阅读说明：本技术 游艺场的作弊检测系统 (Cheating detection system for recreation ground ) 是由 重田泰 于 2016-08-02 设计创作，主要内容包括：本发明提供对游艺场上的游戏中的作弊行为、或进行游艺用代用币的放置和精算时的错误和作弊行为进行检测的新系统。对按照输赢结果进行游艺用代用币的回收及偿还的游戏中的作弊行为进行检测的作弊检测系统,具备：摄像机(2),其对发牌者(5)的游艺用代用币托盘(17)中收纳的游艺用代用币(9)进行拍摄；图像分析装置(12),其对通过摄像机(2)拍摄到的图像进行分析,对游艺用代用币托盘(17)中收纳的游艺用代用币(9)的数额进行检测；卡牌发放装置(13),其对游戏的输赢结果进行判定；以及控制装置(14),其对游戏的输赢结果和游艺用代用币的回收及偿还前后的游艺用代用币托盘(17)中收纳的游艺用代用币(9)的数额进行比较,来检测作弊行为。(The invention provides a new system for detecting cheating actions in games on a casino, or errors and cheating actions when tokens for amusement are placed and counted. A cheating detection system for detecting cheating behaviors in a game for recovering and repaying substitute coins for entertainment according to a win-or-lose result comprises: a camera (2) which shoots the token coin (9) for amusement which is stored in the token coin tray (17) for amusement of the dealer (5); an image analysis device (12) which analyzes the image captured by the camera (2) and detects the amount of the token coin for amusement (9) stored in the token coin tray for amusement (17); a card issuing device (13) for judging the win-or-lose result of the game; and a control device (14) which compares the result of the win or loss of the game with the amount of the token coins (9) for amusement stored in the token coin tray (17) for amusement before and after the collection and repayment of the token coins for amusement, and detects the cheating behavior.)

1. An abnormal situation detection system in a casino having a plurality of game tables, comprising:

an amusement token coin discriminating device for discriminating an amusement token coin placed at each play position;

a game result determination device that determines a game result of each game on the game table; and

a control device for determining a win-or-lose result at each play position by using a recognition result of the token coin recognition device for amusement and the game result determined by the game result determination device,

the controller is configured to extract an abnormal situation by comparing the win-or-lose history of each play position with the statistical data.

2. The abnormal situation detection system of claim 1,

the control device has a configuration capable of determining an individual player associated with an identification number at each play position of the gaming table, and,

when an abnormal condition is extracted, the control means associates the extracted abnormal condition with the identification number of the player extracted as the abnormal condition.

3. The abnormal situation detection system of claim 2,

the control device extracts a face image of each player to determine each player.

4. The abnormal situation detection system according to claim 2 or 3,

the control device issues a warning when the determined player leaves a play position of the gaming table or when the determined player sits on another gaming table.

5. The abnormal situation detection system according to claim 1 or 2,

the control device monitors whether or not a winner of the player has received the placed token coin for amusement and the paid out token coin for amusement after each game is completed.

Technical Field

The present invention relates to a system for detecting cheating in a game in a casino or errors and cheating when a token coin for amusement is placed and counted.

Background

In casinos such as casinos, attempts are being made to prevent various cheating acts. The game house is provided with a monitoring camera for monitoring cheating behaviors, and cheating such as cheating of games and collection and repayment of game tokens different from win-or-lose results is judged according to images obtained by the monitoring camera, so that cheating is prevented.

On the other hand, in order to grasp the number and total amount of tokens for entertainment that have been placed, it has been proposed to attach a wireless ic (rfid) tag to each token for entertainment to grasp the amount of tokens for entertainment.

In the card game monitoring system described in international publication No. 2015/107902, it is determined whether or not a token coin for game placed on a game table is collected or refunded according to a result of win or loss by analyzing a change in a token coin for game, and a cheating action is monitored.

Disclosure of Invention

Problems to be solved by the invention

The invention provides a new system for detecting cheating in games in a casino or errors and cheating when tokens for amusement are placed and counted.

Means for solving the problems

A cheating detection system according to an aspect of the present invention is a cheating detection system in a casino having a plurality of gaming tables, including: a game recording device for recording the state of a game played at the game table as an image by a camera; an image analysis device that performs image analysis on the video of the recorded game state; a win-or-lose result determination device for determining a win-or-lose result of each game on the game table; and a control device for detecting cheating on the game table by using the image analysis result of the image analysis device and the win-or-lose result judged by the win-or-lose result judgment device; the control device grasps the position, kind, and number of tokens for amusement placed by each player via the image analyzing device, and grasps the total amount of tokens for amusement in the token-for-amusement tray of the dealer of the amusement table, adds the total amount of tokens for amusement in the token-for-amusement tray before each game is cleared, subtracts the added or subtracted amount of the tokens for amusement in the game calculated from the position, kind, and number of tokens for amusement placed by all players in the game and the win/lose result of the game obtained by the win/lose result judging device, compares the due total amount of the tokens for amusement in the token-for-amusement tray after the game is cleared at the end of the game with the actual total amount of the tokens for amusement in the token-for-amusement tray at the end of the game obtained by the image analyzing device, it is determined whether there is a difference between the total amount and the actual total amount.

In the cheating detection system, the control device may grasp the position, kind, and number of tokens for play placed by each player via the image analysis device, grasp an actual total amount of tokens for play in the token-for-play tray when collection of all lost tokens for play placed by each player is completed, and compare an applied total amount of tokens for play in the token-for-play tray obtained by adding an increase in the token-for-play tray in the game to the position, kind, and number of tokens for play placed by each player before the game is cleared with the actual total amount of tokens for play in the token-for-play tray to determine whether or not there is a difference between the applied total amount and the actual total amount.

In the cheating detecting system, the control means may compare the total sum of the medals for amusement in the medal tray for amusement obtained by adding the increase amount of the medals for amusement in the game to the position, kind and number of the medals for amusement placed by the losing player before the respective games are cleared, with the actual total sum of the medals for amusement in the medal tray for amusement obtained by comparing the actual total sum of the medals for amusement in the medal tray for amusement with the actual total sum of the medals for amusement in the medal tray for amusement after the game is completed, and may determine that there is no difference between the due total sum and the actual total sum by comparing the due sum in the medal tray for amusement after the game is completed with the actual total sum of the medals for amusement in the medal tray for amusement at the time when the game is completed obtained by the image analyzing means, and may determine that there is a difference between the total sum and the actual total sum, if the payment error is determined, a payment error signal for notifying the payment error is generated.

In the cheating detection system, the amusement token tray may be provided with a collected amusement token tray for collecting and temporarily storing amusement tokens placed by the losing player, and the image analysis device and the control device may compare the applied total amount of the amusement tokens in the collected amusement token tray calculated from the position, the type, and the number of the amusement tokens placed by the losing player with the actual total amount of the amusement tokens in the collected amusement token tray, and determine whether or not there is a difference between the applied total amount and the actual total amount in the collected amusement token tray.

In the cheating detection system, the image analysis device may be configured to obtain an actual total amount of the amusement tokens in the amusement token tray after the game is completed at any one of the following times:

1) when the repayment for the won token money for amusement is completed;

2) when the cards used in the game are recovered and discarded in the discard area of the table;

3) when a predetermined button attached to the win-or-lose result determination device is pressed;

4) when the mark representing win-or-lose returns.

In the cheating detecting system, the game recording device may be configured to give an index or time to the acquired video, or may be configured to specify and reproduce a collection scene or a payment scene of the token money for amusement so that, when the control device determines that the difference does not correspond to the difference between the actual total amount of the token money for amusement grasped in the token money tray for amusement of the dealer at the amusement table and the increase/decrease amount of the token money for amusement calculated from the amount of the token money for amusement placed by all players and the win/lose result of the game.

In the cheating detection system, the image analysis device or the control device may be configured to obtain information on the type, number and position of the medals for playing placed on the game table even if the plurality of medals for playing placed on the game table are partially or entirely hidden due to a blind spot of the camera.

In the cheat detection system, the control device may be configured to perform the following operations:

1) grasping the position, type and number of tokens for amusement placed at each play position of the amusement table, comparing the win-or-lose history of each player obtained from the win-or-lose result of each game and the obtained amount of tokens for amusement with the statistical data of the past games, extracting as abnormal conditions, or

2) The amount of tokens for placing amusement when losing at the playing position of the amusement table is less than the amount of tokens for placing amusement when winning, and the statistical data of the past games are compared and extracted as abnormal conditions.

In the cheating detection system described above, the control means may be capable of identifying the player of the play position extracted as the difference or abnormal situation by means of the image analysis means.

In the cheating detection system, the control device may have a warning function of notifying the other gaming table of the presence of the specific player when the specific player is out of the seat and sits on the other gaming table.

In the cheating detection system, the control device may further include a function of determining at least one of:

1) whether or not there is no change in the game token in each game until the win-or-lose result of the game is displayed by the card dispensing device after the drawing of the card is started or after the game start operation by the dealer,

2) whether or not the dealer has not received the token coin for amusement while the dealer has collected the token coin for amusement placed by the dealer among the game participants after each game is completed,

3) whether or not a token coin for amusement is added while the dealer collects token coins for amusement placed by the loser among the game participants after each game is completed,

4) whether or not the dealer pays out the position of the token coin for amusement placed in the winner among the game participants after each game is finished,

5) after each game is finished, whether the winner among the game participants has taken the placed token for the game and the paid out token for the game.

In the cheating detecting system, the win-or-lose result judging means may be a card issuing means for issuing a card at the game table, or a control means for judging the win-or-lose result of each game based on information of the image analyzing means which reads the card issued at the game table by a camera.

ADVANTAGEOUS EFFECTS OF INVENTION

According to the cheating detection system of the present invention, cheating in the collection and repayment of a token coin for amusement according to the result of win or loss of a game can be detected.

Drawings

Fig. 1 is a diagram showing an overview of the entirety of a cheating detection system in a casino having a plurality of gaming tables according to embodiment 1 of the present invention.

Fig. 2A is a perspective view of a token coin for game showing an example of different overlapping states of token coins for game grasped in embodiment 1 of the present invention.

Fig. 2B is a perspective view of the token money for game showing an example of different overlapping states of the token money for game grasped in embodiment 1 of the present invention.

Fig. 3A is a diagram showing details of a token coin tray for amusement according to embodiment 1 of the present invention.

Fig. 3B is a diagram showing another example of the token money tray for amusement according to embodiment 1 of the present invention.

Fig. 4 is an enlarged view of a mark for explaining the dirt of the grasped medal in embodiment 1 of the present invention.

Fig. 5 is a plan view showing the front surface of a sign according to embodiment 1 of the present invention.

Fig. 6 is a plan view showing the back surface of a sign according to embodiment 1 of the present invention.

Fig. 7 is a plan view showing an overall outline of a cheating detecting system in a card game according to embodiment 2 of the present invention.

Fig. 8 is a plan view schematically showing a progress state of a certain card game in the cheat detecting system according to embodiment 2 of the present invention.

Fig. 9 is an explanatory diagram showing a case where a dealer wins a casino-side game token in a certain card game.

Fig. 10 is an explanatory diagram showing a situation in which a dealer pays out to a winning guest (game participant) in a card game according to embodiment 2 of the present invention.

Fig. 11 is an explanatory diagram showing a situation in which won guests (game participants) in a certain card game according to embodiment 2 of the present invention receive a token for amusement and payout.

Fig. 12A is an explanatory diagram of an image to be subjected to image analysis of collection of gaming tokens won by a dealer in the cheating detection system according to embodiment 2 of the present invention.

Fig. 12B is an explanatory diagram of an image to be subjected to image analysis of collection of gaming tokens won by a dealer in the cheating detection system according to embodiment 2 of the present invention.

Fig. 12C is an explanatory diagram of an image to be subjected to image analysis of collection of gaming tokens won by a dealer in the cheating detection system according to embodiment 2 of the present invention.

Fig. 13 is an explanatory diagram of an image to be subjected to image analysis for cheating collection of game tokens in the cheating detection system according to embodiment 2 of the present invention.

Fig. 14A is an explanatory diagram of an image to be subjected to image analysis of collection of gaming tokens won by the game participants in the cheating detection system according to embodiment 2 of the present invention.

Fig. 14B is an explanatory diagram of an image to be subjected to image analysis of collection of the gaming tokens won by the game participants in the cheating detection system according to embodiment 2 of the present invention.

Fig. 14C is an explanatory diagram of an image to be subjected to image analysis of collection of the gaming tokens won by the game participants in the cheating detection system according to embodiment 2 of the present invention.

Fig. 14D is an explanatory diagram of an image to be subjected to image analysis of collection of the gaming tokens won by the game participants in the cheating detection system according to embodiment 2 of the present invention.

Fig. 15 is a side sectional view of the card issuing device according to embodiment 2 of the present invention.

Fig. 16 is a diagram showing an example of a card according to embodiment 2 of the present invention.

Fig. 17 is a principal part plan view of the card-guiding part of the card-handing device according to embodiment 2 of the present invention, partially cut.

Fig. 18A is a main part side sectional view showing a card dispensing regulation device for regulating the entrance and exit of a card from a card storage part of the card dispensing device according to embodiment 2 of the present invention.

Fig. 18B is a partial side sectional view showing a modified example of the card dispensing regulation device for regulating the entrance and exit of the card from the card storage unit of the card dispensing device according to embodiment 2 of the present invention.

Fig. 19 is a diagram showing a relationship between an output waveform of a sensor and a mark in the card shoe issuing device according to embodiment 2 of the present invention.

Detailed Description

(embodiment 1)

In a casino such as a casino, there is a problem that game tokens are stacked high and placed on a game table, but a reading device for an IC tag provided under the game table cannot accurately read the total amount, and if the sensitivity of the reading device is increased, game tokens placed at different positions (win or lose from right to left of a passing position) are summed up, and the total amount of game tokens at each position cannot be grasped. Further, there is a problem that the image captured by the camera is blocked due to a blind spot or overlapping of the images from the camera, and the total amount of the token money for amusement cannot be grasped.

Moreover, although the game is frequently played in a certain card game, there are problems as follows: the cards are bent by a player's pressure on the cards (a behavior of bending the cards with the back surface facing upward and looking at the size of the face of the cards little by little while looking at the size of the face of the cards) or the like, and the size and the suit of the face of the cards cannot be determined by the image analysis from the camera.

Furthermore, the following new problems are also held: the game table is highly sophisticated in cheating, which cannot be found by simply detecting a large amount of won or the like, by a sophisticated placement method, and cannot be found by tracking the amount won with a camera. Further, cheating based on collusion of the dealer and the player cannot be sufficiently prevented by the related art.

In order to solve the above problems, the cheating detection system of the casino having a plurality of gaming tables of embodiment 1,

a cheating detection system in a casino having a plurality of gaming tables, comprising:

a game recording device for recording the progress status of the game played at the game table, including the dealer and the player, as an image by a camera;

an image analysis device for performing image analysis on the recorded video of the game progress state;

a card issuing device for determining and displaying the result of losing or winning of each game on the game table; and

and a control device for detecting a cheating action performed at the game table by using an image analysis result of the image analysis device and a win-or-lose result determined by the card issuing device.

In addition, in the cheating detection system, the card issuing device has the following structure: the card face size of the issued card can be read, and the control device is constructed as follows: the image analysis device can compare the information of the size of the card face obtained from the image of each card dealt at the game table with the information of the size of the card face read by the card dealing device to determine the coincidence or non-coincidence.

In the cheating detection system, the image analysis device or the control device is an artificial intelligence effective use type or deep learning structure capable of obtaining information on the size of the card surface of a card from the card which is released at the game table and is bent or soiled by the player.

In the cheating detection system, the control device recognizes the position, type and number of tokens for amusement placed by each player by the image analysis device, and analyzes the video of the progress state of the game by the image analysis device, thereby determining whether or not the collection of lost tokens for amusement placed by each player and the payment of the won tokens for amusement have been correctly performed in accordance with the result of win-or-lose of the game.

In the cheating detection system, the image analysis device or the control device is an artificial intelligence effective utilization type or deep learning structure capable of obtaining information on the type, number and position of the placed token coins for amusement even if a plurality of token coins for amusement placed on the amusement table are partially or entirely hidden due to the dead angle of the camera.

In the cheating detection system, the control device is an artificial intelligence effective utilization type or deep learning structure capable of comparing and calculating the amount of grasped game tokens in the game token tray of the dealer of the game table according to the result of losing and winning of the game, and after the game is finished and cleared, whether or not the amount of the grasped game tokens corresponds to the collection of lost game tokens placed by each player and whether or not the amount of the paid money for the won game tokens is increased or decreased.

In the cheating detection system, the control device is an artificial intelligence utilization type or deep learning structure capable of grasping the position and amount of token coins for amusement placed at each playing position of the amusement table, comparing the win-or-lose history of each player obtained from the win-or-lose result of each game and the amount of token coins for amusement obtained with the statistical data of the past games, and extracting the result as an abnormal situation.

In the cheating detection system, the control device is an artificial intelligence effective use type or deep learning structure capable of extracting an abnormal situation by comparing a state in which the amount of tokens for placing amusement when losing at the play position of a certain amusement table is smaller than the amount of tokens for placing amusement when winning with statistical data of a past game.

In the cheating detection system, the control device is configured to be able to specify the individual player at a play position at which the abnormal condition is extracted or a win or more by a predetermined amount is obtained by the image analysis device.

In the cheating detection system, the control device has a warning function of notifying the other game table of the existence of the specific player when the specified player leaves the seat and sits on the other game table.

Hereinafter, an outline of the entirety of the cheat-detecting system in a casino having a plurality of gaming tables according to embodiment 1 of the present invention will be described in further detail. Fig. 1 is a diagram showing an overview of the entire system, and a cheating detection system in a casino having a plurality of game tables 4 includes: a game recording device 11 for recording the progress status of the game played on the game table 4, including the player 6 and the dealer 5, as images by the plurality of cameras 2; an image analysis device 12 for performing image analysis on the video of the recorded progress state of the game; and a card issuing device 3 for determining and displaying the result of winning or losing of each game on the game table 4. The card issuing device 3 is a so-called electronic shoe that has been used by those skilled in the art, and is configured to read information on the issued card C by programming rules of the game in advance and determine win or loss of the game. For example, in a certain card game, winning of a banker, winning of a loser, and a tie (or a tie) are basically determined by the size of the face of 2 to 3 cards, and the result of the determination (win or loss) is displayed by the result display lamp 13.

The cheating detection system further includes a controller 14, and the controller 14 compares the actual size of the card face based on the image analysis result of the image analyzer 12 with the win-or-lose result determined by the card issuing device 3, and detects cheating actions (such as the total size of the card face of the issued cards and the discrepancy between the win-or-lose results) performed at the game table 4. The card issuing device 3 is configured as follows: the card C manually dealt by the dealer 5 can be read in the face size (a, 2 to 10, J, Q, K) and suit (red peach, black peach, etc.), and the control device 14 is configured as follows: the information on the size and the suit of the card face obtained by the image analyzing device 12 (using artificial intelligence) from the image of each card dispensed at the game table 4 (captured by the camera 2) can be compared with the information on the card and the suit read by the card dispensing device 3, and it can be determined that the cards match with each other. The image analysis device 12 and the control device 14 in the present cheating detection system are configured to integrally or compositely include a computer, a program, and a memory that are configured by a plurality of configurations.

The image analysis device 12 and the control device 14 are provided with an artificial intelligence effective use type or deep learning structure capable of obtaining information on the size of the face of a card even if the card C is dealt on the game table 4 and is bent or soiled by the player 6. As shown in fig. 4, the dirty card C is a plum or a black peach, which is difficult to be discriminated. Even in such a case, the determination of the design and color is possible by analyzing and judging the image using an artificial intelligence utility type computer or control system or a deep learning (construction) technique. Even if the cards are bent by the pressure of the player against the cards, which is frequently performed in a certain card game or the like, the cards before being deformed can be recognized in terms of the size of the face and the style of the cards by an artificial intelligence effective use type computer, control system, deep learning (construction) technique, or the like, using self-learning of a modified example of a plurality of images. An artificial intelligence efficient computer or control system, and a deep learning (construction) technique are well known to those skilled in the art and can be used, and a detailed description thereof will be omitted.

The control device 14 of the specific artificial intelligence utilization type or the deep learning structure can grasp, by the camera 2 and the image analyzing device 12, where the token for game 9 is placed in the placement area 8 by each player 6 (a player, a banker, or a pair), the kind of the token for game 9 placed (a value in which the token for game 9 is assigned a different number for each color), and the number of tokens. The tokens 9 for play may be overlapped with each other not only in the vertical direction but also in a staggered manner as shown in fig. 2A. In this case, when the camera 2 is positioned in the direction of the arrow X shown in fig. 2A (or when the orientation of the token coin for amusement 9 is relatively blind), it is assumed that the token coin for amusement 9 is not visible (enters the blind spot) as shown in fig. 2B. The computer, the control system, and the deep learning (construction) technique of the artificial intelligence utilization type can recognize the concealment due to the dead space of the token coin for amusement 9 (the case where a part of one token coin for amusement is concealed or the case where the whole token coin for amusement is concealed) by using the self-learning function or the like, and accurately grasp the number of tokens. In this way, since it is possible to grasp the position (free, banker, or pair) of the placement area 8 where the token 9 for game play is placed, the type (value in which different amounts are assigned to the token 9 for game play for each color) and the number of the placed token 9 for game play, the control device 14 analyzes the video of the progress state of the game determined by the card issuing device 3 in each game according to the win-or-lose result of the game by the image analyzing device 12, and determines whether or not the collection (indicated by arrow L) of the lost token placed by each player 6 has been correctly performed according to the win-or-lose result of the game and the payout (9W) of the lost token for game to the won player 6W.

The control device 14 can analyze and grasp the total amount of the game tokens 9 in the game token tray 17 of the dealer 5 of the game table 4 using the image analyzing device 12, and after the game is completed and cleared, can compare and calculate the amount of the lost game tokens 9 collected by each player 6 and the amount of the won game tokens 9W paid out to the won player 6W according to the result of the win-or-lose of the game, and increase or decrease the total amount of the game tokens 9 in the game token tray 17. The total amount of the game tokens 9 in the game token tray 17 is always grasped by means of RFID or the like, but whether the increase or decrease amount is correct or not is determined by the control device 14 by analyzing the video of the progress state of the game by the image analysis device 12. These also make efficient use of artificial intelligence efficient or deep learning constructs.

In this example, since cheating or errors are detected based on the result of the win or loss of the game, information on which kind of token coin for game 9 is placed several times to which position (free or banker or pair) in the placement area 8, information on the amount of lost token coin for game 9 collected and the amount of increase or decrease of the token coin for game 9 in the token coin tray for game 17 after the end of the refund of the won token coin for game 9, even if the change of the token coin for game 9 after the end of the game, that is, the placed token coin for game 9, is not recognized and moved to the player side or the dealer side, the cheating or errors can be detected.

Here, for example, in the case of a certain card game, the winning or losing result of the game can be determined according to the rules of the card game by reading the size of the face of the card C which is played in the game by the card shoe apparatus 3. The win or loss result of the game can be determined by capturing an image of the game table 4 with the camera 2, analyzing the image with the image analyzer 12, and comparing the analysis result with the rules of the game with the controller 14. In this case, the camera 2, the image analysis device 12, and the control device 14 constitute a win-or-lose result determination device. The information on which kind of token coin for game 9 is placed in which position (free, banker, or pair) of the placement area 8 by the player at each play position 7 can be obtained by capturing an image of the token coin for game 9 placed in the placement area 8 by the camera 2 and analyzing the image for each play position 7 by the image analysis device 12.

The increase or decrease of the amount of the amusement token coin 9 in the amusement token tray 17 before and after the collection of the lost amusement token coin 9 and the repayment of the won amusement token coin 9 can be calculated by comparing the total amount of the amusement token coin 9 in the amusement token tray 17 before the collection of the lost amusement token coin 9 and the repayment of the won amusement token coin 9 with the total amount of the amusement token coin 9 in the amusement token tray 17 after the collection of the lost amusement token coin 9 and the repayment of the won amusement token coin 9. The total amount of the game tokens 9 in the game token tray 17 before collection of the lost game tokens 9 and the repayment of the won game tokens 9, and the total amount of the game tokens 9 in the game token tray 17 after collection of the lost game tokens 9 and the repayment of the won game tokens 9 can be detected by capturing images of the game token tray 17 in which the game tokens 9 are stored by the camera 2 and analyzing the images by the image analyzer 12. Further, the total amount of the amusement token money 9 stored in the amusement token money tray 17 may be detected by embedding RFID indicating the amount of the amusement token money 9 in the amusement token money and providing an RFID reader in the amusement token money tray 17.

For example, the total amount of the amusement tokens 9 in the amusement token tray 17 before the start of the game is Bb, and the total amount of the amusement tokens 9 in the amusement token tray 17 after the end of the game, the collection of the lost amusement tokens and the end of the repayment of the won amusement tokens is Ba. In this game, the total of all the play positions 7 of the token coin 9 for amusement placed in the free area is bp, the total of all the play positions 7 of the token coin 9 for amusement placed in the dealer area is bb, and the total of all the play positions 7 of the token coin 9 for amusement placed in the tie area is bt. For example, when the result of win-loss in the game is a dealer win, Ba-Bb ═ bp-Bb + bt should be established. Alternatively, the total amount Ba of the medals 9 for amusement on the medal tray 17 for amusement after the end of the game is (Bb + bp-Bb + bt). In other cases, it is possible to determine that there is a cheat or an error in the collection or return of the token money for amusement.

Fig. 3A is a diagram showing details of the token coin tray for amusement according to the present embodiment, and fig. 3B is a diagram showing another example of the token coin tray for amusement. The amusement token tray 17 is provided with a collected amusement token tray 171 and a refund amusement token tray 172, the collected amusement token tray 171 collects and temporarily stores the amusement token 9L placed by the lost player 6L, and the refund amusement token tray 172 stores the amusement token 9W to be refunded. The image analysis device 12 and the control device 14 grasp the position, kind, and number of the tokens 9L for amusement placed by the losing player 6L, and calculate the increased amount of the tokens 0L for amusement in the game (the due amount of the tokens 9 for amusement in the collected token tray 171). Further, the image analysis device 12 and the control device 14 grasp the actual total amount of the game tokens 9 in the collected game token tray 171, and compare the actual total amount with the total amount to determine whether there is a difference.

Further, by using the game tokens 9 in the repayment game token tray 172 for repayment of the game tokens 9W for the won player 6W, it is possible to secure a sufficient time for the image analysis device 12 and the control device 14 to grasp the actual total amount of the game tokens 9 in the collected game token tray 171 after collection.

The game table 4 includes a discard area 41 and/or a discard slit 42 for discarding the cards C used in the game. At the end of each game, the cards C used in the game are collected and placed in the discard area 41 or the discard slit 42 on the game table 4 to be discarded.

The game table 4 is further provided with a mark 43 showing win or loss of the game. Fig. 5 is a plan view showing the front surface of the marker, and fig. 6 is a plan view showing the rear surface of the marker. In a card game, two kinds of marks 43a and 43b are used, namely, a flag 43a indicating a win of a player and a flag 43b indicating a win of a dealer, and when a game result is obtained, the dealer 5 turns over the flag of the winning player or the winning player. Thereby, the win or loss of the game is displayed clearly on the table. The reversed logo is restored by the dealer 5 after the collection and repayment of the amusement token 9 is completed. By resetting the flag, it also means that the next game can be started.

As described above, in the present embodiment, the control device 14 calculates the balance of the token coin for play for each game based on the amount of the token coin for play placed on the play table 4 and the win-or-lose result of the game, and verifies the increase in the balance of the token coin for play in the token coin tray 17 for play after the game. When the control device 14 detects a difference in the verification, it issues an alarm or adds a record indicating the difference to the record of the video captured by the camera 2. The casino operator can investigate the cause of the difference by confirming the video.

In the cheating detection system of the present embodiment, the increase/decrease amount of the tokens for amusement in the game calculated from the positions, kinds and number of tokens 9 for amusement placed by all players 6 in the game and the win/lose result of the game obtained by the win/lose result determination means is subtracted from the total amount of the tokens for amusement 9 in the token tray for amusement 17 before the respective games are cleared, and the due total amount of the tokens for amusement 9 in the token tray for amusement 17 after the completion of the game is compared with the actual total amount of the tokens for amusement 9 in the token tray for amusement 17 at the completion of the game obtained by the image analysis means 12, and it is determined whether or not there is a difference between the due total amount and the actual total amount.

The control device 14 grasps the position, type, and number of tokens for amusement placed by each player via the image analysis device 12, grasps the actual total amount of tokens for amusement in the token tray for amusement when the collection of all the lost tokens for amusement placed by each player is completed, and determines whether or not there is a difference between the due total amount of tokens for amusement 9 in the token tray for amusement 17 and the actual total amount of tokens for amusement 9 in the token tray for amusement 17 obtained by adding the increase amount of the token tray for amusement 17 in the game to the position, type, and number of tokens for amusement placed by the lost player, based on comparison between the due total amount of tokens for amusement 9 in the token tray for amusement 17 for amusement and the actual total amount of tokens for amusement 9 in the token tray for amusement 17.

When the total amount of the tokens 9 for amusement in the token tray 17 for amusement before being cleared from each game is compared with the actual total amount of the tokens 9 for amusement in the token tray 17 for amusement obtained by adding the added amount of the token tray 17 for amusement in the game to the position, kind and number of the tokens 9 for amusement placed by the lost player, and it is judged that there is no difference between the due total amount and the actual total amount, and the due total amount in the token tray 17 for amusement after being counted at the end of the game and the actual total amount of the tokens 9 for amusement in the token tray 17 for amusement at the end of the game obtained by the image analyzing means 12 are compared, it is judged that there is a difference between the due total amount and the actual total amount, if the payment error is determined, a payment error signal for notifying the payment error is generated.

The token-for-game tray 17 is provided with a token-for-game tray 171 for collecting and temporarily storing token-for-game 9 placed by the lost player, and the image analysis device 12 compares the total amount of token-for-game 9 to be collected in the token-for-game tray 171 for collecting token-for-game tray 171 obtained by adding the amount of tokens 9 to be played in the game calculated based on the position, kind, and number of token-for-game 9L placed by the lost player to the actual total amount of token-for-game 9 in the token-for-game tray 171 for collecting token-for-game tray 171, and determines whether or not there is a difference between the total amount to be collected and the actual total amount.

In order to analyze the record of the game in which the difference is generated in the game recording device 11 when the control device 14 determines that the difference is not a difference between the actual total amount of the gaming tokens 9 grasped in the gaming token tray 17 of the dealer 5 of the gaming table 4 and the increase/decrease amount of the gaming tokens calculated from the amount of the gaming tokens placed by all the players and the win/lose result of the game, the game recording device 11 is configured to give an index or time to the acquired video, or to specify and reproduce the collection scene or the payout scene of the gaming tokens 9.

In this way, the control device 14 obtains the total amount of the token coins for amusement in the token coin tray 17 for amusement after the clearing at the end of the game by the image analysis device 12, but the determination after the clearing in this case is a time when any of the following cases 1) to 4) occurs:

1) when the repayment for the won token coin for amusement 9 is completed;

2) when the card C used in the game is collected and discarded in the discard area 41 or the discard slit 42 of the table;

3) when a predetermined button attached to the win-or-lose result judgment device is pressed;

4) the flag 43 indicating the win-or-lose is restored.

The control device 14 is an artificial intelligence effective use type or deep learning structure as follows: the position (position for the player, the banker, or the pair) and the amount (kind and number) of tokens for amusement placed at each play position 7 of the amusement table 4 can be grasped, and the win/lose history of each player 6 obtained from the win/lose result of each game and the amount (win amount) of tokens for amusement obtained can be compared with the statistical data of many (big data) games in the past, and extracted as an abnormal situation (set by the casino). Typically, the apparatus is provided with a control device 14 of an artificial intelligence effective utilization type or a deep learning structure as follows: the occurrence of a winning amount equal to or larger than a certain amount (100 ten thousand points), or the continuation of a state in which the amount of the medals for putting the amusement is small when losing at the play position 7 of the certain amusement table 4 and the amount of the medals for putting the amusement is large when winning is continued in a plurality of games, can be extracted as an abnormal situation by comparing the amount with statistical data (big data or the like) of the past games.

Further, the control device 14 (integrated with the image analysis device 12) of the cheating detection system is configured to identify the individual player 6 at the play position 7 at which the abnormal situation is extracted or a win or more by a predetermined amount is obtained. For such player 6 identification, the image analysis device 12 obtains an image of a face by feature point extraction or the like, and assigns an identification number (ID or the like) thereto. The control device 14 has a warning function of notifying the presence of the specified player 6 to another gaming table when the specified player is out of the seat and sits on the other gaming table. Specifically, a manager who manages each game table 4 or each table responsible person (or dealer) is notified to prevent further abnormality.

In the above example, the individual player 6 may be unsure and the win-or-lose history for each play position 7 and the amount of the obtained token for amusement (winning amount) may be monitored. In this case, although each player 6 cannot be tracked and investigated when the player 6 is out of seat, it is possible to detect an abnormal situation such as a state in which the amount of the token money for putting game is small when losing at the specific play position 7 of one game table 4 and the amount of the token money for putting game is large when winning continues in a plurality of games. When such a play position 7 is detected, it is suspected that a cheat or an error is present at the play position 7. By verifying the video in which the play position 7 is captured, cheating or an error can be found.

Specifically, the camera 2 is provided to photograph at least the token coin for amusement 9 placed in the placement area 8 of the amusement table 4. The image analysis device 12 analyzes the image captured by the camera 2, and detects, for each user position 7, which position of the free house, the banker, and the tie of the placement area 8 the token coin for amusement and the amount of the placed token coin for amusement are placed. The card dispensing device 3 functions as a win-or-lose result determination device for determining a win-or-lose result of the game. The controller 14 records (monitors) a win-or-lose history of each play position 7 and the amount of the game tokens (the token acquisition amount for the game) obtained based on the position (the player, the banker, or the tie) in the placement area 8 where the token 9 for the game is placed and the win-or-lose result of the game. In addition, only one of the win-or-lose history and the token money earning amount for amusement may be recorded. When the win-or-lose history and/or the history of the amount of tokens obtained for the game is compared with the statistical data of a plurality of (big data) games in the past to determine an abnormal situation (set by the casino), the controller 14 determines the player position 7 as a play position where cheating is suspected.

In the case where a cheating action is suspected for a certain player position 7, the cheating detection system may generate an alarm (light, sound, or vibration) in a manner that is at least noticeable to the dealer at that point in time. This can prevent continuation of the cheating act by interrupting the subsequent game or the like at least at the location. In addition, information indicating that a cheating action is suspected may be added to the video captured and recorded by the camera 2. Thus, the reason of the suspicious cheating behavior can be investigated by confirming the video.

(embodiment 2)

There is a certain card game in a large number of table games played in casinos such as casinos. These games use a standard deck of 52 playing cards, and dispense playing cards onto a game table from a card dispensing device having a plurality of previously shuffled decks (from 6 to 9 or 10 decks), and win or loss is determined based on the number of cards dispensed (card face size) and game rules.

The dealing of the cards from the card dealing device and the settlement of the amount of money placed by the guest (game participant) are performed by a dealer or the like who is in charge of the game table. In casinos such as casinos, attempts have been made to prevent errors and cheating in the settlement of the amount of money placed by the guest (game participant).

In the card game monitoring system described in international publication WO2015/107902, a change in a token for play is read by a monitoring camera, and it is checked whether or not a set amount is paid to a winner.

The following problems are involved in the placement of a guest in a card game and the settlement of the amount of placement of the guest (game participant) by a dealer: the timing of these operations, who placed the token money for amusement, or whether or not the token money has been removed cannot be detected, and therefore, whether or not these operations are correct cannot be grasped.

In order to solve the above-described problems, a cheating detection system according to embodiment 2 is a cheating detection system for a casino having a game table, and includes:

a game monitoring device for monitoring the progress of a game played at a game table by using a camera;

an image analysis device that performs image analysis on the video obtained by the camera;

a card issuing device that determines and displays a result of losing or winning of each game on the game table; and

a control device for determining the position of the token for amusement placed on the amusement table by the game participant using the analysis result of the image analysis device in each game, and determining the winner and loser among the game participants using the win-loss result;

the control device further has a function of determining at least one of:

1) whether or not there is no change in the game token in each game until the win-or-lose result of the game is displayed by the card dispensing device after the drawing of the card is started or after the game start operation by the dealer,

2) whether or not there is no change in the token coin for amusement by a person other than the dealer while the dealer collects the token coin for amusement placed by the loser among the game participants after each game is completed,

3) whether or not a token coin for amusement is added while the dealer collects token coins for amusement placed by the loser among the game participants after each game is completed,

4) whether or not the dealer pays out the position of the token coin for amusement placed in the winner among the game participants after each game is finished,

5) after each game is finished, whether the winner among the game participants has taken the placed token for the game and the paid out token for the game.

Further, the control device may be configured to detect a change in the hands of the dealer and the game participants, a change in the game tokens, or a change in the hands and a change in the game tokens by using the analysis result of the image analysis device, and determine at least one of the above 1) to 5).

Further, the control device may be configured to determine whether or not the amount of the amusement token paid out to the winner by the dealer is correct, based on the amount of the winner among the game participants.

Further, the cheating detection system for a game may further include a monitor or a lamp that receives the determination result and gives a warning or displays the result.

According to the cheating detection system of the present embodiment, the time when a token coin for amusement is placed, who the token coin for amusement is placed by, or whether the token coin for amusement is taken away can be detected with respect to placement of a guest in a certain card game and the amount of money placed by a dealer for the guest (game participant), and errors and cheating actions can be detected, and a warning or display can be given, thereby preventing the recurrence.

Before proceeding to the detailed description of the present embodiment, a flow of a certain card game performed in a casino such as a casino will be described. In embodiment 2, the same components as those in embodiment 1 will be denoted by the same reference numerals.

As shown in fig. 7, on the game table 4, a guest (game participant/player) 6 sits on a play position (chair) 7 in a face-to-face relationship with the dealer 5. The guest (game participant) 6 places (hereinafter, this is referred to as "placement") a token coin 9 for amusement in the placement area 8 in front of the guest, as a result of winning or losing a certain card game, which of a PLAYER (PLAYER) and a BANKER (BANKER) wins or becomes a TIE (TIE). The dealer 5 also performs an operation (the state shown in fig. 7) such as shouting "No More Bet (placement acceptance end)" to measure the time for ending the placement by the guest (game participant) 6, and moving the hand laterally. In a certain card game, during a period from when the drawing of a card is started or when the dealer 5 starts the operation of starting the game by shouting "No More Bet (placement acceptance end)", until the win-or-lose result of the game is displayed by the card issuing device 3, the guest (game participant) 6 cannot move a token for play, place an additional token for play, or retrieve the placed token for play.

Thereafter, the playing cards 1 are drawn out one by one from the card issuance apparatus 3 to the game table 4 with the back surfaces thereof facing upward. First, 4 cards are extracted, and as shown in (r) to (r) of fig. 8, the first card is assigned to a PLAYER (PLAYER), the second card is assigned to a BANKER (BANKER), the third card is assigned to a PLAYER (PLAYER), and the fourth card is assigned to a BANKER (BANKER), and the cards are divided into areas 10 (PLAYER area 10P and BANKER area 10B) before the dealer 5 on the game table 4. Then, the fifth card 1 and the sixth card 1 are drawn by the dealer 5 according to the face sizes (numbers) of the first to fourth cards 1 and the conditions in the detailed rules of a certain card game, and these cards are owned by the PLAYER (PLAYER) or the BANKER (BANKER). Subsequently, the win or loss of the game is determined based on the face size (number) of the first to fourth cards 1 (the fifth and sixth cards are counted as appropriate) and the detailed rules of a certain card game. Here, the card issuing device 3 is configured to program rules of a game in advance, and read information (card face size (number) and suit) of the issued cards 1 to determine win or loss of the game. Whether or not the win-or-lose determination result (win-or-lose result) determined by the card issuance apparatus 3 matches the win-or-lose result determined by the dealer or the like as described above is determined.

The following describes an overview of the overall cheat detection system in a casino game according to an embodiment of the present invention. Fig. 7 is a diagram showing an overview of the entire system, and the cheating detection system in a game in a casino includes: a game recording device 11 that records the progress status of a game played on the game table 4, including a guest (game participant) 6 and a dealer 5, as an image by the camera 2; an image analysis device 12 for performing image analysis on the video of the recorded progress state of the game; and a card issuing device 3 having a function of determining and displaying a result of winning or losing of each game on the game table 4. The card issuing device 3 is a so-called electronic shoe that has been used by those skilled in the art, and is configured to program rules of a game in advance, detect the timing at which the dealer 5 starts issuing the cards 1 at the start of each game, and read information (the face size (number) and suit) of each card 1 issued to determine win or lose of the game. For example, in a certain card game, the winning of the banker, the winning of the player, and the tie (and the round) are basically determined by the size of the face of 2 to 3 cards, and the determination result (win-or-lose result) is displayed by the display lamp 13.

The control device 14 of the cheat detecting system has a token-for-game detecting function for identifying which of the holding areas 8 on the free-house side or the dealer side of the game table 4a guest 6 (game participant) has placed a token 9 for game, using the analysis result of the image analyzing device 12 for each game. It is assumed that the position and total amount of the token coin for amusement 9 (in which of the holding areas 8 on the free house side or the dealer side the token coin for amusement 9 is placed) cannot be read in general when the token coins for amusement 9 are overlapped with each other in a staggered manner or when the token coin for amusement falls into a dead space from the position of the camera 2. The control device 14 is configured to recognize the concealment of the token coin for play 9 due to dead space (a case where a part of one token coin for play is concealed or a case where the whole token coin for play is concealed) and to accurately grasp the position, the number of tokens, and the like, by using a self-learning function or the like based on an existing artificial intelligence utility type computer, a control system, a deep learning (construction) technique, and the like. The structure for detecting the position and type of the token coin for game 9 in the placement area 8 is not limited to this, and may be configured to detect by reading an ID embedded in the token coin for game, for example.

As described above, the control device 14 can grasp the position (the position where the tokens 9 for amusement are placed for the player 6, the house, or the pair), the type (the value obtained by assigning different amounts to the tokens 9 for amusement for each color) and the number of tokens 6 placed by each player by means of the camera 2 and the image analysis device 12, and can detect which guest 6 is placed for the house (which guest 6 is placed for the house if there are a plurality of guests 6 placed for the house), and which guest 6 is placed for the house (which guest 6 is placed for the house if there are a plurality of guests 6 placed for the house). The image analysis device 12 and the control device 14 in the present cheating detection system are configured to integrally or compositely include a computer, a program, and a memory that are configured by a plurality of configurations.

The control device 14 is configured as follows: the information on the size and the suit obtained by the image analysis device 12 from the video (using the camera 2) of each card 1 dealt at the game table 4 can be compared with the information on the size and the suit read by the card dealing device 3, and it can be determined that the cards are not matched. In accordance with the win-or-lose result of the game determined by the card issuing device 3 in each game, the control device 14 analyzes the video of the progress state of the game by the image analyzing device 12, and determines whether or not the collection of the lost token coin 9 placed by the guest (game participant) 6 and the payment of the won token coin for the game to the won guest (game participant) 6 have been correctly performed in accordance with the win-or-lose result of the game.

The control device 14 has the functions shown in the following 1) to 5) according to the rules of a certain card game, and determines whether or not a cheating against the rules is not performed as a characteristic function of the invention of the present application. That is to say that the first and second electrodes,

1) in each game, whether or not there is no change in the token 9 for game use during a period from a signal for starting drawing of a card obtained from the card issuing device 3 or a game start operation performed by the dealer 5 pressing the start button 4s until the win-or-lose result of the game is displayed by the card issuing device 3 is monitored (as shown in fig. 8) by the information obtained by the image analyzing device 12 using the camera 2.

2) After each game is completed, the dealer 5 monitors whether or not the dealer 6 has taken the game tokens 9 without cheating while the dealer 5 collects the game tokens 9 placed by the loser among the game participants 6 (as shown in fig. 9) by using the information obtained by the image analysis device 12 using the camera 2.

3) After each game is completed, the image analyzing device 12 monitors whether or not a person other than the dealer 5 (winner or loser) has added the won token 9W for game or whether or not the token 9 for game has not been newly placed on the side of the winner that has not been placed, while the dealer 5 has collected the token 9 for game placed by the loser among the game participants using the information obtained by the camera 2.

4) After each game is completed, the image analysis device 12 monitors whether or not the dealer 5 has placed the token coin for game 9 for the winner among the game participants 6 using the information obtained by the camera 2, and correctly places the token coin for game 9W paid out (see fig. 10).

5) After each game is completed (the dealer 5 operates the card issuing device 3 and displays the win-or-lose result on the display lamp 13), whether or not the winner 6W among the game participants 6 has taken the placed token for game 9 and the paid-out token for game 9W is monitored by the information obtained by the image analyzing device 12 using the camera 2 (as shown in fig. 11).

The control device 14 analyzes the information obtained by the image analysis device 12 using the camera 2 as follows. That is, the monitoring of 1) to 5) is performed by detecting the change in the hands of the dealer 5 and the game participant 6, the change in the amusement token, or the change in the hands and the change in the amusement token using the analysis result of the image analysis device 12, but in this basic analysis, it is necessary to know at least who has taken the amusement token 9. The method of this analysis will be described below with reference to fig. 12 to 14.

The dealer 5 has taken an analysis of the token 9 for amusement placed by the game participant 6L (fig. 12).

The dealer 5 collects the token 9 for amusement placed in the game participant 6L lost in the game. The information obtained by the image analysis device 12 using the camera 2 is analyzed and monitored, and whether it is collected is ensured. First, a change from the present state (fig. 12A) to the absent state (fig. 12C) of the placed medal for game 9 is detected by image analysis. Then, the images (fig. 12B) of the period from the present state to the absent state of the game token 9 are analyzed. In the image (fig. 12B) of the period from the existing state to the non-existing state of the token coin for game 9, which side the hand 5h is projected from (whether from above or not in fig. 12) is analyzed, and when the hand is projected from above (or the hand appears from above and moves back to the hand), the hand 5h is determined to be the dealer 5, and when the hand is projected from a direction other than this, it is determined to be cheating, and the cheating is detected based on such a rule.

While the dealer 5 collects the token 9 for amusement placed by the game participant 6L who loses the game, it is monitored whether or not another person has taken the lost token 9 for amusement without cheating (fig. 12 and 11). As shown in fig. 13, in the image of the period from the present state to the absent state of the token 9 for game, the image analysis of the state taken by the loser 6L among the game participants 6 detects that the hand 6h protrudes or moves from the lower side (originally from the upper side) of fig. 13, determines that the hand 6h other than the dealer 5 has taken the token 9 for game, and determines that the cheating exists.

The dealer 5 correctly pays out (sets) the amusement token 9W for the won amusement token 9 and analyzes the fact that the won token is taken by the winner 6W among the game participants 6. First, the won token coin for amusement shown in fig. 14A is replaced with a token coin for amusement 9W according to the rules of the game as shown in fig. 14B. A change from the state of the diagram shown in fig. 14A to the state shown in fig. 14B is detected, and at the same time, whether or not the hand is the hand 5h of the dealer 5 is detected by image analysis. Then, the control device 4 checks whether or not the hands 6h of the winners 6W among the game participants 6 are extended (moved) to the same placement area as shown in fig. 14C, and then the token coins 9 for amusement disappear (the state of fig. 14D), based on the result of the image analysis, in accordance with the rules of the game, thereby determining whether or not there is no cheating.

Further, the control device 4 is configured to determine whether or not the amount of the amusement token paid out to the winner by the dealer 5 is correct, based on the amount of the winner 6W among the game participants 6. Specific examples are shown below. It is assumed that the positions and total amounts of the tokens 9 for amusement (in which of the placement areas 8 on the free house side or the dealer side the tokens 9 for amusement are placed) cannot be read in general when the tokens 9 for amusement are overlapped with each other in a staggered manner or when the tokens fall into a dead space from the position of the camera 2. The control device 14 is configured to recognize the concealment of the token coin for play 9 due to dead space (a case where a part of one token coin for play is concealed or a case where the whole token coin for play is concealed) and to accurately grasp the position, the number of tokens, and the like, by using a self-learning function or the like based on an existing artificial intelligence utility type computer, a control system, a deep learning (construction) technique, and the like. The structure for detecting the position 8 and the type of the token coin for game 9 in the placement area 8 is not limited to this, and may be configured to detect by reading an ID embedded in the token coin for game, for example.

As described above, the control device 14 can grasp the position 8 (the position where the token 9 for game is placed for the player 6, the player, the dealer, or the pair), the type (the value in which the token 9 for game is assigned a different amount for each color) and the number of tokens (the positions where the tokens are placed for the players), the number of guests 6 placed for the player (which guest 6 the highest number is placed when there are a plurality of guests 6 placed for the player), and the number of guests 6 placed for the dealer (which guest 6 the highest number is placed when there are a plurality of guests 6 placed for the dealer) by the camera 2 and the image analysis device 12.

Further, the control device 14 of the cheating detection system of the present game analyzes and monitors the information obtained by the image analysis device 12 using the camera 2 by the above-described technique according to the rule of a certain card game. The monitoring shown in 1) to 5) above is performed to determine whether or not the cheating against the rule is not performed. When the cheat is detected, the card issuance detection device 14C turns on the abnormality display lamps 16 provided in the card issuance device 3 and the game table 4, respectively, and outputs 15 the detected cheat to the casino management or the like wirelessly or by wire. Further, a monitor or a lamp for receiving the determination result and giving a warning or display may be provided in another place.

As described above, the control device 14 detects the cheating behavior and sends the display signal to the indicator lamp 13 or the abnormality indicator lamp 16 of the card issuance device 3 at the detected time or at an appropriate timing, but the card issuance device 3 may be activated to stop the issuance of the card 1 after the cheating or erroneous time is detected in addition to the warning.

An embodiment of a card shoe machine 3 used in the table game system of the present invention will be described below with reference to fig. 15 to 19. The card issuing apparatus 3 includes: a card storage unit 102 that stores a plurality of shuffled playing cards 1 s; a card guide section 105 that guides the shuffled playing cards 1 when the dealer 5 or the like manually draws the shuffled playing cards 1 one by one from the card storage section 102 to the game table 4; an opening 106 for taking out the card 1 guided by the card guide 105; a card detection unit (card detection sensors 22 and 23) that detects that the shuffled playing cards 1 are being extracted when the shuffled playing cards 1 are being extracted; a card reading unit 108 that reads information indicating at least a few (card face sizes) of the shuffled playing cards 1; a control unit 109 for determining win or loss of the card game based on the number (the size of the card face) of the shuffled playing cards 1 sequentially read by the card reading unit 108; a result display lamp 13 for displaying the win-or-lose result determined by the control unit 109; a dispensing restriction device 30 provided in the opening 106 and restricting the card 1 from entering and exiting the card storage section 102; and a management control unit 114 having the same functions as those of the control device 14, which are integrated, and having the following functions: when the control device 14 detects an error or a cheating action by the dealer in the game, the card dispensing device 3 is prevented from dispensing cards again at a time point after or at a predetermined timing when the error or the cheating action is detected.

Next, the dealing regulation device 30 for regulating the entrance and exit of the card 1 into and out of the card housing 102 will be described with reference to fig. 17 and 16. The dealing regulation device 30 is provided in a card guide 107 of a card guide portion 105, and the card guide portion 105 guides the cards 1 taken out one by one from an opening 106 in front of the card housing portion 102 onto the game table 4. The dispensing restricting device 30 has the following structure: when the card 1 passes through the slit 33 between the card guide section 105 and the guide cover of the card guide 107, the locking member 34 presses the card 1 to prevent the card 1 from entering and exiting the slit 33. The lock member 34 is moved as indicated by an arrow m so as to be able to take two states, a position (restricting position) where the card 1 is pressed and a passing position where the card 1 can pass, by a driving unit 35 such as an electromagnetic solenoid or a piezoelectric element. The driving unit 35 is controlled by a control unit 109 connected directly or indirectly to the control device 14 by wire or wirelessly, and moves the locking member 34 between two states, i.e., a position where the card 1 is pressed and a position where the card 1 can pass through. The control unit 109 stores a rule of a certain card game by programming in advance.

Next, a modified example of the dispensing restricting apparatus 30 will be described with reference to fig. 18B. The dispensing restricting device 40 in the modification has the following configuration: when the card 1 passes through the slit 33 between the card guide portion 105 and the card guide 107 (guide cover), the lock member 36 protrudes into the slit 33 to prevent the card 1 from moving. The lock member 36 is movable as indicated by an arrow m so as to be able to take two states, a position (restricting position) at which the card 1 is prevented from moving and a passing position at which the card 1 can pass, by a driving unit 37 such as an electromagnetic solenoid or a piezoelectric element. The driving unit 37 is controlled by a control unit 109 connected to the control device 14 to move the locking member 36 between two states, i.e., a position for preventing the card 1 from moving and a position for allowing the card 1 to pass through.

Next, the details of the code reading unit 108 that reads the code 52 indicating the number (number, face size) of the card 1 from the card 1 when the card 1 is manually drawn out from the card housing unit 102 will be described. Fig. 17 is a plan view of a main part of the card issuance apparatus 3. In the figure, the code reading unit 108 is provided in a card guide section 105, and the card guide section 105 guides the cards 1 manually taken out one by one from an opening 106 in front of the card housing section 102 onto the game table 4. The card guide 105 is an inclined surface, and card guides 107 serving as sensor covers are attached to both side edges. Each of the 2 card guides 107 can be attached and detached by a screw or the like (not shown). When the card guide 107 is removed, the sensor group 115 of the code reading unit 108 is exposed. The sensor group 115 includes 4 sensors, and includes 2 ultraviolet ray sensitive sensors (UV sensors) 20 and 21 and object detection sensors 22 and 23.

The object detection sensors 22 and 23 are optical fiber sensors for detecting the presence or absence of the card 1, and can detect the movement direction of the card 1. The object detection sensor 22 is located on the upstream side of the card guide 105 along the moving direction of the card 1, and the other object detection sensor 23 is located on the downstream side. As shown in the drawing, two object detection sensors 22, 23 are disposed on the upstream side and the downstream side with the UV sensors 20, 21 interposed therebetween. The UV sensors 20 and 21 include an LED (ultraviolet LED) emitting ultraviolet light and a detector. The mark M of the code 52 is printed on the card 1 with ultraviolet light emitting ink which emits a color when irradiated with ultraviolet light, and the card 1 is irradiated with ultraviolet light (near ultraviolet light) and reflected light of the mark M of the code 52 of the card 1 is detected by a detector. The UV sensors 20, 21 are connected to the code reading section 108 and the control section 109 by cables. The code reading unit 108 determines the combination of the marks M and the number (card size) corresponding to each code 52 based on the output signals of the detectors of the UV sensors 20 and 21.

The code reading unit 108 controls the start and end of reading by the UV sensors 20 and 21 by the control unit 109 based on the detection signals of the object detection sensors 22 and 23. The control unit 109 also determines whether or not the card 1 has normally passed through the card guide 105 based on the detection signals of the object detection sensors 22 and 23. As shown in fig. 19, marks M in the form of squares indicating the size (number) and suit (e.g., a red peach or a black peach) of the face of the card are arranged in 2 columns and 4 rows at the edge of the card 1. When the UV sensors 20 and 21 detect the mark M, an ON signal is output. The code reading unit 108 determines the relative relationship between the two signals input from the 2 UV sensors 20 and 21. Thus, the code reading unit 108 identifies the code based on the relative difference or the like of the 2 marks M detected by the 2 UV sensors 20 and 21, and identifies the number (card face size) and the type (suit) of the corresponding card 1.

The relationship between the code 52 and the outputs of the ON signals of the 2 UV sensors 20, 21 is shown in fig. 19. From the results of comparison of the relative changes in the outputs of the ON signals of the UV sensors 20, 21, a predetermined combination of the marks M can be specified. As a result, if 4 combinations of the marks M in the upper and lower 2 rows are printed in 4 rows, 256 kinds of codes of 4 kinds, i.e., 4 powers can be formed. Each of the 52 cards of the playing card is assigned to one of the 256 codes and stored as a look-up table in a memory or a program, and the code reading unit 108 is configured to specify the number (card face size) and the type (suit) of the card 1 by specifying each code 52 and using a preset look-up table (not shown). Further, since 256 kinds of codes can be associated with 52 cards in a freely combinable manner and stored in the lookup table, the combination is complicated, and the combination of 256 kinds of codes and 52 cards can be changed depending on time and place. The code is preferably printed with a paint that is visible by exposure to ultraviolet light, in a position that is not coincident with the indicia of the card's type and the index 103.

In the above embodiment, the image analyzer 12 and the controller 14 are devices having an artificial intelligence efficient use type or deep learning structure, but the image analyzer 12 and the controller 14 may analyze the image and perform the various kinds of control described above using a Scale-Invariant Feature Transform (SIFT) algorithm, a Convolutional Neural Network (CNN), deep learning (deep learning), mechanical learning (machine learning), or the like. These techniques are techniques for recognizing an object included in a captured image by performing image recognition on the captured image, and in particular, in recent years, the object is recognized with high accuracy by using a deep learning technique in which a neural network is multilayered. The deep learning technique generally identifies objects with high accuracy by overlapping layers across multiple stages at an intermediate layer between an input layer and an output layer of a neural network. This deep learning technique, in particular, a convolutional neural network, has a higher performance than identifying an object from an existing image feature amount, and is thus attracting attention.

In the convolutional neural network, an identification target image with a tag is learned, and a main object included in the identification target image is identified. When a plurality of main objects exist in a learning image, the main objects are specified by a region rectangle, and the image corresponding to the specified region is labeled for learning. Further, in the convolutional neural network, it is also possible to determine a main object in an image and the position of the object.

When the convolutional neural network is further described, in the identification flow of the object, an edge extraction process or the like is performed on the identification target image to extract a candidate region from local features, and the candidate region is input to the convolutional neural network, and is classified based on the extracted feature vector, so that the classified candidate region with the highest reliability is obtained as the identification result. The reliability is an amount indicating how much the similarity between a certain image region and a subject of an image learned together with a tag is relatively higher than the similarity of other levels.

Further, devices having an artificial intelligence effective use type or deep learning structure are described in U.S. Pat. No. 9361577, U.S. patent publication No. 2016-.

While various embodiments of the present invention have been described above, it is needless to say that the above embodiments can be modified by those skilled in the art within the scope of the present invention, and the device of the present embodiment may be modified as appropriate according to the requirements in the game to which it is applied.

Description of the symbols

1 playing card

1s multiple shuffle playing cards

2 Surveillance camera

3 card distributing device

4 recreation table

5 dealer

6 Guest (Game participant/player)

7 chair

8 Placement area

9 token coin for amusement

10 area

10P leisure area

10B dealer area

11 Game recording device

12 image analysis device

13 result display lamp

14 control device

14C card issuing and detecting device

15 output (abnormality determination result, etc.)

16 abnormal display lamp

30 dispensing restriction device

33 slit

34 locking part

35 drive part

36 latching member

37 drive part

40 dispensing restriction device

102 card accommodation part

103 index

105 card guiding part

106 opening part

107 card guide

109 control unit

112 side monitor