Fraudulence monitoring system of table game and fraudulence monitoring program of table game

Abstract

To prevent damage due to a fraudulent act by continuously monitoring an amount bet by a game participant and win or lose of a game, comparing data of a probability statistic calculation or past statistical data, and detecting a suspicious pattern suspected of being fraudulent.SOLUTION: A detection system of the present disclosure stores positions and the amount of game tokens (9) that a game participant (6) places on a game table (4) based on a measurement result by a bet chip measuring device (19) in the same persons for each game participant (6) or player positions (45) of the game table (4). A management control device (50) compares an actual winning rate and a total return amount with figures obtained by a probability statistic calculation at the time of an end of the predetermined number of games to determine whether there is a significant difference therebetween and specifies any one of the game participant (6), the player position (45), the game table (4), or a room having the game table (4) where the significant difference is occurring.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a fraud monitoring system for table games in gaming facilities, and more particularly to a technique for detecting fraudulent activity. [Background technology]

[0002] Casinos and other gaming establishments are attempting to prevent various types of cheating. Gaming establishments are equipped with surveillance cameras to monitor cheating, and prevent cheating by detecting cheating in games and cheating by collecting or redeeming chips that do not match the winning or losing results from images captured by the surveillance cameras. On the other hand, it has been proposed to attach an IC tag to each chip to keep track of the number of chips bet and the total amount.

[0003] The card game monitoring system described in international application PCT / JP2015 / 000171 monitors for fraudulent activity by determining whether chips placed on a gaming table have been collected or redeemed in accordance with the outcome of the game through image analysis of the movement of the chips. Summary of the Invention [Problem to be solved by the invention]

[0004] In gaming facilities such as casinos, surveillance cameras record gaming tables, and when a player wins a large amount at once, the recorded video is played back to check for any cheating. However, there is a problem that cheating at gaming tables has become more sophisticated, and there are cases of cheating that cannot be detected simply by detecting a large amount of winnings, such as continuous wins with small amounts. In addition, conventional technology is not sufficient to prevent cheating by collusion between dealers and players.

[0005] In consideration of the above problems, the present invention aims to prevent damage caused by fraudulent activity by continuously monitoring the amounts bet by game participants and their wins and losses, and comparing the data from probability statistical calculations and past statistical data to detect suspicious patterns that may indicate fraudulent activity. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems, the fraud monitoring system of the present invention is a fraud monitoring system for table games, and includes a win / loss determination device that determines the win / loss result of each game at a gaming table, a betting chip measurement device that grasps the amount of gaming tokens bet by game participants on the gaming table in each game, and a management control device that stores the number of games played at the gaming table and, using the measurement results of the betting chip measurement device in each game, identifies and stores the positions and amounts of gaming tokens placed by game participants on the gaming table for each game participant or for each player position at the gaming table. and the management control device has a calculation function that uses the total number of games participated by the same person for each game participant or player position, the transition of the amount of chips bet in each game, the win / loss of each game, and the amount of return obtained in each game as the basis for calculation, compares the actual winning rate and total return amount of the same person for each game participant or player position at the time when a predetermined number of games have been completed with the numbers calculated as probability statistics, determines whether there is a significant difference between the two, and identifies either the game participant or player position, or the game table, or the room in which the game table is located, where there is a significant difference.

[0007] Furthermore, the management control device may further include a function for identifying a dealer in each game, and may also include a function for identifying a dealer in charge of a series of games in which the significant difference occurs.

[0008] Furthermore, the management control device may have a calculation function that factorizes the total return amount in the series of games in which the significant difference occurred, and determines whether the cause of the significant difference in the series of games in which the significant difference occurred is due to the winning rate or the amount of betting chips bet.

[0009] Furthermore, the management control device may be equipped with a calculation function that stores the betting chip amounts for each gaming game at betting positions with different return multipliers, and determines whether the cause of the significant difference in the series of games in which the significant difference occurred is related to the betting positions with different return multipliers.

[0010] Furthermore, the management control device may have a plurality of criteria for determining whether or not there is a significant difference, and may issue a different type of warning for each criterion.

[0011] Furthermore, the management control device may have different criteria for determining a significant difference for each number of games separated by a predetermined rule or for each predetermined number of games, and may issue different types of warnings.

[0012] Furthermore, the management control device may determine whether or not there is a significant difference for each game as multiple games progress, or every time a predetermined number of games have elapsed, and issue a warning when the criterion for determining whether or not there is a significant difference is exceeded.

[0013] Furthermore, the management control device may have multiple criteria for determining whether there is a significant difference, identify game participants or player positions that may be potentially cheating for each criterion, score them for each different criterion, and display a warning as a score or rank.

[0014] Furthermore, the management control device may use a conditional probability calculation in which a specific element is a fixed value to determine the probability statistical calculation value used to determine whether or not there is a significant difference.

[0015] Furthermore, the management control device may further have a function of sequentially grasping the numbers rolled on playing cards or dice in a series of games, and may infer the cause of cheating when a significant difference is determined to exist based on the data on the numbers rolled in the series of games, and may list possible causes of cheating.

[0016] Furthermore, the management control device may infer the cause of fraud when a significant difference is determined to exist, using either big data related to past games or data on past fraud, or both, and list possible causes of fraud.

[0017] In order to solve the above-mentioned problems of the related art, the fraud monitoring system of the present invention may be in the following form: That is, the fraud monitoring system for table games includes an income / expense grasping device that grasps the casino's income and expenditure for each of a plurality of gaming tables, and a management control device that stores the number of games played at each of the gaming tables and grasps and stores the transition of the casino's income and expenditure for each game, the management control device having a calculation function that uses the total number of games played at each of the gaming tables and the transition of the casino's income and expenditure for each game as the basis for calculation, compares the casino's income or expenditure at the time a predetermined number of games have been played with a figure calculated based on probability statistics to determine whether there is a significant difference between the two, and identifies either a gaming table or a room in which a gaming table is located where a significant difference occurs.

[0018] In order to solve the above-mentioned conventional problems, the fraud monitoring system of the present invention may be in the following form: That is, the fraud monitoring system for table games includes a win / loss determination device for determining the outcome of each game at a gaming table, a betting chip measurement device for grasping the amount of gaming tokens bet by game participants on the gaming table in each game, and a management control device for storing the number of games played at the gaming table and for identifying and storing the position and amount of gaming tokens placed by game participants on the gaming table for each game participant or player position of the gaming table using the measurement results of the betting chip measurement device, the management control device having an artificial intelligence taught by data of past fraudulent games as teacher data, and further including a calculation function for determining whether or not a significant difference that may be fraudulent has occurred at the time a predetermined number of games have been completed in any of the game participants or player positions, the gaming tables, or the rooms in which the gaming tables are located, based on the total number of games in which the same person has participated for each game participant or player position, the transition of the amount of betting chips in each game, the win / loss of each game, and the amount of return obtained in each game.

[0019] Furthermore, the management control device may further include a function for identifying a dealer in each gaming game, and may include a function for identifying a dealer in charge of a series of games in which the significant difference occurs.

[0020] Furthermore, the management control device may have a calculation function that factorizes the total return amount in the series of games in which the significant difference occurred, and determines whether the cause of the significant difference in the series of games in which the significant difference occurred is due to the winning rate or the amount of betting chips bet.

[0021] Furthermore, the management control device may be equipped with a calculation function that stores the betting chip amounts for each gaming game at betting positions with different return multipliers, and determines whether the cause of the significant difference in the series of games in which the significant difference occurred is related to the betting positions with different return multipliers.

[0022] Furthermore, the management control device may have a plurality of criteria for determining whether or not there is a significant difference, and may issue a different type of warning for each criterion.

[0023] Furthermore, the management control device may have different criteria for determining a significant difference for each number of games separated by a predetermined rule or for each predetermined number of games, and may issue different types of warnings.

[0024] Furthermore, the management control device may have multiple criteria for determining whether there is a significant difference, identify game participants or player positions that may be potentially cheating for each criterion, score them for each different criterion, and display a warning as a score or rank.

[0025] Furthermore, the management control device may further have a function of sequentially grasping the numbers rolled on playing cards or dice in a series of games, and may infer the cause of cheating when a significant difference is determined to exist based on the data on the numbers rolled in the series of games, and may list possible causes of cheating.

[0026] Furthermore, the management control device may infer the cause of fraud when a significant difference is determined to exist by using big data related to past games as an additional item, and list possible causes of fraud.

[0027] In order to solve the above-mentioned problems of the related art, the fraud monitoring system of the present invention may be in the following form: That is, the fraud monitoring system for table games includes an income / expense grasping device for grasping the casino's income and expenditure for each of a plurality of gaming tables, and a management control device for storing the number of games played at each of the gaming tables and grasping and storing the transition of the casino's income and expenditure for each game, the management control device having an artificial intelligence taught with data of past fraudulent games as training data, and further including a calculation function for using the total number of games played at each of the gaming tables and the transition of the casino's income and expenditure for each game as the basis for calculation, comparing the amount of income or expenditure of the casino at the time when a predetermined number of games have been completed with a calculated figure of the artificial intelligence taught with data of past fraudulent games as training data, determining whether there is a significant difference between the two, and determining the possibility of fraud occurring at a gaming table where a significant difference occurs or at a room in which a gaming table is located.

[0028] In order to solve the above-mentioned problems of the prior art, the fraud monitoring system of the present invention may be in the following form: That is, the fraud monitoring program for table games includes a win / loss determination step of grasping the win / loss results of each game at a gaming table, a betting chip grasping step of grasping the amount of gaming tokens bet by game participants on the gaming table in each game, a game progress grasping and management step of storing the number of games played at the gaming table and specifying and storing the betting patterns of gaming tokens placed by game participants on the gaming table for each game participant or player position at the gaming table using transition data of the betting chips in the betting chip grasping step, and a fraudulent activity occurrence possibility verification step, wherein the fraudulent activity occurrence possibility verification step is configured to verify the same winning / losing result of each game at a gaming table, a betting chip grasping step of grasping the amount of gaming tokens bet by game participants on the gaming table in each game, a game progress grasping and management step of specifying and storing the betting patterns of gaming tokens placed by game participants on the gaming table for each game participant or player position at the gaming table using transition data of the betting chips in the betting chip grasping step, and a fraudulent activity occurrence possibility verification step. A table game fraud monitoring program comprising a step of calculating the actual winning percentage of the same person for each game participant or player position at the end of a specified number of games and a step of calculating a total return amount by grasping the total number of games participated in by a person, the change in the amount of bet chips, and the amount of return for each game as the basis for calculation, and a significant difference determination step of comparing the obtained winning percentage and total return amount with a winning percentage and total return amount for a specified number of games based on probability statistical calculations or past data to determine whether there is a significant difference between them, and an abnormality determination step of identifying either a game participant or player position where a significant difference occurs, or a gaming table, or a room in which a gaming table is located.

[0029] Furthermore, the step of verifying the possibility of fraudulent activity may further include a dealer identification step of identifying the dealer in each gaming game, and the abnormality determination step may include a step of identifying the dealer in charge of a series of games in which a significant difference has occurred.

[0030] Furthermore, the step of verifying the possibility of occurrence of fraudulent activity may further include an artificial intelligence learning step of learning data of past fraudulent games as training data.

[0031] Furthermore, the table game fraud monitoring system may further include a device for identifying the game participants at the gaming game table using a face authentication system or information on the credit card used. Effect of the Invention

[0032] With the above configuration, the present invention continuously monitors the amounts bet by game participants and their wins and losses, and compares the data from probability statistical calculations and past statistical data to detect suspicious patterns that may indicate fraudulent activity, thereby making it possible to prevent damage caused by fraudulent activity. [Brief description of the drawings]

[0033] [Figure 1] FIG. 1 is a diagram showing an overall outline of a fraud monitoring system in a gaming arcade having a gaming table according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a perspective view of chips showing an example of a state in which chips are stacked according to an embodiment of the present invention. [Diagram 3] FIG. 3 is a front cross-sectional view of a game token in an embodiment of the present invention. [Figure 4] FIG. 4 is a perspective view of a chip showing examples of different stacking states of the chip grasped in the embodiment of the present invention. [Diagram 5] FIG. 5 is a schematic diagram of a casino facility having a plurality of gaming tables according to an embodiment of the present invention. [Figure 6] FIG. 6 is an example of a graph showing the amount of return for each game player in the embodiment of the present invention. [Figure 7] FIG. 7 is a graph showing the win rate of game participants in the embodiment of the present invention. [Figure 8] FIG. 8 is another graph showing the win rate of game participants in the embodiment of the present invention. [Figure 9] FIG. 9 is a diagram showing an outline of a fraud monitoring system for a gaming table according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0034] Below, an overview of the fraud monitoring system 10 for table games in game arcades will be described with reference to the drawings. Here, table games include various table games such as baccarat, blackjack, poker, bridge, roulette, dice, etc. Below, the explanation will be given using baccarat as an example.

[0035] 1 and 2 are diagrams showing the entirety of the fraud monitoring system 10. The gaming table 4 is a baccarat table, and is provided with betting areas 44 for each of the seven game participants 6. The betting areas 44 for each game participant 6 include areas for a player (PLAYER) 441, a banker (BANKER) 442, a tie (TIE) 443, a player pair (P) 444, and a banker pair (B) 445. Each game participant 6 places a bet by placing gaming tokens 9 in the amount to be bet on each area corresponding to the bet target.

[0036] The card distribution device 3 is used as a winning / losing determination device that determines the winning / losing results of each game at the game table 4. The card distribution device 3 is a so-called electronic shoe that is already used by those skilled in the art, and is pre-programmed with the rules of the game and is structured to read the information of the playing cards that are distributed and determine the winning / losing results of the game. For example, in a baccarat game, the banker's win, the player's win, a tie (draw), and a pair (the player's or banker's cards have the same rank) are basically determined by the ranks of two or three cards, and the determination results (winning / losing results) are displayed by the result display lamp 13. The card distribution device 3 also sends the winning / losing results of the determined games to the management control device 50.

[0037] However, as another embodiment, instead of the card dealing device 3, the drawn playing cards may be image-recognized by the camera device 2 to determine the outcome of the game, thereby forming a win / loss determination device.

[0038] The camera device 2 is provided above the gaming table 4 and photographs the betting area 44. The gaming table 4 in this embodiment has seven player positions 45 ("1", "2", "3", "5", "6", "7", and "8" in FIG. 1), and seven game participants 6 can join the game. The camera device 2 photographs the betting areas 44 of all the player positions 45. Note that it is not necessary to photograph the betting areas 44 of all the player positions 45 with one camera device 2, and photographing all the betting areas 44 may be shared among multiple camera devices 2.

[0039] The gaming tokens 9 have markings, patterns and / or colors that enable the type (value) to be identified from the surface and / or side, and the betting chip measuring device 19 identifies the position (betting target), type and number of the gaming tokens 9 placed in the betting area 44 from the color of the edge of the side of the gaming tokens 9 based on the image captured by the camera device 2. At this time, the betting chip measuring device 19 may be configured to recognize the hiding of the gaming tokens 9 due to blind spots (when part of a chip is hidden, or when the entire chip is hidden) using an existing artificial intelligence-based computer or control system, a self-learning function based on deep learning (structure) technology, etc., to accurately grasp the position, number, etc.

[0040] 4, the gaming token 9 may have a structure in which a plurality of plastic layers of different colors are laminated, with at least a colored layer 121 in the middle, with white or light-colored layers 120 laminated on both sides of the colored layer 121, forming a striped pattern in the lamination direction on the side, and the type of gaming token 9 can be identified by the colored layers. In this case, the type of gaming token 9 (10 points, 20 points, 100 points, 1000 points, etc.) can be identified by changing the color of the colored layer 121 (red, green, yellow, blue, etc.).

[0041] In such a case, as shown in FIG. 4 (compared to the conventional chip shown in FIG. 2), a multi-layered structure is formed, and the side has a clearly formed striped pattern in the stacking direction, so that the betting chip measuring device 19 can easily and accurately measure the type and number of gaming tokens 9.

[0042] Alternatively, the structure for detecting the position 8 and type of the gaming token 9 in the betting area 8 is not limited to this, and for example, as shown in FIG. 3, an RFID 125 storing information capable of identifying at least its type may be embedded in the gaming token 9, and an RFID antenna (not shown) installed on the gaming table 4 may read the ID to detect and identify the position (bet target), type, and number of the gaming token 9 placed in the betting area 44.

[0043] A gaming token tray 17 for the dealer 5 to store the gaming tokens 9 is provided in front of the gaming table 4. At least a part of the gaming token tray 17 is embedded in the gaming table 4. In the gaming token tray 17, a betting chip measuring device 19 analyzes and measures the gaming tokens 9 stacked laterally, making it possible to determine the type and number of the gaming tokens 9. Even in this case, the betting chip measuring device 19 may identify the position (bet target), type, and number of the gaming tokens 9 stored in the gaming token tray 17 from the color of the edge of the side of the gaming tokens 9 based on the image captured by the camera device 2, or, for example, as shown in Figure 3, the gaming tokens 9 may be embedded with an RFID 125 storing information capable of identifying at least their type, and the ID may be read by an RFID antenna (not shown) installed on the gaming table 4 to detect and identify the position (bet target), type, and number of the gaming tokens 9 stored in the gaming token tray 17.

[0044] The management control device 50 grasps the position, type and number of gaming tokens 9 on the gaming table 4 obtained by the betting chip measuring device 19, and has a function of calculating the casino's income and expenditure at the gaming table 4 using the win / loss results obtained from the card distribution device 3 as a win / loss determination device, stores the number of games played at the gaming table 4, and uses the measurement results of the betting chip measuring device 19 in each game to identify and store the position and amount of the gaming tokens 9 placed on the gaming table 4 by the game participants 6 for each game participant 6 or for each player position 45 of the gaming table 4.

[0045] Furthermore, the management control device 50 has a calculation function for calculating the settlement amount of each game (calculation of the casino's income and expenditure at the gaming table 4) for each game, and is structured to be able to compare and calculate whether the increase or decrease in the gaming tokens corresponding to the settlement is correct after the total amount of the gaming tokens 9 in the gaming token tray 17 of the dealer 5 at the gaming table 4 is settled with the gaming tokens 9 bet by each game player 6 after each game. The management control device 50 can constantly (or at predetermined time intervals) verify the total amount of the gaming tokens 9 in the gaming token tray 17.

[0046] The management control device 50 stores the total number of games played by each game participant 6 or the same person at each player position 45, the transition of the amount of bet chips in each game, and the amount of return obtained in each game. The return amount is the amount of money obtained by the game participant 6 as a result of betting. The management control device 50 calculates the winning percentage at the end of each game as shown in FIG. 7 using the total number of games, the transition of the amount of bet chips in each game, and the amount of return obtained in each game as the basis for calculation, and calculates the actual winning percentage and total return amount of the same person at each game participant 6 or player position 45 at the time when a predetermined number of games have been completed. The management control device 50 has a calculation function for comparing the actual winning percentage and total return amount obtained with the numbers calculated by probability statistics to determine whether there is a significant difference between them, and for identifying one of the game participants 6 or player positions 45, the game tables 4, or the rooms in which the game tables 4 are located, in which there is a significant difference.

[0047] Specifically, the management control device 50 may, for example, determine that there is a significant difference when the ratio of the win rate to the number of games is higher than a predetermined threshold value compared to the number obtained from the probability statistical calculation, may determine that there is a significant difference when the ratio of the total return amount to the number of games is higher than a predetermined threshold value compared to the number obtained from the probability statistical calculation, may determine that there is a significant difference when the ratio of the total return amount to the win rate is higher than a predetermined threshold value compared to the number obtained from the probability statistical calculation, may determine that there is a significant difference when the total return amount obtained by one person is higher than a predetermined threshold value compared to the number obtained from the probability statistical calculation, or may determine that there is a significant difference based on other criteria.

[0048] 5 shows an overall view of a room having a plurality of game tables 4. The management control device 50 stores the total number of games played, the transition of the amount of chips bet on each game, and the amount of return obtained in each game for each game participant 6 at each of the plurality of game tables 4 or for each player at each player position 45, and obtains the winning percentage and the total return amount. The management control device 50 may further store the total number of games played, the transition of the amount of chips bet on each game, and the amount of return obtained in each game for each game table 4 and for each room, and obtain the winning percentage and the total return amount. In this way, it is possible to detect cooperative cheating across all game tables or all rooms.

[0049] The management control device 50 may further have a function of identifying the dealer 5 in each game, and may have a function of identifying the dealer 5 in charge of a series of games in which the significant difference occurs. In addition, the management control device 50 stores the total number of games played by each dealer 5, the transition of the amount of chips bet on each game, and the amount of return obtained in each game, and can obtain the winning rate and total amount of return to detect dealers 5 who are likely to be leading fraudulent activities.

[0050] FIG. 6 shows the change over time in the total return amount for each game participant 6. By comparing this with probability calculation data or past statistical data, or by comparing this with other game participants 6, it is possible to detect game participants 6 who are likely to be engaging in cheating.

[0051] The management control device 50 has a calculation function that factorizes the total return amount in a series of games in which a significant difference occurred, and determines whether the cause of the significant difference in a series of games in which a significant difference occurred is the win rate or the amount of bet chips. In other words, it can determine whether the total return amount is large because the probability of winning is simply high because the same amount is bet both when winning and when losing, or whether the probability of winning is not that high, but the total return amount is large because a large amount is bet when winning and a small amount is bet when losing.

[0052] Furthermore, the management control device 50 has a calculation function for storing the betting chip amounts of each game at betting positions with different return multipliers, and for identifying whether the cause of the significant difference in a series of games where a significant difference occurred is related to the betting positions with different return multipliers. For example, in baccarat, the return multipliers are 2x for player / banker, 8x for tie, and 11x for pair. Therefore, it is possible to detect betting methods that win only with betting methods with high return multipliers such as tie or pair, which are likely to be fraudulent. In particular, a tie is well known as a fraudulent act because it can be established by consecutive 10s or picture cards (J, Q, K).

[0053] Furthermore, the management control device 50 has a plurality of criteria for determining whether or not there is a significant difference, and issues a different type of warning for each criterion. For example, if the detected significant difference is a significant difference that may occur about once every three days, a first level warning is issued, if the detected significant difference is a significant difference that may occur about once a week, a second level warning is issued, and if the detected significant difference is a significant difference that may occur about once a month, a third level warning is issued. When fraud is detected, the abnormality display lamp 16 provided on both the winning / losing determination device 3 and the gaming table 4 is turned on, and the detection of fraud is output 15 to the casino management department or the like by wireless or wired communication. A monitor or lamp that issues a warning or display in response to the judgment result may be provided in another location. Also, the configuration may be such that a VTR recorded in the camera device 2 can be played back when a warning is received.

[0054] FIG. 7 is a graph showing the change in the winning percentage of game participant 6 for each number of games. For example, in this case, the winning percentage is high from the 100th to 200th games, but if the number of games is small, this is within the range that can occur according to probability calculations, so a first level warning is issued. Since the winning percentage then drops, the warning disappears for a moment, but the winning percentage rises again by the 1000th game. In this case, since the winning percentage is high despite the large number of games, this is an unlikely event according to probability calculations, so a third level warning is issued.

[0055] Furthermore, the management control device 50 has different criteria for determining the significant difference for each number of games divided by a predetermined rule or each predetermined number of games, and issues different types of warnings. For example, in a table game using shuffled playing cards such as baccarat, 8 to 10 decks of playing cards are shuffled and stored in a box, and the shuffled playing cards are set in the card distribution device 3 to play the game. FIG. 8 shows a graph that aggregates data on the winning rate (for shoe 1S, shoe 2S, and shoe 3S) divided for each box of shuffled playing cards used in the game (called a shoe). This makes it possible to detect cheating such as knowing the order of arrangement in a specific box of shuffled playing cards, which cannot be detected by simply monitoring the overall winning rate. In the graph in the upper part of FIG. 8, which continuously monitors the overall winning rate, the winning rate is high at the beginning and there seems to be no characteristics thereafter. However, looking at the graph divided for each shoe in the lower part of FIG. 8, the winning rate is high for shoe 3S, and it is possible to detect a high possibility of cheating. In addition, the graphs show similar winning rates for shoe 1S and shoe 3S, and it is possible to detect a high possibility of cheating.

[0056] Furthermore, the management control device 50 judges whether or not there is a significant difference for each game as multiple games progress, or for each predetermined number of games, and issues a warning when the criterion for judging a significant difference is exceeded. For example, even if the win rate is 70% after 50 games, it is difficult to judge that there is a significant difference, but if the win rate is 70% after 500 games, it is judged that there is a significant difference, and suspicion of cheating increases.

[0057] The management control device 50 can calculate the occurrence probability for each game unit and determine whether there is a significant difference. If the game unit is, for example, 75 games, the probability of consecutive results from the 1st to 75th games is calculated, then the probability of consecutive results from the 2nd to 76th games is calculated, and so on, and a warning is output when the occurrence probability exceeds the criterion for determining a significant difference. The game unit is not limited to 75 games, and may be 100 games, 300 games, etc. The management control device 50 can also calculate the occurrence probability for each bet type and box number.

[0058] Furthermore, the management control device 50 has multiple criteria for determining whether there is a significant difference, and can identify game participants 6 or player positions 45 that may be potentially cheating for each criterion, score them for each different criterion, and display a warning as a score or rank.

[0059] Furthermore, the management control device 50 can use conditional probability calculations with a specific factor set as a fixed value to calculate the probability statistics used to determine whether or not there is a significant difference. For example, the number of games may be fixed and the calculations may be performed using conditional probability calculations.

[0060] The management control device 50 further has a function of sequentially grasping the numbers of playing cards or dice rolled in a series of games, and can infer the cause of cheating when a significant difference is determined based on the data of the numbers rolled in a series of games, and can list possible causes of cheating. For example, if it is known based on the data of the numbers rolled in a series of games that a winning combination occurs when the number of the first playing card drawn is 9, it can be inferred that the cause of cheating is knowing the position of the 9 playing card.

[0061] Furthermore, the management control device 50 may infer the cause of fraud when a significant difference is determined to exist, using either big data related to past games or data on past fraud, or both, and may list possible causes of fraud.

[0062] The management control device 50 may also have a calculation function that uses the total number of games played at each gaming table 4 and the transition of the casino's income and expenditure in each game as the basis for calculation, compares the casino's income or expenditure at the end of a predetermined number of games with the numbers calculated from probability statistics, and determines whether there is a significant difference between the two, and identifies either the gaming table 4 or the room in which the gaming table 4 is located where there is a significant difference. In this case, there is no need to know the amount bet by each game participant 6, and it is necessary to know only the income and expenditure of the gaming token tray.

[0063] Furthermore, the fraud monitoring system for table games may further include a device for identifying the game participants 6 at the gaming tables 4 using a face authentication system or information on the ID cards, credit cards, or the like used.

[0064] In the case of a face authentication system, an image of a face is obtained by extracting features or the like, and an identity number (ID, etc.) is assigned to identify the player. When an identified game participant 6 leaves his / her seat and sits at another gaming table, the system has a warning function that notifies the other gaming table of the presence of the identified player. Specifically, the system can warn the pit manager who manages each gaming table 4 and each table manager.

[0065] In another embodiment of the present invention, the management control device 50 stores the total number of games played by each game participant 6 or the same person for each player position 45, the transition of the bet chip amount for each game, and the return amount obtained in each game. The return amount is the amount obtained by the game participant 6 as a result of betting. The management control device 50 has an artificial intelligence taught by data of past fraudulent games as training data, and has a calculation function for determining whether or not a significant difference that may be fraudulent has occurred in any of the game participants 6 or player positions 45, the gaming tables 4, or the rooms in which the gaming tables 4 are located at the time when a predetermined number of games have been completed, based on the total number of games, the transition of the bet chip amount for each game, and the return amount obtained in each game.

[0066] For example, the management control device 50 is an intelligent control device that grasps the position (position where the player, banker, or pair is bet) and amount (type and number) of the gaming tokens 9 bet on the gaming table 4, and is an artificial intelligence-based or deep learning structure that can extract the win / loss history and the amount of chips (winning amount) of each game participant 6 obtained by the winning / losing result of each game as a unique situation (set by the casino) by comparing it with statistical data of many past games (big data). Typically, the management control device 50 is equipped with an artificial intelligence-based or deep learning structure that can extract the occurrence of a winning amount of more than a certain amount (1 million dollars) or a situation in which the amount of bet chips when losing is small and the amount of bet chips when winning is large for several games at a playing position 7 of a certain gaming table 4 as a unique situation by comparing it with statistical data of past games (big data, etc.). Furthermore, the management control device 50 can identify the cause of the fraudulent behavior by comparing it with the statistical data of past games.

[0067] The management control device 50 has artificial intelligence that has been taught using data from past fraudulent games as training data, and is therefore capable of detecting new types of fraudulent behavior that are newly detected by the artificial intelligence.

[0068] When it is determined that there is a significant difference, the management control device 50 infers the cause of the fraud by using big data related to past games as additional items, and lists possible causes of the fraud.

[0069] In addition, in a fraud monitoring system according to another embodiment of the present invention, the management control device 50 has an artificial intelligence that has been taught using data on past fraudulent games as training data, and further has a calculation function that uses the total number of games played by the same person for each game participant 6 or player position 45, the changes in the amount of chips bet on each game, the win / loss of each game, and the amount of return obtained from each game as the basis for calculation, to determine whether or not a significant difference that may indicate fraud has occurred for either the game participant 6 or player position 45, the gaming table 4, or the room in which the gaming table 4 is located at the time a predetermined number of games have been completed.

[0070] In still another embodiment of the present invention, a fraud monitoring program for table games includes a win / loss determination step of grasping the win / loss result of each game at a gaming table 4, a betting chip grasping step of grasping the amount of gaming tokens 9 bet by game participants 6 on said gaming table 4 in each game, a game progress grasping and management step of storing the number of games played at the gaming table 4 and specifying and storing the betting pattern of the gaming tokens 9 placed by the game participants 6 on the gaming table 4 for each game participant 6 or player position 45 of the gaming table 4 using transition data of the betting chips in each game at the betting chip grasping step, and a fraudulent activity occurrence possibility verification step. The system has a step of calculating the actual winning percentage of the same person for each player 6 or player position 45 at the time a predetermined number of games have been completed by grasping the total number of games played by the same person for each player 6 or player position 45, the transition of the bet chip amount, and the return amount for each game as the basis for calculation, and a step of calculating the total return amount, and is provided with a significant difference determination step of comparing the obtained winning percentage and total return amount with the winning percentage and total return amount for the predetermined number of games based on probability statistical calculation or past data to determine whether there is a significant difference between them, and an abnormality location determination step of identifying either the game participant or player position, or the game table, or the room in which the game table is located, where a significant difference has occurred.

[0071] The fraudulent activity occurrence possibility verification step further includes a dealer specifying step of specifying the dealer 5 in each game, and the abnormality location determination step includes a step of specifying the dealer 5 in charge of a series of games in which a significant difference occurs.

[0072] The step of verifying the possibility of occurrence of fraudulent activity further includes an artificial intelligence learning step of learning from data of past fraudulent games as training data.

[0073] The fraud monitoring system for table games further includes a device for identifying a game participant 6 at a gaming table 4 using a face authentication system or information on a credit card used.

[0074] Although baccarat has been described as an example above, the table game may be roulette as shown in FIG. 9. In roulette, one type of gaming token 9 with a different color is used for each player, and each game participant 6 places a bet by placing gaming token 9 of a different color in the betting area 54. The camera device 2 grasps the number and amount of gaming token 9 bet in the betting area 54, and by recognizing the color of the gaming token 9, identifies the game participant 6 who placed the gaming token 9. Winning or losing in roulette is determined by the camera device 2 recognizing the pocket 58 in the wheel 57 that contains the ball 59. Alternatively, a winning or losing determination device in which the dealer 5 manually inputs the winning or losing result may be provided.

[0075] In addition, although details will not be described, various table games such as blackjack, poker, bridge, dice, etc. may also be used. [Explanation of symbols]

[0076] 1. Playing Cards 2 Cameras, surveillance cameras 3 Card distribution device 4 Gaming Table 5 Dealer 6 Game Participants 9 Chips (gaming tokens) 44 Betting Area 441 Player Area 442 Bunker Area 443 Thailand Area 444 Player Pair Area 445 Bunker Pair Area 121 Colored layer 122 White or light-colored layer 125 RFID

Claims

1. a first detector that detects a dealt state of one or more cards dealt from a card shoe onto a gaming table to generate first detection data; a second detector that generates second detection data indicative of a bet amount and a bet position for each game participant or player position in one or more games played via the gaming table; one or more supervisory control devices connected to the first detector and the second detector; The management control device determines a return amount obtained as a result of a bet on each of the one or more games based on the bet amount identified based on the second detection data and the betting position identified based on the second detection data, and determines a winning or losing result of the game based on the first detection data; Furthermore, the management control device determines a winning percentage and a total return amount in a predetermined number of games among the one or more games based on the return amount in a predetermined number of games among the one or more games; Furthermore, the management control device determines whether or not there is a significant difference between the first numerical characteristic and the second numerical characteristic based on a comparison between the determined winning rate and the total return amount and a second numerical characteristic obtained by probability statistical calculation and the winning rate and the total return amount.

2. The fraud monitoring system of claim 1 , wherein the second detector includes a camera or an RFID sensor.

3. The fraud monitoring system of claim 1 , wherein the one or more administrative controllers are further configured to track the one or more games based on the first detection data.

4. The fraud monitoring system of claim 1, wherein the one or more management control devices are further configured to obtain a winning or losing result for each game of the one or more games determined based on the first detection data, and to identify the betting amount and the betting position based on the second detection data.

5. The fraud monitoring system of claim 1, wherein the one or more management control devices are further configured to receive a winning or losing result determined based on the first detection data, and to store information on the betting amount and the betting position for each game of the one or more games.

6. 2. The fraud monitoring system of claim 1, comprising a card shoe, said card shoe including said first detector.

7. The fraud monitoring system of claim 1, wherein the comparison between the first numerical characteristic based on the determined winning rate and total return amount and the second numerical characteristic based on the winning rate and total return amount obtained by the probability statistical calculation is performed by probability statistical calculation.

8. The fraud monitoring system of claim 1, wherein the one or more management control devices include a first management control device that determines the winning or losing result, and a second management control device that determines the betting amount and the betting position.

9. The fraud monitoring system of claim 1 , wherein the one or more supervisory controllers are configured to identify and output causes of the significant differences.

10. 2. The fraud monitoring system of claim 1, wherein the one or more management controllers are configured to perform the comparison for each of at least two consecutively played games of the one or more games, or for a predetermined number of played games of the one or more games, to determine whether there is a significant difference, and to issue an alert based on a determination of the significant difference.

11. The fraud monitoring system of claim 1 , wherein the one or more management control devices have a plurality of criteria for determining whether or not there is a significant difference, and are configured to identify according to which of the plurality of criteria the significant difference was determined.

12. 12. The fraud monitoring system of claim 11, wherein the one or more management controllers are configured to select one or more of a plurality of alerts for output in response to the determined significant difference depending on which of the plurality of criteria the significant difference is determined to.

13. The fraud monitoring system of claim 11 , wherein the one or more administrative controllers are configured to determine a respective score for each of the plurality of criteria.

14. The fraud monitoring system of claim 1 , wherein the one or more supervisory controllers are configured to determine whether there is a significant difference based on a conditional probability calculation.

15. The fraud monitoring system of claim 1, wherein the one or more management control devices are configured to determine whether or not there is a significant difference by executing an artificial intelligence program that has been taught using data from past fraudulent games as teacher data, and to identify the gaming table in which the significant difference has occurred from among the multiple gaming tables or rooms in which the gaming tables are located.

16. The fraud monitoring system of claim 15, wherein the one or more management controllers are configured to identify a dealer in charge of each game of a plurality of games and determine which of the plurality of game participants and which of the plurality of dealers, or which of a plurality of player positions, is associated with the occurrence of the significant difference.

17. The fraud monitoring system of claim 1 , wherein the bet amount is an amount of chips or currency.

18. the one or more supervisory controllers correlate the movement of chips or currency on the gaming table to one of the one or more games and determine casino revenues and expenses at the gaming table for each of the one or more games based on the correlated movement; 2. The fraud monitoring system of claim 1, further comprising: for the predetermined number of one or more games, a comparison is made based on the determined casino income and expenditure at the gaming table for each of the one or more games and the numerical value obtained by the probability statistical calculation; based on the comparison, a determination is made as to whether or not there are other significant differences between the compared numerical values ​​according to each of a plurality of criteria; if it is determined that there is another significant difference, a determination is made as to which of the plurality of criteria the significant difference was identified according to; and a determination is made as to which gaming table the other significant difference occurred from among the plurality of gaming tables or rooms in which the gaming tables are located.

19. receiving, in one or more management control devices, first detection data from a first detector based on a dealing state of one or more cards dealt from a card shoe onto a gaming table; receiving, at the one or more supervisory control devices, second detection data from a second detector for each game of one or more games played via the gaming table, the second detection data indicating a bet amount and a bet position of a game participant or player position; determining, for each of the one or more games, a return amount to be obtained by the game player in the game based on the betting amount of the game identified based on the second detection data, the betting position of the game identified based on the second detection data, and a winning or losing result of the game determined based on the first detection data; determining a winning percentage and a total return amount for the player of a predetermined number of the one or more games based on one or more return amounts associated with a predetermined number of the one or more games; and determining whether or not there is a significant difference between the first numerical characteristic and the second numerical characteristic based on a comparison between the determined winning rate and total return amount and a second numerical characteristic obtained by a probability statistical calculation and the first numerical characteristic based on the determined winning rate and total return amount.

20. 20. The method of claim 19, further comprising issuing an alert based on the determined significant difference.

Images