle deviation between the actual trajectory 140 of the initial segment and the theoretical trajectory is output as the angle deviation degree E. This embodiment comprehensively considers three abnormal factors, namely the behavior abnormality degree R, the trajectory difference degree G, and the angle deviation degree E, and the evaluation of the cheating probability is more comprehensive and sufficient, and the reference data is accurate and effective.

[0086] As an alternative embodiment, the expression of the evaluation model is:

[0087] Q = (n + 1) (K1 R + K2 G + K3 E);

[0088] In the formula, Q is the cheating evaluation value, n is the historical cheating times of the target player, K1 is the first conversion coefficient, K2 is the second conversion coefficient, and K3 is the third conversion coefficient.

[0089] In this embodiment, the sum of the behavior abnormality degree R, the trajectory difference degree G, and the angle deviation degree E is used to characterize the size of the cheating evaluation value Q. However, the attributes of the three parameters are different. Because the first conversion coefficient K1, the second conversion coefficient K2, and the third conversion coefficient K3 are respectively used to convert and adjust each parameter, so that each parameter can be quantitatively superimposed. At the same time, the first conversion coefficient K1, the second conversion coefficient K2, and the third conversion coefficient K3 may also include weight coefficients, so as to characterize the influence degree of different parameter factors on the cheating evaluation value Q. At the same time, the historical cheating times n of the target player are also introduced. The historical cheating times n can be obtained from the historical competition database of the target player. If it is detected that there is a cheating record, the cheating evaluation value Q will increase exponentially to characterize that the cheating probability of the player is greater on the premise of having a cheating record. Therefore, based on the above formula, not only the abnormal factor parameters are considered, but also the historical cheating times of the player are considered to comprehensively evaluate the cheating probability of the player. Its quantitative calculation is effective and accurate, and has good reference and guidance.

[0090] As an optional embodiment, the theoretical trajectory information includes the theoretical trajectory travel of the cue ball and the theoretical trajectory travel of the struck ball, and both the theoretical trajectory travel of the cue ball and the theoretical trajectory travel of the struck ball are range values;

[0091] According to the monitoring image, obtain the trajectory difference degree G between the actual trajectory information and the theoretical trajectory information, including:

[0092] According to the monitoring image, obtain the actual trajectory travel 110 of the cue ball and the actual trajectory travel 120 of the struck ball;

[0093] Compare the actual trajectory travel 110 of the cue ball with the theoretical trajectory travel of the cue ball to obtain the first trajectory difference degree G1;

[0094] Compare the actual trajectory travel 120 of the struck ball with the theoretical trajectory travel of the struck ball to obtain the second trajectory difference degree G2;

[0095] According to the first trajectory difference degree G1 and the second trajectory difference degree G2, obtain the trajectory difference degree G; where G = W1 G1 + W2 G2, W1 is the first weight value, W2 is the second weight value, and W1 < W2.

[0096] In this embodiment, when a player cheats using equipment, there is a probability that the actual trajectory of the cue ball and the struck ball does not match the theoretical trajectory. According to the hitting angle, movement amplitude, hitting speed, etc., the kinetic energy at the time of hitting can be comprehensively characterized, so that the kinetic energy consumption after the cue ball and the struck ball collide and subsequent collisions can be predicted, and then the range values of the theoretical trajectory of the cue ball and the theoretical trajectory of the struck ball can be predicted respectively. Then, based on the monitoring image, the actual trajectory 110 of the cue ball and the actual trajectory 120 of the struck ball can be accurately calculated, and the difference comparison is made with the maximum or minimum value of the theoretical trajectory of the cue ball and the theoretical trajectory of the struck ball respectively, so as to obtain the corresponding first trajectory difference degree G1 and the second trajectory difference degree G2. At the same time, considering that in the case of cheating, the movement trajectory of the struck ball changes more significantly than that of the cue ball, so here a greater weight value W2 is given to the second trajectory difference degree G2, so that the finally calculated trajectory difference degree G is more real and reliable, providing an accurate data basis.

[0097] As an alternative embodiment, comparing the actual trajectory 110 of the cue ball with the theoretical trajectory of the cue ball to obtain the first trajectory difference degree G1 includes:

[0098] Determine whether the actual trajectory 110 of the cue ball is within the range of the theoretical trajectory of the cue ball;

[0099] If so, output the first trajectory difference degree G1 as 0;

[0100] If not, obtain the first difference between the actual trajectory 110 of the cue ball and the maximum or minimum value of the theoretical trajectory of the cue ball, and output the first difference as the first trajectory difference degree G1.

[0101] In this embodiment, since the theoretical trajectory of the cue ball is a range value, it is necessary to first determine whether the actual trajectory 110 of the cue ball is within the range of the theoretical trajectory of the cue ball. If so, it means that the total travel of the movement trajectory of the cue ball is normal, and at this time the first trajectory difference degree G1 is output as 0. If not, at this time, if the actual trajectory 110 of the cue ball is greater than the theoretical trajectory of the cue ball, the difference between the actual trajectory 110 of the cue ball and the maximum value of the theoretical trajectory of the cue ball is the first trajectory difference degree G1. If the actual trajectory 110 of the cue ball is less than the theoretical trajectory of the cue ball, the difference between the minimum value of the theoretical trajectory of the cue ball and the actual trajectory 110 of the cue ball is the first trajectory difference degree G1, and the calculation is accurate and reliable.

[0102] As an alternative embodiment, comparing the actual trajectory 120 of the struck ball with the theoretical trajectory of the struck ball to obtain the second trajectory difference degree G2 includes:

[0103] Determine whether the actual trajectory 120 of the struck ball is within the range of the theoretical trajectory of the struck ball;

[0104] If so, the second trajectory difference degree G2 is output as 0;

[0105] If not, obtain the second difference between the maximum or minimum value of the actual trajectory travel 120 of the hit ball and the theoretical trajectory travel of the hit ball, and output the second difference as the second trajectory difference degree G2.

[0106] In this embodiment, similarly, when calculating the second trajectory difference degree G2, first determine whether the actual trajectory travel 120 of the hit ball is within the range of the theoretical trajectory travel of the hit ball. If so, output the second trajectory difference degree G2 as 0. If not, it is necessary to determine whether the actual trajectory travel 120 of the hit ball is greater than the maximum value or less than the minimum value of the theoretical trajectory travel of the hit ball, so as to calculate the corresponding second difference, and the second trajectory difference degree G2 can be accurately and effectively calculated.

[0107] As an optional embodiment, according to the monitoring image, obtain the angle deviation degree E of the hit ball, including:

[0108] According to the monitoring image, obtain the hitting state parameters of the target player to obtain the initial segment theoretical trajectory of the hit ball; wherein, the initial segment theoretical trajectory is the theoretical movement trajectory of the hit ball from the moment of being hit to the next collision moment or the stop moment, and the initial segment theoretical trajectory has two trajectory boundaries 130;

[0109] According to the monitoring image, obtain the initial segment actual trajectory 140 of the hit ball;

[0110] Judge whether the initial segment actual trajectory 140 is between the two trajectory boundaries 130;

[0111] If so, output the angle deviation degree E as 0;

[0112] If not, obtain the included angle between the initial segment actual trajectory 140 and the adjacent trajectory boundary 130, and output the included angle as the angle deviation degree E.

[0113] In this embodiment, when a player cheats by making a micro movement of the hand to touch the cue ball, the moving angle and speed of the cue ball will change. At the same time, considering that when the cue ball is used as the reference object, it is difficult to track the trajectory change of the initial segment of the cue ball when the distance between the cue ball and the struck ball is very close. Therefore, here the struck ball is used as the reference object. When the trajectory of the cue ball is abnormal, the struck ball will also become abnormal accordingly, and the struck ball generally has an obvious change in the initial segment of the trajectory. Here, the hitting state parameters of the target player (including hitting angle, amplitude of the cue action, speed, etc.) can be obtained based on the monitoring image first, so as to predict the theoretical trajectory of the initial segment of the struck ball. Considering the calculation error, two trajectory boundaries 130 are set here to represent the deviation ranges in different directions of the theoretical trajectory of the initial segment. If the actual trajectory 140 of the initial segment of the struck ball is between the two trajectory boundaries 130, it can be indicated that there is no such suspicion of cheating, that is, the output angle deviation degree E is 0. If it is not between the two trajectory boundaries 130, it means that there is such suspicion of cheating. At this time, the angle between the actual trajectory 140 of the initial segment and the adjacent trajectory boundary 130 (i.e., Figure 3 the included angle θ in

[0114] As an alternative embodiment, identify the image information of the target player in the monitoring image to obtain the behavior abnormality degree R of the target player, including:

[0115] Identify the image information of the target player in the monitoring image;

[0116] Compare the image information of the target player with a preset image database to obtain the behavior action abnormality index R1 and the micro-expression abnormality index R2; among them, the image database stores the image information of the behavior norms of the billiards game.

[0117] Obtain the behavior abnormality degree R according to the behavior action abnormality index R1 and the micro-expression abnormality index R2; where R = R1 + R2.

[0118] In this embodiment, when calculating the behavior abnormality degree R, it is comprehensively considered from two indexes of the behavior action abnormality index R1 and the micro-expression abnormality index R2. Here, the existing similarity algorithm can be combined to compare the image information of the target player with the image database, and R1 and R2 are calculated respectively according to the comparison difference degree. The sum of the two is the behavior abnormality degree R, so that the behavior abnormality degree R can be accurately quantified and calculated.

[0119] As an alternative embodiment, according to the cue posture information, obtain the theoretical trajectory information from the historical game database of the target player, including:

[0120] Obtain the current billiards game type of the target player to obtain the historical game database of the same type;

[0121] Input the cue stick attitude information into the historical game database and perform similarity association to determine whether associated trajectory information matching the input cue stick attitude information can be found;

[0122] If so, output the matching associated trajectory information as theoretical trajectory information;

[0123] If not, input the cue stick attitude information into a preset trajectory prediction model to obtain theoretical trajectory information.

[0124] In this embodiment, when obtaining the theoretical trajectory information, first identify the current type of billiards game, so as to match the historical game database of the same type participated by the target player. If no match is found, it means that the target player is participating in this type of billiards game for the first time. At this time, the general historical game database obtained by combining the historical game data of multiple relatively standard billiards players and based on big data algorithms can be used as the currently matched database. After inputting the cue stick attitude information into the historical game database, similarity association can be performed, that is, it can be judged whether associated trajectory information matching the input cue stick attitude information can be found. If so, output the matching associated trajectory information as theoretical trajectory information. If not, at this time, input the cue stick attitude information into the trained trajectory prediction model, so as to predict and obtain the corresponding theoretical trajectory information, ensuring the accuracy of the basic information. Embodiment 2

[0125] Based on the same inventive concept as the foregoing embodiment, this embodiment also provides an intelligent monitoring system for billiards game data, including:

[0126] A video acquisition module for acquiring a monitoring video of the target billiards game area;

[0127] An image decomposition module for decomposing the monitoring video frame by frame to obtain multiple monitoring images;

[0128] An image recognition module for obtaining the cue stick attitude information of the target player and the actual trajectory information of the target billiard according to the multiple monitoring images; wherein, the cue stick attitude information includes the action amplitude, displacement speed and hitting angle of the cue stick, and the target billiard includes the cue ball and the ball to be hit;

[0129] An information matching module for obtaining theoretical trajectory information from the historical game database of the target player according to the cue stick attitude information; wherein, the historical game database stores the analysis data of the target player's billiards games of the same type;

[0130] A first data processing module for judging whether the actual trajectory information matches the theoretical trajectory information. If so, return to acquiring the monitoring video of the target billiards game area. If not, obtain multiple abnormal factors based on the monitoring images; wherein, the abnormal factors are environmental state parameters when the target player cheats.

[0131] A cheating evaluation module, configured to input multiple abnormal factors into a preset evaluation model to obtain a cheating evaluation value;

[0132] A second data processing module, configured to determine whether the cheating evaluation value is greater than a preset standard value. If so, it outputs cheating confirmation information. If not, it returns to obtain the monitoring video of the target billiards game area.

[0133] For the relevant explanations and examples of each module in the device of this embodiment, reference may be made to the method of the foregoing embodiment, which will not be elaborated here. Embodiment III

[0134] Based on the same inventive concept as the foregoing embodiment, this embodiment provides a computer device, which includes a memory and a processor. A computer program is stored in the memory, and the processor executes the computer program to implement the above method. Embodiment IV

[0135] Based on the same inventive concept as the foregoing embodiment, this embodiment provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium, and the processor executes the computer program to implement the above method.

[0136] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be included in the patent protection scope of the present application by the same token.

Claims

1. An intelligent monitoring method for billiards game data, characterized in that, It includes the following steps: Obtain the monitoring video of the target billiards game area; Decompose the monitoring video frame by frame to obtain multiple monitoring images; Based on the multiple monitoring images, obtain the cue stick pose information of the target player and the actual trajectory information of the target billiards balls; wherein, the cue stick pose information includes the action amplitude, displacement speed, and hitting angle of the cue stick, and the target billiards balls include the cue ball and the struck ball; According to the cue stick pose information, obtain the theoretical trajectory information from the historical game database of the target player; wherein, the historical game database stores the analysis data of the target player's same type of billiards games; Judge whether the actual trajectory information matches the theoretical trajectory information. If so, return to obtain the monitoring video of the target billiards game area. If not, obtain multiple abnormal factors based on the monitoring images; wherein, the abnormal factors are the state parameters when the target player cheats; including: identifying the target player image information in the monitoring image to obtain the behavior abnormality degree R of the target player; wherein, the behavior abnormality degree R represents the abnormality degree of the target player's behavior actions and micro-expressions; according to the monitoring images, obtain the trajectory difference degree G between the actual trajectory information and the theoretical trajectory information; wherein, the trajectory difference degree G represents the difference between the actual trajectory travel of the cue ball and the struck ball and the theoretical trajectory travel; according to the monitoring images, obtain the angle deviation degree E of the struck ball; wherein, the angle deviation degree E represents the included angle deviation between the actual trajectory of the initial segment of the struck ball and the theoretical trajectory, and the actual trajectory of the initial segment is the actual movement trajectory of the struck ball from the moment of being struck to the next collision moment or the stop moment; according to the monitoring images, obtain the angle deviation degree E of the struck ball, including: according to the monitoring images, obtain the hitting state parameters of the target player to obtain the theoretical trajectory of the initial segment of the struck ball; wherein, the theoretical trajectory of the initial segment is the theoretical movement trajectory of the struck ball from the moment of being struck to the next collision moment or the stop moment, and the theoretical trajectory of the initial segment has two trajectory boundaries; according to the monitoring images, obtain the actual trajectory of the initial segment of the struck ball; judge whether the actual trajectory of the initial segment is between the two trajectory boundaries; if so, output the angle deviation degree E as 0; if not, obtain the included angle between the actual trajectory of the initial segment and the adjacent trajectory boundary, and output the included angle as the angle deviation degree E; Input the multiple abnormal factors into a preset evaluation model to obtain a cheating evaluation value; the expression of the evaluation model is: Q = (n + 1) * (K1 * R + K2 * G + K3 * E); In the formula, Q is the cheating evaluation value, n is the historical cheating times of the target player, K1 is the first conversion coefficient, K2 is the second conversion coefficient, and K3 is the third conversion coefficient; Judge whether the cheating evaluation value is greater than the preset standard value. If so, output the cheating confirmation information. If not, return to obtain the monitoring video of the target billiards game area.

2. The intelligent monitoring method for billiards game data according to claim 1, characterized in that, The theoretical trajectory information includes the theoretical trajectory travel of the cue ball and the theoretical trajectory travel of the struck ball, and both the theoretical trajectory travel of the cue ball and the theoretical trajectory travel of the struck ball are range values; According to the monitoring images, obtain the trajectory difference degree G between the actual trajectory information and the theoretical trajectory information, including: According to the monitoring images, obtain the actual trajectory travel of the cue ball and the actual trajectory travel of the struck ball; Compare the actual trajectory of the cue ball with the theoretical trajectory of the cue ball to obtain the first trajectory difference degree G1; Compare the actual trajectory of the struck ball with the theoretical trajectory of the struck ball to obtain the second trajectory difference degree G2; Obtain the trajectory difference degree G according to the first trajectory difference degree G1 and the second trajectory difference degree G2; where G = W1 * G1 + W2 * G2, W1 is the first weight value, W2 is the second weight value, and W1 < W2.

3. The intelligent monitoring method for billiards game data according to claim 2, wherein Compare the actual trajectory of the cue ball with the theoretical trajectory of the cue ball to obtain the first trajectory difference degree G1, including: Judge whether the actual trajectory of the cue ball is within the range of the theoretical trajectory of the cue ball; If so, output the first trajectory difference degree G1 as 0; If not, obtain the first difference between the maximum or minimum value of the actual trajectory of the cue ball and the theoretical trajectory of the cue ball, and output the first difference as the first trajectory difference degree G1.

4. The intelligent monitoring method for billiards game data according to claim 2, characterized in that, Compare the actual trajectory of the struck ball with the theoretical trajectory of the struck ball to obtain the second trajectory difference degree G2, including: Judge whether the actual trajectory of the struck ball is within the range of the theoretical trajectory of the struck ball; If so, output the second trajectory difference degree G2 as 0; If not, obtain the second difference between the maximum or minimum value of the actual trajectory of the struck ball and the theoretical trajectory of the struck ball, and output the second difference as the second trajectory difference degree G2.

5. The intelligent monitoring method for billiards game data according to claim 1, characterized in that, Identify the target player image information in the monitoring image to obtain the behavior abnormality degree R of the target player, including: Identify the target player image information in the monitoring image; Compare the target player image information with the preset image database to obtain the behavior action abnormality index R1 and the micro-expression abnormality index R2; where the image database stores the image information of the billiards game behavior specifications; Obtain the behavior abnormality degree R according to the behavior action abnormality index R1 and the micro-expression abnormality index R2; where R = R1 + R2.

6. The intelligent monitoring method for billiards game data according to claim 1, characterized in that, Obtain the theoretical trajectory information from the historical game database of the target player according to the cue stick attitude information, including: Obtain the current billiards game type of the target player to obtain the historical game database of the same type; Input the cue stick attitude information into the historical game database and perform similarity association to judge whether an associated trajectory information matching the input cue stick attitude information can be found; If so, output the matching associated trajectory information as the theoretical trajectory information; If not, input the cue stick attitude information into the preset trajectory prediction model to obtain the theoretical trajectory information.

7. An intelligent monitoring system for billiards game data, characterized in that, Including: A video acquisition module for acquiring the monitoring video of the target billiards game area; An image decomposition module for decomposing the monitoring video frame by frame to obtain multiple monitoring images; An image recognition module for obtaining the cue stick attitude information of the target player and the actual trajectory information of the target billiard according to multiple monitoring images; where the cue stick attitude information includes the action amplitude, displacement speed, and hitting angle of the cue stick, and the target billiard includes the cue ball and the struck ball; An information matching module for obtaining the theoretical trajectory information from the historical game database of the target player according to the cue stick attitude information; where the historical game database stores the analysis data of the same type of billiards games of the target player; The first data processing module is used to determine whether the actual trajectory information matches the theoretical trajectory information. If so, it returns to obtaining the monitoring video of the target billiards game area. If not, it obtains multiple abnormal factors based on the monitoring image. Among them, the abnormal factor is the state parameter when the target player cheats, including: identifying the target player image information in the monitoring image to obtain the behavior abnormality degree R of the target player. The behavior abnormality degree R represents the abnormal degree of the behavior actions and micro-expressions of the target player; obtaining the trajectory difference degree G between the actual trajectory information and the theoretical trajectory information according to the monitoring image. The trajectory difference degree G represents the difference between the actual trajectory travel and the theoretical trajectory travel of the cue ball and the struck ball; obtaining the angle deviation degree E of the struck ball according to the monitoring image. The angle deviation degree E represents the angle deviation between the actual trajectory of the initial segment of the struck ball and the theoretical trajectory. The actual trajectory of the initial segment is the actual movement trajectory of the struck ball from the moment of being struck to the next collision moment or the stop moment; obtaining the angle deviation degree E of the struck ball according to the monitoring image, including: obtaining the hitting state parameter of the target player according to the monitoring image to obtain the theoretical trajectory of the initial segment of the struck ball. The theoretical trajectory of the initial segment is the theoretical movement trajectory of the struck ball from the moment of being struck to the next collision moment or the stop moment, and the theoretical trajectory of the initial segment has two trajectory boundaries; obtaining the actual trajectory of the initial segment of the struck ball according to the monitoring image; determining whether the actual trajectory of the initial segment is between the two trajectory boundaries. If so, output the angle deviation degree E as 0. If not, obtain the included angle between the actual trajectory of the initial segment and the adjacent trajectory boundary, and output the included angle as the angle deviation degree E. The cheating evaluation module is used to input multiple abnormal factors into a preset evaluation model to obtain a cheating evaluation value. The expression of the evaluation model is: Q = (n + 1) * (K1 * R + K2 * G + K3 * E); In the formula, Q is the cheating evaluation value, n is the historical cheating times of the target player, K1 is the first conversion coefficient, K2 is the second conversion coefficient, and K3 is the third conversion coefficient. The second data processing module is used to determine whether the cheating evaluation value is greater than a preset standard value. If so, it outputs cheating confirmation information. If not, it returns to obtaining the monitoring video of the target billiards game area.

Images