HK40007719B - Fraud detection system in casino - Google Patents

Description

Cheating detection system for amusement parks

Divisional application

This application is a divisional application of patent application number 201610624380.5 filed in China on August 2, 2016, and the invention name is "Cheating Detection System for Amusement Parks."

References to related applications

This application claims the rights to patent application No. 2015-163213 filed in Japan on August 3, 2015, and patent application No. 2015-206735 filed in Japan on October 1, 2015, the contents of which are incorporated herein by reference.

Technical Field

The present invention relates to a system for detecting cheating in a game at an amusement park, or errors and cheating in placing and calculating tokens for gaming.

Background Art

Gaming venues such as card rooms (including casinos) are attempting to prevent various cheating behaviors. Gaming venues are equipped with surveillance cameras to monitor cheating. Based on images captured by the surveillance cameras, they identify cheating in the game, or cheating in the collection and redemption of gaming tokens that differ from winning or losing outcomes, thereby preventing cheating.

On the other hand, in order to grasp the number and total amount of game tokens placed, it has been proposed to attach a wireless IC (RFID) tag to each game token to grasp the amount of the game tokens.

The card game monitoring system described in International Publication No. 2015/107902 monitors cheating behavior by analyzing the changes in gaming tokens through images to determine whether gaming tokens placed on a gaming table are being collected or redeemed based on winning or losing outcomes.

Summary of the Invention

Problems to be solved by the invention

The object of the present invention is to provide a new system for detecting cheating in casino games that complies with local laws and regulations, or errors and cheating in the placement and calculation of tokens used in casinos.

Solutions to the Problem

One embodiment of the present invention is a cheating detection system for a casino having a plurality of gaming tables, comprising: a game recording device for recording the status of games played at the gaming tables as images via a camera; an image analyzing device for performing image analysis on the recorded images of the game status; a win/loss result determining device for determining the win/loss results of each game played at the gaming tables; and a control device for detecting cheating at the gaming tables using the image analysis results of the image analyzing device and the win/loss results determined by the win/loss result determining device; the control device uses the image analyzing device to determine the position, type, and number of gaming tokens placed by each player, and to determine the number of gaming tokens placed at the gaming tables. The total amount of gaming tokens in the dealer's gaming token tray is calculated by adding or subtracting the increase or decrease in gaming tokens in the game calculated based on the positions, types and numbers of gaming tokens placed by all players in the game and the win or loss result of the game obtained by the win or loss result determination device from the total amount of gaming tokens in the gaming token tray before the settlement of each game. The expected total amount of gaming tokens in the gaming token tray after settlement at the end of the game is compared with the actual total amount of gaming tokens in the gaming token tray at the end of the game obtained by the image analysis device to determine whether there is a difference between the expected total amount and the actual total amount.

In the above-mentioned cheating detection system, the control device can also grasp the position, type and number of gaming tokens placed by each player with the help of the image analysis device, and when the recovery of all lost gaming tokens placed by each player is completed, grasp the actual total amount of gaming tokens in the gaming token tray, and compare the total amount of gaming tokens in the gaming token tray before the settlement of each game with the expected total amount of gaming tokens in the gaming token tray, which is obtained by adding the position, type and number of gaming tokens placed by the losing players to the increase in the gaming token tray in the game, with the actual total amount of gaming tokens in the gaming token tray to determine whether there is a difference between the expected total amount and the actual total amount.

In the above-mentioned cheating detection system, the control device may also compare the total amount of gaming currency in the gaming currency tray before the settlement of each game, the expected total amount of gaming currency in the gaming currency tray obtained by adding the position, type and number of gaming currency placed by the losing player to the increase in the gaming currency tray in the game, with the actual total amount of gaming currency in the gaming currency tray, and determine that there is no difference between the expected total amount and the actual total amount. The control device may also compare the expected total amount in the gaming currency tray after actuation at the end of the game with the actual total amount of gaming currency in the gaming currency tray at the end of the game obtained by the image analysis device. If it is determined that there is a difference between the expected total amount and the actual total amount, it is determined that a payment error occurs and a payment error signal is generated to notify the payment error.

In the above-mentioned cheating detection system, a recycling amusement currency tray for collecting and temporarily storing the amusement currency placed by the losing player may be provided in the amusement currency tray. The image analysis device and the control device compare the expected amount of amusement currency in the recycling amusement currency tray calculated based on the position, type and number of amusement currency placed by the losing player with the actual total amount of amusement currency in the recycling amusement currency tray to determine whether there is a difference between the expected total amount and the actual total amount in the recycling amusement currency tray.

Alternatively, in the cheating detection system, the image analysis device may determine that the actual total amount of gaming tokens in the gaming token tray after settlement at the end of the game is at any one of the following times:

1) When the repayment of tokens won for gaming is completed;

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

3) When a prescribed button attached to the winning/losing result determination device is pressed;

4) When the symbols indicating winning or losing are restored.

In the above-mentioned cheating detection system, when the control device determines that the actual total amount of gaming currency held in the gaming currency tray of the dealer of the gaming table does not correspond to the increase or decrease in gaming currency calculated based on the amount of gaming currency placed by all players and the winning or losing results of the game, the game recording device can analyze the record of the game that produces the difference, and the game recording device is constructed to assign an index or time to the obtained image, or to be able to determine and reproduce the recovery scene or payment scene of gaming currency.

In the above-mentioned 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 placed gaming currency even if a portion or a whole of the gaming currency placed on the gaming table is hidden due to a blind spot of the camera.

Alternatively, in the cheating detection system, the control device may be configured to perform the following operations:

1) Assess the location, type, and number of tokens placed at each gaming position on the gaming table, compare the winning and losing history of each player and the amount of tokens obtained from the winning and losing results of each game with the statistical data of past games, and extract any abnormalities; or

2) The situation where the amount of game tokens placed at the gaming table during a loss is less than the amount of game tokens placed during a win is compared with past game statistics and extracted as an abnormal situation.

In the above-mentioned cheating detection system, the control device can also compare and determine the amount of gaming currency substitutes held in the gaming currency substitute tray of the dealer of the gaming table, and after the exchange of banknotes and gaming currency substitutes, whether the amount has been increased or decreased based on the payment amount of the gaming currency substitutes corresponding to the exchanged banknotes, or the payment amount of the banknotes corresponding to the exchanged gaming currency substitutes.

In the above-mentioned cheating detection system, the control device may further include a database that records the exchange history of banknotes and gaming currency, and may refer to the database at regular intervals or daily intervals to compare and determine whether the amount of gaming currency held in the gaming currency tray of the dealer at the gaming table has increased or decreased based on the payment amount of the gaming currency corresponding to the exchanged banknotes, or the total payment amount of the banknotes corresponding to the exchanged gaming currency.

Alternatively, in the cheating detection system, the control device may be capable of identifying the player at the playing position extracted as the difference or abnormality by means of the image analyzing device.

In the cheating detection system, the control device may have a warning function for notifying another gaming table of the presence of the specific player when the identified player leaves his seat and sits at another gaming table.

Alternatively, in the above cheating detection system, the control device may further have a function of determining at least one of the following:

1) In each game, whether there is no change in the game token currency from the time the start card is drawn or the dealer's game start operation until the card distribution device displays the game win or loss result,

2) After each game is over, during the period when the dealer collects the gaming tokens placed by the losers among the game participants, whether the losers do not take the gaming tokens,

3) After each game is over, during the period when the dealer collects the game tokens placed by the losers among the game participants, whether additional game tokens are added,

4) After each game is over, whether the dealer pays the gaming tokens placed by the winner among the game participants,

5) After each game is over, whether the winner among the game participants has taken the placed game tokens and the paid game tokens.

Alternatively, in the cheating detection system, the win/loss result determination device may be a card issuing device that issues cards at a gaming table, or a control device that determines the win/loss result of each game based on information from the image analysis device that reads cards issued at the gaming table using a camera.

Effects of the Invention

According to the cheating detection system of the present invention, it is possible to detect cheating in the collection and redemption of gaming tokens according to the winning or losing results of a game.

Furthermore, the system of the present invention can determine the card size and suit through image analysis, even if the cards are bent due to repeated squeezing by players in a card game. It can also determine the total amount and location of blind spots and overlapping game tokens. Furthermore, it can detect cheating when exchanging paper money for game tokens.

BRIEF DESCRIPTION OF THE DRAWINGS

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

FIG. 2A is a perspective view of amusement tokens, each showing examples of different stacking states of the amusement tokens understood in the first embodiment of the present invention.

FIG. 2B is a perspective view of the amusement tokens, each showing an example of different stacking states of the amusement tokens grasped in the first embodiment of the present invention.

FIG3A is a diagram showing details of the gaming token tray according to the first embodiment of the present invention.

FIG3B is a diagram showing another example of the gaming token tray according to the first embodiment of the present invention.

FIG. 4 is an enlarged view of a mark for explaining dirt on a grasped card in the first embodiment of the present invention.

FIG5A is a plan view showing the front side of the sign according to the first embodiment of the present invention.

FIG5B is a plan view showing the back side of the sign according to the first embodiment of the present invention.

FIG. 6 is an explanatory diagram showing a simplified image of the exchange state of grasped banknotes and gaming tokens in the first embodiment of the present invention.

7 is a plan view showing an overall outline of a cheating detection system for a card game according to a second embodiment of the present invention.

8 is a plan view schematically showing the progress of a certain card game in the cheating detection system according to the second embodiment of the present invention.

FIG. 9 is an explanatory diagram showing a situation in which a dealer in a certain card game collects game tokens won at a card room.

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

11 is an explanatory diagram showing how a winning guest (game participant) receives game tokens and payouts in a card game according to the second embodiment of the present invention.

12A is an explanatory diagram of an image that is the subject of image analysis, showing the recovery of gaming tokens won by the dealer in the card room in the cheating detection system according to the second embodiment of the present invention.

12B is an explanatory diagram of an image that is the subject of image analysis, showing the recovery of gaming tokens won by the dealer in the card room in the cheating detection system according to the second embodiment of the present invention.

12C is an explanatory diagram of an image that is the subject of image analysis, showing the recovery of gaming tokens won by the dealer in the card room in the cheating detection system according to the second embodiment of the present invention.

13 is an explanatory diagram of an image to be analyzed for cheating recovery of gaming tokens in the cheating detection system according to the second embodiment of the present invention.

14A is an explanatory diagram of an image to be analyzed for the recovery of gaming tokens won by a game participant in the cheating detection system according to the second embodiment of the present invention.

14B is an explanatory diagram of an image to be analyzed for recovery of gaming tokens won by a game participant in the cheating detection system according to the second embodiment of the present invention.

14C is an explanatory diagram of an image to be analyzed for the recovery of gaming tokens won by a game participant in the cheating detection system according to the second embodiment of the present invention.

14D is an explanatory diagram of an image to be analyzed for the recovery of gaming tokens won by a game participant in the cheating detection system according to the second embodiment of the present invention.

FIG15 is a side sectional view of the card issuing device in the second embodiment of the present invention.

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

17 is a plan view of a main portion of a card guide portion of the card dispensing device according to the second embodiment of the present invention, with the card guide portion partially cut away.

18A is a side sectional view of a main part showing a distribution restriction device that restricts the entry and exit of cards from a card storage portion of a card distribution device according to a second embodiment of the present invention.

18B is a side sectional view of a main part showing a modified example of the issuing restriction device for restricting the entry and exit of cards from the card storage portion of the card issuing device according to the second embodiment of the present invention.

FIG. 19 is a diagram showing the relationship between output waveforms of sensors and marks in the card issuing device according to the second embodiment of the present invention.

DETAILED DESCRIPTION

(First embodiment)

In casinos such as chess and card rooms, gaming tokens are piled high on gaming tables. However, IC tag readers installed beneath the gaming tables cannot accurately read the total amount. Increasing the reader's sensitivity can cause gaming tokens placed at different locations (betting for wins and losses) to be totaled, making it impossible to determine the total amount of gaming tokens at each location. Furthermore, when shooting from a camera, blind spots due to the camera's viewing angle or overlapping images can obscure the total amount of gaming tokens.

Moreover, although this is often done in certain card games, there is a problem: due to the player squeezing the cards (bending the cards toward the back while enjoying themselves while looking at the size of the cards little by little), the cards are bent, and the image analysis from the camera cannot determine the size and color of the cards.

Furthermore, new challenges are emerging: the increasing sophistication of gaming table cheating. Cheating based on sophisticated poker game methods, which cannot be detected simply by detecting large wins at the gaming table, cannot be detected using cameras or tracking winnings. Furthermore, existing technologies cannot adequately prevent cheating based on collusion between dealers and players.

In order to solve the above-mentioned problems, a cheating detection system for a casino having a plurality of gaming tables according to a first embodiment is provided.

It is a cheating detection system for casinos with multiple gaming tables, with:

a game recording device for recording the progress of the game played at the gaming table, including the dealer and the players, as an image via a camera;

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

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

The control device detects cheating behavior at the gaming table using the image analysis result of the image analysis device and the win/loss result determined by the card issuing device.

In addition, in the cheating detection system, the card issuing device is constructed as follows: it is capable of reading the card size of the issued card, and the control device is constructed as follows: it is capable of comparing the card size information obtained by the image analysis device from the image of each card issued on the gaming table with the card size information of the card read by the card issuing device to determine whether they are consistent.

In addition, in the cheating detection system, the image analysis device or the control device is an artificial intelligence-utilizing type or deep learning structure capable of obtaining information on the face size of cards dealt at a gaming table and bent or soiled by players.

In addition, in the cheating detection system, the control device uses the image analysis device to grasp the position, type and number of gaming tokens placed by each player, and by analyzing the image of the game progress status with the help of the image analysis device, it is determined whether the lost gaming tokens placed by each player are properly recovered and the won gaming tokens are paid out according to the winning or losing results of the game.

In addition, in the cheating detection system, the image analysis device or the control device is an artificial intelligence effective utilization type or deep learning structure that can obtain information on the type, number and location of the placed gaming tokens even if a plurality of gaming tokens placed on the gaming table are partially or entirely hidden due to the blind spot of the camera.

In addition, in the cheating detection system, the control device is an artificial intelligence-based or deep learning structure that can compare and calculate the amount of gaming tokens held in the dealer's gaming token tray at the gaming table according to the winning or losing results of the game, and after the game is over and settled, it can determine whether the amount of lost gaming tokens placed by each player is recovered and the amount of payment for winning gaming tokens is increased or decreased accordingly.

In addition, in the cheating detection system, the control device is an artificial intelligence effective utilization type or deep learning structure that can grasp the position and amount of gaming tokens placed at each playing position on the gaming table, compare the winning and losing history of each player and the amount of gaming tokens obtained from the winning and losing results of each game with the statistical data of past games, and extract abnormal conditions.

In addition, in the cheating detection system, the control device is an artificial intelligence or deep learning structure that can compare the situation in which the amount of game tokens placed at a playing position on a certain game table during a loss is less than the amount of game tokens placed during a win with statistical data from past games to extract abnormal conditions.

In addition, in the cheating detection system, the control device is configured to be able to identify an individual player at a playing position who has been extracted as the abnormal situation by the image analysis device or has obtained a winning amount exceeding a predetermined amount.

In addition, in the cheating detection system, the control device has a warning function for notifying the other gaming table of the presence of the specific player when the identified player leaves and sits at another gaming table.

To address the aforementioned issues, the present invention provides a cheating detection system for a casino with multiple gaming tables, comprising:

a game recording device for recording the progress of the game played at the gaming table, including the dealer and the players, as an image via a camera;

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

An image analyzing device for performing image analysis on the recorded image of the progress of the game; and

a control device capable of detecting banknotes and gaming tokens at the gaming table using the image analysis results of the image analysis device;

The image analysis device or control device is an artificial intelligence effective utilization type or deep learning structure as follows, that is, based on the information obtained from the card issuing device or the dealer, it can detect that the exchange of banknotes and gaming currency is taking place on the gaming table under conditions other than the dealing of cards, and recognize the total amount of genuine banknotes verified by near-ultraviolet light. Furthermore, even if multiple gaming currency substitutes taken out at the gaming table as exchange objects are partially or entirely hidden due to the blind spot of the camera, the total amount of gaming currency substitutes can be recognized, and the total amount of banknotes taken out from the players on the gaming table and the total amount of gaming currency substitutes taken out from the dealer can be compared to determine whether the two amounts are consistent.

In addition, in the cheating detection system, the control device is an artificial intelligence effective utilization type or deep learning structure that can compare and calculate the amount of gaming currency tokens held in the gaming currency token tray of the dealer of the gaming table, and after the exchange and settlement of banknotes and gaming currency tokens, whether the payment amount of the gaming currency tokens corresponding to the exchanged banknotes has increased or decreased.

In the cheating detection system, the control device utilizes artificial intelligence or a deep learning architecture, capable of comparing and calculating, after the exchange and settlement of banknotes and gaming tokens, the banknote deposit amount generated by the dealer's input based on the dealer's input and the total amount of banknotes determined based on the image analysis results of the image analysis device. Furthermore, the control device utilizes artificial intelligence or a deep learning architecture, capable of comparing and calculating the total amount of banknote deposits generated based on the dealer's input at the gaming table for which the dealer is responsible and the total amount of banknotes determined based on the image analysis results of the image analysis device.

The cheating detection system of this embodiment can determine the card size and suit through image analysis, even if the cards are bent due to repeated squeezing by players in a card game. It can also determine the total amount and location of blind spots and overlapping game tokens. Furthermore, it can detect cheating when exchanging banknotes for game tokens.

The following is a detailed description of the overall overview of a cheating detection system for a casino with multiple gaming tables, according to a first embodiment of the present invention. Figure 1 shows the overall overview of the system. The cheating detection system for a casino with multiple gaming tables 4 comprises a game recording device 11 that records the progress of games played at the gaming tables 4, including players 6 and dealers 5, as video using multiple cameras 2; an image analysis device 12 that analyzes the recorded video of the progress of the games; and a card issuing device 3 that determines and displays the win/loss results of each game played at the gaming tables 4. The card issuing device 3 is a so-called electronic card, commonly used by those skilled in the art. It is configured with pre-programmed game rules, capable of reading information from dealt cards C, and determining the outcome of the game. For example, in a certain card game, the outcome of a banker win, player win, or tie (draw) is essentially determined by the rank of two or three cards, and the result (win/loss) is displayed by an outcome indicator 13.

This cheating detection system also includes a control device 14 that compares the actual card rank based on the image analysis results of the image analysis device 12 with the win/loss results determined by the card distribution device 3 to detect cheating behavior at the gaming table 4 (e.g., discrepancies between the total rank of the dealt cards and the win/loss results). The card distribution device 3 is configured to read the rank (A, 2-10, J, Q, K) and suit (hearts, spades, etc.) of the cards C manually dealt by the dealer 5. The control device 14 is configured to compare the rank and suit information obtained by the image analysis device 12 (using artificial intelligence) from the images of each card dealt at the gaming table 4 (captured by the camera 2) with the card and suit information read by the card distribution device 3 to determine whether the cards match or disagree. The image analysis device 12 and control device 14 in this cheating detection system are configured to include a computer, a program, and a memory, either integrated or composed of multiple components.

The image analysis device 12 and control device 14 are constructed using artificial intelligence (AI) or deep learning technology, enabling them to determine the card's rank even if a card C dealt on the gaming table 4 and bent or soiled by a player 6. As shown in FIG4 , soiled card C makes it difficult to distinguish between clubs and spades. Even in such situations, the AI-enabled computer or control system utilizes deep learning (structuring) technology to analyze and determine the image, enabling suit identification. Furthermore, even if a card is bent due to repeated squeezing by players in a card game, the AI-enabled computer or control system and deep learning (structuring) technology can identify the card's suit and rank before deformation by utilizing self-learning of multiple image deformations. AI-enabled computers or control systems and deep learning (structuring) technology are well known and readily available to those skilled in the art, so detailed descriptions are omitted.

The control device 14, which utilizes artificial intelligence or a deep learning architecture, can use the camera 2 and image analysis device 12 to determine where each player 6 places their tokens 9 in the placement area 8 (player, banker, or pair), the type of token 9 placed (different values are assigned to each color of token 9), and the number of tokens. If the tokens 9 overlap vertically, they will be offset and overlapped, as shown in FIG2A . In this case, if the camera 2 is positioned in the direction of arrow X in FIG2A (or if the tokens 9 are in a blind spot relative to the direction of the tokens), as shown in FIG2B , it is conceivable that the tokens 9 will be invisible (entering a blind spot). Computers or control systems that effectively utilize artificial intelligence, or deep learning (architecture) technology, can use self-learning functions to identify hidden gaming tokens 9 due to blind spots (e.g., partially hidden gaming tokens or entirely hidden gaming tokens), and accurately determine the number of gaming tokens. This allows for accurate identification of the placement of gaming tokens 9 in the placement area 8 (player, banker, or pair), the type of gaming token 9 placed (game tokens 9 are assigned different values for each color), and the number of gaming tokens. Therefore, in each game, based on the win/loss outcome determined by the card distribution device 3, the control device 14 analyzes the game progress image using the image analysis device 12 to determine whether the lost gaming tokens placed by each player 6 (indicated by arrow L) and the winning gaming tokens (9W) are appropriately paid to the winning player 6W.

The control device 14 can use the image analysis device 12 to analyze and grasp the total amount of token currency 9 in the dealer's token currency tray 17 at the gaming table 4. After the game ends and the settlement is completed, the control device 14 can compare and calculate whether the total amount of token currency 9 in the token currency tray 17 corresponds to the amount of lost token currency 9 placed by each player 6 and the amount of 9W paid to the winning player 6W for the winning token currency, based on the game's winning or losing results. Even if the total amount of token currency 9 in the token currency tray 17 is constantly grasped using RFID or other means, the control device 14 can determine whether the increase or decrease is correct by analyzing the video of the game progress using the image analysis device 12. This also effectively utilizes artificial intelligence or deep learning architectures.

In this example, cheating or errors are detected based on the information of the winning or losing results of the game, how many pieces of which type of gaming tokens 9 are placed at which position in the placement area 8 (player, banker, or pair), and the increase or decrease in the amount of gaming tokens 9 in the gaming token tray 17 after the recovery of lost gaming tokens and the repayment of won gaming tokens 9. Therefore, even if the changes in gaming tokens 9 after the end of the game are not known, that is, whether the placed gaming tokens 9 are moved to the player side or to the dealer side, cheating or errors can be detected.

For example, in the case of a card game, the game's win or loss can be determined according to the rules of the card game by reading the face value of the cards C released successively by the card issuing device 3. Furthermore, the game's win or loss can also be determined by using the camera 2 to capture the gaming table 4, analyzing the image using the image analysis device 12, and comparing the analysis results with the game rules by the control device 14. In this case, the camera 2, the image analysis device 12, and the control device 14 constitute the win or loss determination device. Information about which type of gaming token 9 the player at each playing position 7 placed, how many pieces, and where in the placement area 8 (player, banker, or pair) can be obtained by capturing the gaming token 9 placed in the placement area 8 using the camera 2 and analyzing the image for each playing position 7 using the image analysis device 12.

Moreover, the increase or decrease in the amount of gaming currency 9 in the gaming currency tray 17 before and after the recovery of lost gaming currency 9 and the repayment of won gaming currency 9 can be calculated by comparing the total amount of gaming currency 9 in the gaming currency tray 17 before the recovery of lost gaming currency 9 and the repayment of won gaming currency 9 with the total amount of gaming currency 9 in the gaming currency tray 17 after the recovery of lost gaming currency 9 and the repayment of won gaming currency 9. The total amount of gaming tokens 9 in the gaming token tray 17 before the recovery of lost gaming tokens 9 and the redemption of won gaming tokens 9, and the total amount of gaming tokens 9 in the gaming token tray 17 after the recovery of lost gaming tokens 9 and the redemption of won gaming tokens 9, can be detected by imaging the gaming token trays 17 containing the gaming tokens 9 with the camera 2 and analyzing the images with the image analysis device 12. Alternatively, the total amount of gaming tokens 9 stored in the gaming token tray 17 can be detected by embedding an RFID indicating the amount of gaming tokens 9 in the gaming token tray 17 and providing an RFID reader on the gaming token tray 17.

For example, assume that before the game begins, the total amount of tokens 9 in the token tray 17 is Bb. After the game ends and the collection of lost tokens and the redemption of won tokens are completed, the total amount of tokens 9 in the token tray 17 is Ba. Furthermore, in this game, the total amount of tokens 9 placed in all play positions 7 in the player area is bp, the total amount of tokens 9 placed in all play positions 7 in the banker area is bb, and the total amount of tokens 9 placed in all play positions 7 in the tie area is bt. For example, if the banker wins the game, Ba - Bb = bp - bb + bt should hold. Alternatively, the total amount of tokens 9 in the token tray 17 after the game ends, Ba, should be (Bb + bp - bb + bt). If this is not the case, it can be determined that cheating or an error occurred in the collection or redemption of tokens.

Figure 3A shows a detailed diagram of the token currency tray of this embodiment, and Figure 3B shows another example of the token currency tray. The token currency tray 17 includes a token currency collection tray 171 and a token currency repayment tray 172. The token currency collection tray 171 collects and temporarily stores token currency 9L placed by the losing player 6L, while the token currency repayment tray 172 stores token currency 9W to be repaid. The image analysis device 12 and control device 14 determine the location, type, and number of token currency 9L placed by the losing player 6L and calculate the increase in the token currency 0L used in the game (the expected amount of token currency 9L in the token currency collection tray 171). Furthermore, 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 corresponding total amount with the actual total amount to determine whether there is a discrepancy.

In addition, by using the game currency 9W in the game currency tray 172 to repay the winning player 6W, it is possible to ensure that there is sufficient time for the image analysis device 12 and the control device 14 to grasp the actual total amount of game currency 9 in the recovered game currency tray 171 after the recovery.

The gaming table 4 is provided with a discard area 41 and/or a discard slot 42 for discarding 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 slot 42 on the gaming table 4 for discarding.

The gaming table 4 also includes a sign 43 indicating the outcome of the game. Figure 5A is a top view of the front of the sign, and Figure 5B is a top view of the back of the sign. In certain card games, two types of signs are used: a sign 43a indicating a player win and a sign 43b indicating a banker win. When the game results are announced, the dealer 5 flips over the sign indicating the winning player or banker. This makes it easy to clearly display the outcome of the game on the table. After the game tokens 9 have been collected and redeemed, the flipped sign is restored by the dealer 5. Restoring the sign also means that the next game is ready to begin.

As described above, in this embodiment, the control device 14 calculates the balance of gaming tokens for each game based on the amount of gaming tokens placed on the gaming table 4 and the game's winning or losing results. It then verifies the increase in the gaming token balance in the gaming token tray 17 after the game. If the control device 14 detects a discrepancy during this verification, it issues an alarm or adds a record indicating this to the video footage captured by the camera 2. The card room operator can investigate the cause of the discrepancy by reviewing the video footage.

In the cheating detection system of this embodiment, the total amount of gaming currency 9 in the gaming currency tray 17 before each game is settled is added or subtracted from the increase or decrease in gaming currency in the game calculated based on the position, type and number of gaming currency 9 placed by all players 6 in the game and the win or loss result of the game obtained by the win or loss result determination device. The expected total amount of gaming currency 9 in the gaming currency tray 17 after settlement at the end of the game is compared with the actual total amount of gaming currency 9 in the gaming currency tray 17 at the end of the game obtained by the image analysis device 12 to determine whether there is a difference between the expected total amount and the actual total amount.

The control device 14 uses the image analysis device 12 to grasp the position, type and number of gaming tokens placed by each player. When the recovery of all lost gaming tokens placed by each player is completed, the control device 14 grasps the actual total amount of gaming tokens in the gaming token tray. The total amount of gaming tokens in the gaming token tray before the settlement of each game is compared with the expected total amount of gaming tokens 9 in the gaming token tray 17 obtained by adding the position, type and number of gaming tokens placed by the losing players to the increase in the gaming token tray 17 in the game, and the actual total amount of gaming tokens 9 in the gaming token tray 17 to determine whether there is a difference between the expected total amount and the actual total amount.

The control device 14 compares the total amount of gaming tokens 9 in the gaming token tray 17 before settlement of each game with the expected total amount of gaming tokens 9 in the gaming token tray 17, which is obtained by adding the position, type and number of gaming tokens 9 placed by the losing player to the increase in the gaming token tray 17 in the game, and determines that there is no difference between the expected total amount and the actual total amount. The control device 14 compares the expected total amount in the gaming token tray 17 after actuation at the end of the game with the actual total amount of gaming tokens 9 in the gaming token tray 17 at the end of the game obtained by the image analysis device 12. If it is determined that there is a difference between the expected total amount and the actual total amount, it is determined that a payment error has occurred and a payment error signal is generated to notify the payment error.

In the game currency tray 17, there is provided a recovery game currency tray 171 for recovering and temporarily storing the game currency 9 placed by the losing player. The image analysis device 12 compares the expected total amount of game currency 9 in the recovery game currency tray 171 obtained by adding the increase in the game currency 9 in the game calculated based on the position, type and number of the game currency 9L placed by the losing player, with the actual total amount of game currency 9 in the recovery game currency tray 171 to determine whether there is a difference between the expected total amount and the actual total amount.

In order to analyze the records of the game that caused the above-mentioned difference in the game recording device 11 when the control device 14 determines that the actual total amount of the game token 9 held in the game token tray 17 of the dealer 5 of the game table 4 does not correspond to the increase or decrease in the game token amount calculated based on the amount of the game token placed by all players and the winning or losing results of the game, the game recording device 11 is constructed to assign an index or time to the obtained image, or to be able to determine and reproduce the recovery scene or payment scene of the game token 9.

In this manner, the control device 14 uses the image analysis device 12 to determine the total amount of gaming tokens in the gaming token tray 17 after settlement at the end of the game. However, the determination of settlement in this case is made when any of the following situations 1) to 4) occurs:

1) When the redemption of the gaming tokens 9 won is completed;

2) When the card C used in the game is recovered and discarded in the discard area 41 or discard slot 42 of the table;

3) When a prescribed button attached to the winning/losing result determination device is pressed;

4) When the symbol 43 indicating winning or losing is restored.

Furthermore, the control device 14 utilizes artificial intelligence or a deep learning structure, capable of understanding the position (for the player, banker, or pair) and amount (type and number) of tokens placed at each playing position 7 of the gaming table 4. The control device 14 compares the win/loss history of each player 6 and the amount of tokens (won) obtained from the win/loss results of each game with statistical data from many past games (big data), thereby identifying abnormal conditions (set by the gaming room). Typically, the control device 14 utilizes artificial intelligence or a deep learning structure, capable of identifying abnormal conditions (set by the gaming room) by comparing the win/loss history of each player 6 and the amount of tokens (won) obtained from the win/loss results of each game with statistical data from many past games (big data, etc.). Typically, the control device 14 utilizes artificial intelligence or a deep learning structure, capable of identifying abnormal conditions (set by the gaming room) by comparing the win/loss history of each player 6 and the amount of tokens (won) obtained from the win/loss results of each game with statistical data from many past games (big data, etc.).

Furthermore, the control device 14 (integrated with the image analysis device 12) of this cheating detection system is configured to identify individual players 6 at playing positions 7 who have been identified as abnormal or who have achieved a winning amount exceeding a specified amount. To identify these players 6, the image analysis device 12 extracts facial images through feature point extraction and assigns an identification number (ID, etc.) to the identified players 6. Furthermore, the control device 14 includes a warning function that notifies the administrator or person in charge (or dealer) of each gaming table 4 of the presence of the identified player 6 when the identified player 6 leaves and relocates to another gaming table. Specifically, the control device 14 notifies the administrator or person in charge (or dealer) of each gaming table 4, further preventing abnormalities.

The control device 14 also has a database that records the exchange history of banknotes K and gaming currency 9. The control device 14 refers to the database at regular intervals or on a daily basis to compare and determine whether the amount of gaming currency 9 held in the gaming currency tray 17 of the dealer 5 of the gaming table 4 has increased or decreased based on the payment amount of the gaming currency 9 corresponding to the exchanged banknotes K, or the total amount of the payment amount of the banknotes K corresponding to the exchanged gaming currency 9.

Furthermore, in the above example, the win/loss history and the amount of gaming tokens (wins) earned at each playing position 7 can be monitored without identifying individual players 6. In this case, while it is impossible to track and investigate each player 6 if they leave their seat, it is possible to detect abnormal situations such as a situation where a specific playing position 7 at a gaming table 4 consistently deposits a small amount of gaming tokens during losses and a large amount during wins, across multiple games. Furthermore, if such a playing position 7 is detected, cheating or errors at that playing position 7 may be suspected. Furthermore, by verifying a video recording of that playing position 7, cheating or errors can be discovered.

Specifically, the camera 2 is configured to capture at least the gaming tokens 9 placed within the placement area 8 of the gaming table 4. The image analysis device 12 analyzes the images captured by the camera 2 and, for each user position 7, detects at which position (Player, Banker, or Tie) within the placement area 8 the gaming tokens are placed, as well as the amount of gaming tokens placed. Furthermore, the card distribution device 3 functions as a win/loss determination device, determining the outcome of the game. The control device 14 records (monitors) the win/loss history and the amount of gaming tokens earned (gaming token amount) for each playing position 7 based on the position (Player, Banker, or Tie) within the placement area 8 where the gaming tokens 9 are placed and the outcome of the game. Alternatively, only one of the win/loss history and the amount of gaming tokens earned may be recorded. When the control device 14 compares the win and loss history and/or the gaming currency acquisition history with the statistical data of many past (big data) games and finds an abnormal situation (set by the chess and card room), the player position 7 is determined to be a playing position suspected of cheating.

If cheating is suspected at a player position 7, the cheating detection system generates an alarm (light, sound, or vibration) in a manner that is at least perceptible to the dealer at that point in time. This prevents the cheating from continuing by, for example, interrupting the game at that location. Furthermore, information indicating the suspected cheating can be added to the video footage captured and recorded by camera 2. This allows the cause of the suspected cheating to be investigated by verifying the video.

The cheating detection system for an amusement arcade with a gaming table according to this embodiment also has a function for checking the frequent exchange of banknotes and gaming tokens at the gaming table 4. In a gaming arcade, such as a card room, before a game, players 6 exchange banknotes (cash, etc.) for gaming tokens at a designated gaming token exchange location. However, if a player 6 runs out of gaming tokens, they can continue playing at the gaming table (such as a card game table) by exchanging cash (banknotes) for gaming tokens 9 without leaving the gaming table 4. However, this creates an opportunity for cheating between the dealer 5 and the players. Exchanging cash (banknotes) for gaming tokens 9 at the gaming table (such as a card game table) must be done when the game is not in progress. The card issuing device 3 can detect the start and end of card dealing (the winning and losing periods) to determine the outcome of the game. Therefore, the card issuing device 3 detects situations other than card issuing (dealing). The control device 14 detects, in situations other than card dealing, that the exchange of banknotes and gaming tokens 9 is taking place on the gaming table 4 (as shown in FIG6 ). The situation of card dealing (or other situations) can be detected based on information obtained from the card issuing device 3 or the actions of the dealer 5.

The control device 14 performs image analysis of the surface of the banknotes K to identify the number and amount of the banknotes. Furthermore, on the gaming table 4, the authenticity of the banknotes K exchanged for gaming tokens 9 is verified by irradiating the banknotes with near-ultraviolet light to detect the authenticity mark G. As shown in Figure 6 , the control device 14 utilizes artificial intelligence or a deep learning architecture that also performs image analysis to verify the authenticity mark G, thereby identifying the total amount of genuine banknotes. This allows the control device 14 to identify the total amount of gaming tokens even if multiple gaming tokens are withdrawn from the gaming table as exchange objects and are hidden due to a blind spot in the camera 2. Furthermore, the control device 14 can compare the total amount of banknotes K withdrawn from the gaming table 4 by a player with the total amount of gaming tokens 9 withdrawn from the dealer 5 to determine whether the two amounts match.

The control device 14 utilizes artificial intelligence or a deep learning architecture that can compare and calculate the total amount of gaming tokens 9 in the dealer's 5 gaming table 4's gaming token tray 17. After banknotes and gaming tokens are exchanged and settled, the control device 14 can determine whether the amount of gaming tokens 9 has increased or decreased based on the amount paid out in accordance with the exchanged banknotes. Alternatively, the total amount of gaming tokens 9 in the dealer's 5 gaming token tray 17 can be monitored in advance using, for example, RFID tags attached to the gaming tokens 9. Alternatively, the camera 2 can capture the gaming token tray 17 containing the gaming tokens 9, and the image analysis device 12 can analyze the resulting image to detect the total amount of gaming tokens 9 stored in the gaming token tray 17.

Furthermore, the control device 14 verifies whether the increase or decrease in the amount of gaming currency 9 in the gaming currency tray 17 before and after the exchange of banknotes and gaming currency tokens is consistent with the gaming currency token exchange amount as determined by image analysis of the gaming table 4. The amount of the paid banknotes can be input into the control device 14 by the dealer 5 via keyboard input, or can be determined by analyzing the image of the gaming table 4 where the banknotes are being paid using the camera 2 and the image analysis device 12.

As described above, the control device 14 determines whether the deduction of the gaming tokens 9 from the gaming token tray 17 resulting from the exchange of banknotes and gaming tokens is consistent with the amount of banknotes paid by the player 6 to the dealer 5. Furthermore, the control device 14 is an intelligent control device capable of comparing and calculating the consistency between the amount of banknotes deposited by the dealer 5 (typically based on keyboard input, etc.) after the exchange and settlement of banknotes and gaming tokens, and the calculated amount of banknotes based on the image analysis results of the image analysis device 12. Furthermore, the control device 14 may be a device that effectively utilizes artificial intelligence or a deep learning architecture.

In addition, the control device 14 is an artificial intelligence effective utilization type or deep learning structure that can compare and calculate the consistency or inconsistency between the total amount of banknotes deposited in the gaming table 4 for which the dealer is responsible based on the input of the dealer and the total amount of banknotes based on the image analysis results of the image analysis device 12.

The control device 14 compares and determines whether the amount of gaming currency 9 held in the gaming currency tray 17 of the dealer 5 of the gaming table 4 has been increased or decreased after the exchange of banknotes and gaming currency 9, or the payment amount of the gaming currency 9 corresponding to the exchanged banknotes, or the payment amount of the banknotes corresponding to the exchanged gaming currency 9.

(Second embodiment)

Among the many table games played in casinos and other amusement parks, some are card games. These games use a standard deck of 52 playing cards. Cards are dealt onto the gaming table from a card distribution device equipped with pre-shuffled decks (ranging from 6 to 9 or 10 cards). Winning is determined by the number of cards dealt (the rank of the cards) and the rules of the game.

The distribution of cards from the card distribution device and the calculation of the placement of the guests (game participants) are performed by the dealer in charge of the game table, etc. In amusement parks such as card rooms, attempts are being made to prevent errors and cheating in the calculation of the placement of the guests (game participants).

In the card game monitoring system described in International Publication WO2015/107902, a monitoring camera is used to read the changes in gaming tokens and check whether the bet has been paid to the winner.

Regarding the calculation of the guest's placement in a card game and the dealer's placement on the guest (game participant), there is the following problem: it is impossible to detect the time when these are performed, who placed the game token, or whether it was taken, so it is impossible to determine whether these are correct.

To solve the above-mentioned problems, a cheating detection system according to a second embodiment is a cheating detection system for a casino having gaming tables, and includes:

A game monitoring device that uses a camera to monitor the progress of the game being played at the gaming table;

an image analysis device for performing image analysis on the image obtained by the camera;

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

a control device for determining the position of gaming tokens placed on the gaming table by the game participants using the analysis results of the image analysis device in each game, and for determining the winner and loser among the participants in each game using the win/loss results;

The control device also has the function of determining at least one of the following:

1) In each game, whether there is no change in the game token currency from the time the start card is drawn or the dealer's game start operation until the card distribution device displays the game win or loss result,

2) After each game is over, during the period when the dealer collects the gaming tokens placed by the losing players, is there any change in the gaming tokens caused by anyone other than the dealer?

3) After each game is over, during the period when the dealer collects the game tokens placed by the losers among the game participants, whether additional game tokens are added,

4) After each game is over, whether the dealer pays the gaming tokens placed by the winner among the game participants,

5) After each game is over, whether the winner among the game participants has taken the placed game tokens and the paid game tokens.

Furthermore, the control device can also be configured to detect changes in the hands of the dealer and game participants, changes in gaming currency, or changes in the hands and changes in gaming currency by using the analysis results of the image analysis device, thereby making a judgment on at least one of 1) to 5).

Furthermore, the control device may be configured to determine whether the amount of gaming tokens paid by the dealer to the winner is correct based on the amount bet by the winner among the game participants.

Furthermore, the game cheating detection system may further include a monitor or a light that receives the determination result and issues a warning or displays.

According to the cheating detection system of this embodiment, the placements made by guests in a certain card game and the placements made by the dealer against guests (game participants) can be accurately calculated. The time when these are made, who placed the game tokens, or whether they were taken can be detected. These errors and cheating behaviors can be detected and warnings or displays can be issued, thereby preventing them from happening again.

Before entering into the detailed description of this embodiment, the flow of a card game played in a casino such as a card room will be described. Furthermore, in the second embodiment, the same components as those in the first embodiment will be described with the same reference numerals.

As shown in FIG7 , at a gaming table 4, a guest (game participant/player) 6 and a dealer 5 sit facing each other in a playing position (chair) 7. The guest (game participant) 6 then places gaming tokens 9 on a betting area 8 in front of them and places a bet (hereinafter referred to as a "bet") on the outcome of the card game, determining whether the player or banker will win, or whether the game ends in a tie. Furthermore, the dealer 5 times the bet by the guest (game participant) 6, shouts "No More Bet," and performs other actions, such as moving his hand sideways (see the state shown in FIG7 ). In this card game, from the moment "No More Bet" is called out (placement acceptance ends) and the card drawing begins, or the dealer 5 performs the game start operation, until the game win or loss result is displayed on the card issuing device 3, the guest (game participant) 6 cannot move the game tokens, place additional game tokens, or withdraw placed game tokens.

Next, playing cards 1 are drawn one by one from the card distribution device 3 onto the gaming table 4, facing up. First, four cards are drawn, as shown in Figures 1 through 4 of FIG7 : the first card goes to the player, the second to the banker, the third to the player, and the fourth to the banker. These cards are arranged in areas 10 (player area 10P and banker area 10B) visible from the dealer 5 on the gaming table 4. Furthermore, based on the rank (number) of the first through fourth cards 1 and the conditions specified in the detailed rules of the card game, the dealer 5 draws the fifth and sixth cards 1, each of which goes to the player or banker. Next, the game is determined based on the rank (number) of the first through fourth cards 1 (and the fifth and sixth cards, if applicable) and the detailed rules of the card game. Here, the card dealing device 3 is configured so that the game rules are pre-programmed and the information (card rank (number) and suit) of the dealt cards 1 can be read to determine the outcome of the game. A determination is then made as to whether the outcome (win/lose result) determined by the card dealing device 3 is consistent with the outcome determined by the dealer, etc., as described above.

The following describes an overall overview of a cheating detection system for casino games according to an embodiment of the present invention. Figure 7 shows an overall overview of the system. The cheating detection system for casino games includes: a game recording device 11 that records the progress of a game played at a gaming table 4, including the progress of guests (game participants) 6 and the dealer 5, using a camera 2; an image analysis device 12 that analyzes the recorded video of the progress of the game; and a card dealing device 3 that determines and displays the winning or losing results of each game played at the gaming table 4. The card dealing device 3 is a so-called electronic boot, already used by those skilled in the art. It is configured with pre-programmed game rules and can detect the moment when the dealer 5 deals a card 1 at the start of each game. It also reads information (such as the rank (number) and suit) of each dealt card 1 to determine the winning or losing result of the game. For example, in a certain card game, for the banker win, the player win, and the draw (tie), it is basically determined by the size of the cards of 2-3 cards, and the judgment result (win or lose result) is displayed by the display light 13.

The control device 14 of this cheating detection system includes a token detection function. This function uses the analysis results of the image analysis device 12 to determine, during each game, which slot 8 on the player or banker side of the gaming table 4 the guest 6 (game participant) has placed tokens 9. The position and total amount of tokens 9 (in which slot 8 on the player or banker side the tokens 9 have been placed) are often difficult to read if tokens 9 are staggered or overlapped, or if they are in a blind spot from the camera 2. The control device 14 is configured to detect hidden tokens 9 due to blind spots (e.g., partial or complete concealment of a token) using a computer or control system utilizing existing artificial intelligence, or a self-learning function based on deep learning (structuring) technology, and accurately determine their position and number. Furthermore, the structure for detecting the position and type of the gaming token 9 in the placement area 8 is not limited thereto, and for example, detection may be performed by reading an ID embedded in the gaming token.

As described above, the control device 14, using the camera 2 and image analysis device 12, can detect the position (position of the game tokens 9 placed on the player, banker, or pair), type (game tokens 9 are assigned different amounts for each color), and number of game tokens 9 placed by each player 6. This allows the control device 14 to detect which player 6 placed the game token 9 on the player (if multiple players 6 placed the game token 9 on the player, which player 6 placed the highest amount) and which player 6 placed the game token 9 on the banker (if multiple players 6 placed the game token 9 on the banker, which player 6 placed the highest amount). The image analysis device 12 and control device 14 in this cheating detection system are constructed by combining a computer, a program, and a memory, either integrally or in combination.

The control device 14 is configured to compare the card size and suit information obtained by the image analysis device 12 from the images of each card 1 dealt at the gaming table 4 (using the camera 2) with the card size and suit information read by the card distribution device 3 to determine whether they match or not. Based on the win/loss results determined by the card distribution device 3 in each game, the control device 14 analyzes the images of the game progress using the image analysis device 12 to determine whether lost gaming tokens 9 placed by the guests (game participants) 6 and won gaming tokens have been appropriately collected and paid to winning guests (game participants) 6 in accordance with the game win/loss results.

The control device 14 has the following functions 1) to 5) as a characteristic function of the present invention, and determines whether cheating that violates the rules is not performed according to the rules of a certain card game.

1) In each game, the image analysis device 12 uses the information obtained by the camera 2 to monitor (as shown in Figure 8) whether there is no change in the game token currency 9 from the time the card drawing start signal is obtained from the card distribution device 3, or from the time the dealer 5 presses the start button 4s to start the game, until the card distribution device 3 displays the winning or losing result of the game.

2) After each game is over, the image analysis device 12 uses the information obtained by the camera 2 to monitor whether the loser 6 has taken the game token 9 without cheating during the period when the dealer 5 recovers the game token 9 placed by the loser among the game participants 6 (as shown in Figure 9).

3) After each game is over, the image analysis device 12 uses the information obtained by the camera 2 to monitor whether people other than the dealer 5 (winners or losers) have added the winning gaming tokens 9W during the period when the dealer 5 recovers the gaming tokens 9 placed by the losers among the game participants, or whether they have not replaced the gaming tokens 9 on the winning side.

4) After each game is over, the image analysis device 12 uses the information obtained by the camera 2 to monitor whether the dealer 5 has correctly placed the paid entertainment token 9W at the position where the winner among the game participants 6 placed the entertainment token 9 (as shown in Figure 10).

5) After each game is over (the dealer 5 operates the card issuing device 3 and displays the winning or losing results on the display light 13), the image analysis device 12 uses the information obtained by the camera 2 to monitor whether the winner 6W among the game participants 6 has taken the placed entertainment token 9 and the paid entertainment token 9W (as shown in Figure 11).

The control device 14 analyzes the information obtained by the image analysis device 12 using the camera 2 as follows. Specifically, the control device 14 uses the analysis results of the image analysis device 12 to detect hand movements of the dealer 5 and game participants 6, changes in gaming tokens, or both, to perform monitoring 1) through 5) above. However, this fundamental analysis requires at least knowing who has taken the gaming tokens 9. This analysis method will be described below using Figures 12 through 14.

The dealer 5 analyzes the situation of the gaming tokens 9 placed by the game participants 6L (FIG. 12).

The dealer 5 collects the token money 9 placed by the losing player 6L. The dealer monitors whether the token money 9 is actually collected by analyzing information obtained by the image analysis device 12 using the camera 2. First, image analysis detects the change from the placed token money 9 being present (Figure 12A) to the absence (Figure 12C). Next, the image from the transition from the token money 9 being present to the absence (Figure 12B) is analyzed. The image from the transition from the token money 9 being present to the absence (Figure 12B) is analyzed to determine from which direction the hand 5h is extended (from above or elsewhere in Figure 12). If the hand is extended from above (or appears from above and then withdraws upward), the hand 5h is determined to be the dealer's 5 hand. If the hand is extended from any other direction, it is determined to be cheating. This rule is used to detect cheating.

While the dealer 5 is collecting the gaming token 9 placed by the losing game participant 6L, monitoring is performed to determine whether other players have taken the lost gaming token 9 without cheating (Figures 12 and 11). Analysis of the situation in which the losing player 6L or the like has taken the gaming token 9, as shown in Figure 12, in the image during the transition from the gaming token 9 being present to the non-present state, detects that the hand 6h extends or moves from the bottom of Figure 12 (originally from the top). This is determined to indicate that a hand other than the dealer 5's 6h has taken the gaming token 9, and cheating is therefore determined.

Analysis of the situation in which the dealer 5 correctly pays (places) the winning tokens 9W for the winning tokens 9, and the winning tokens 9W are taken by the winner 6W among the game participants 6. First, the winning tokens shown in FIG14A are redeemed for the tokens 9W according to the game rules, as shown in FIG14B. The change from the state shown in FIG14A to the state shown in FIG14B is detected, and image analysis is performed to determine whether the opponent is the dealer 5's hand 5h. Then, based on the image analysis results, the control device 4 checks to determine whether the winning token 6h among the game participants 6 extends (moves) to the same placement area, as shown in FIG14C, and all the tokens 9 disappear (the state shown in FIG14D), in accordance with the game rules, to determine whether cheating has occurred.

Furthermore, the control device 4 is configured to determine whether the amount of gaming tokens paid by the dealer 5 to the winner 6W is correct based on the amount placed by the winner 6W among the game participants 6. A specific example is described below. The position and total amount of gaming tokens 9 (i.e., which placement area 8 on the player or banker side the gaming tokens 9 are placed) are typically not readable if the gaming tokens 9 are staggered or overlapped, or if they are in a blind spot from the camera 2's perspective. The control device 14 is configured to utilize a self-learning function based on existing artificial intelligence-based computers or control systems, deep learning (structuring) technology, and other techniques to identify hidden gaming tokens 9 due to blind spots (e.g., partial or complete concealment of a gaming token), and accurately determine their position and number. Furthermore, the structure for detecting the position 8 and type of the amusement token 9 in the placement area 8 is not limited thereto, and for example, detection may be performed by reading an ID embedded in the amusement token.

As described above, the control device 14 can grasp the position 8 (the position for the player, the banker, or the pair), the type (the gaming currency 9 is assigned different amounts of value for each color) and the number of gaming currency 9 placed by each player 6 through the camera 2 and the image analysis device 12, and can detect which guest 6 placed the gaming currency 9 on the player (if there are multiple guests 6 placing the gaming currency on the player, which guest 6 placed the highest amount), and which guest 6 placed the gaming currency on the banker (if there are multiple guests 6 placing the gaming currency on the banker, which guest 6 placed the highest amount).

Furthermore, the control device 14 of the cheating detection system for this game analyzes the information obtained by the image analysis device using the camera 2 according to the rules of the card game using the aforementioned method. The monitoring described in 1) through 5) above is performed to determine whether cheating that violates the rules has occurred. When cheating is detected, the card distribution detection device 14C illuminates an abnormality indicator 16 provided on either the card distribution device 3 or the gaming table 4, and outputs the detection of cheating via wireless or wired communication 15 to the card room management department. Monitors or lights may also be provided at other locations to receive the determination results and issue warnings or displays.

As described above, cheating behavior is detected by the control device 14, and a display signal is sent to the display light 13 or the abnormality display light 16 of the card issuing device 3 at the time of detection or at an appropriate moment. However, in addition to issuing a warning, after the time of detecting cheating or an error, the card issuing device 3 can also have a function of preventing the issuance of cards to prevent the issuance of cards 1.

An embodiment of the card issuing device 3 used in the table game system of the present invention will be described below with reference to Figures 15 to 19. The card issuing device 3 includes a card storage unit 102 for storing a plurality of shuffled playing cards 1s; a card guide unit 105 for guiding the shuffled playing cards 1 as the dealer 5 or the like manually draws the shuffled playing cards 1 one by one from the card storage unit 102 onto the gaming table 4. 1 for guiding; the opening portion 106 is used to take out the guided card 1 from the card guide portion 105; the card detection portion (card detection sensors 22 and 23) detects that the shuffled playing card 1 is drawn out when the shuffled playing card 1 is drawn out; the card reading portion 108 reads information indicating the minimum number (card size) of the shuffled playing card 1; the control portion 109 determines the win or loss of the card game based on the number (card size) of the shuffled playing card 1 read in sequence by the card reading portion 108; the result display light 13 displays the win or loss result determined by the control portion 109; the issuance restriction device 30 is provided at the opening portion 106, which restricts the entry and exit of the card 1 from the card storage portion 102; and the management control portion 114 has the same function as the control device 14, which is integrated and has the function of being controlled by the control device 14 When a dealer's mistake or cheating behavior is detected during a game, a function is provided to prevent further cards from being drawn from the card issuing device 3 after the time when the mistake or cheating behavior is detected or at a predetermined time.

Next, the issuance restriction device 30 that restricts the entry and exit of cards 1 from the card holding unit 102 will be described using Figures 17 and 16 . The issuance restriction device 30 is provided on the card guide 107 of the card guide unit 105. This card guide unit 105 guides the cards 1, which are removed one by one from the opening 106 in the front of the card holding unit 102, onto the gaming table 4. The issuance restriction device 30 has a structure in which, when a card 1 passes through a slit 33 between the card guide unit 105 and the guide cover of the card guide 107, a locking member 34 presses against the card 1, preventing the entry and exit of the card 1 within the slit 33. The locking member 34 is driven by a drive unit 35, such as an electromagnetic solenoid or a piezoelectric element, so that it can assume two states: a position that presses the card 1 (a restriction position) and a passable position that allows the card 1 to pass through, as indicated by arrow m. The drive unit 35 is controlled by the control unit 109, which is directly or indirectly connected to the control device 14 via wired or wireless communication, to move the locking member 34 between two positions: a position that presses the card 1 and a passable position that allows the card 1 to pass through. The rules of the card game are pre-programmed and stored in the control unit 109.

Next, a modified example of the issuance restriction device 30 will be described using FIG18B . The issuance restriction device 40 in this modified example has the following structure: when a card 1 passes through the slit 33 between the card guide 105 and the card guide 107 (guide cover), a locking member 36 protrudes into the slit 33 to prevent the movement of the card 1. The locking member 36 is moved, as indicated by arrow m, by a drive unit 37 such as an electromagnetic solenoid or a piezoelectric element, to assume two positions: a position that prevents the movement of the card 1 (a restricted position) and a passable position that allows the card 1 to pass. The drive unit 37 is controlled by a control unit 109 connected to the control device 14 to move the locking member 36 between the position that prevents the movement of the card 1 and the passable position that allows the card 1 to pass.

Next, the details of the code reading unit 108, which reads the code 52 indicating the number (number, rank) of a card 1 when it is manually removed from the card storage unit 102, will be described. Figure 17 is a top view of the main components of the card dispensing device 3. In the figure, the code reading unit 108 is located in the card guide 105, which guides the cards 1 manually removed from the opening 106 in front of the card storage unit 102 one by one onto the gaming table 4. The card guide 105 is an inclined surface, with card guides 107, which also serve as sensor covers, attached to both edges. Each of the two card guides 107 can be attached and detached using screws (not shown). Removing the card guides 107 exposes the sensor group 115 of the code reading unit 108. The sensor group 115 consists of four sensors: two ultraviolet-sensitive sensors (UV sensors) 20 and 21, and object detection sensors 22 and 23.

Object detection sensors 22 and 23 are fiber-optic sensors that detect the presence of a card 1 and can detect the movement of a card 1. Object detection sensor 22 is located upstream of the card guide 105 in the direction of card 1 movement, while object detection sensor 23 is located downstream. As shown in the figure, the two object detection sensors 22 and 23 are arranged upstream and downstream, sandwiching UV sensors 20 and 21. UV sensors 20 and 21 include ultraviolet LEDs (UV LEDs) that emit ultraviolet light and detectors. Markings M representing the codes 52 are printed on the cards 1 using ultraviolet luminescent ink that emits a color when exposed to ultraviolet light. Ultraviolet light (near-ultraviolet light) illuminates the cards 1, and the detectors detect the reflected light from the markings M representing the codes 52 on the cards 1. UV sensors 20 and 21 are connected to the code reader 108 and the control unit 109 via cables. The code reader 108 determines the combination of markings M based on the detector output signals from the UV sensors 20 and 21, and determines the number (card size) corresponding to each code 52.

In the code reading unit 108, the control unit 109 controls the start and end of reading by the UV sensors 20 and 21 based on the detection signals from the object detection sensors 22 and 23. Furthermore, the control unit 109 also determines whether the card 1 has passed normally through the card guide 105 based on the detection signals from the object detection sensors 22 and 23. As shown in Figure 19, square marks M representing the card size (number) and suit (such as hearts or spades) are arranged in two columns and four rows along the edge of the card 1. When the UV sensors 20 and 21 detect the marks M, they output an ON signal. The code reading unit 108 determines the relative relationship between the two signals input from the two UV sensors 20 and 21. Thus, the code reading unit 108 determines the code based on the relative difference between the two marks M detected by the two UV sensors 20 and 21, thereby identifying the corresponding number (size) and type (suit) of the card 1.

FIG19 shows the relationship between the code 52 and the output of the ON signals from the two UV sensors 20 and 21. A predetermined combination of markers M can be determined based on the comparison of the relative changes in the output of the ON signals from the UV sensors 20 and 21. Consequently, if there are four combinations of markers M in two upper and lower columns, and these are printed in four columns, then 256 codes can be formed (four to the power of four). Each of the 52 playing card types is assigned to one of the 256 codes, and this is stored as a lookup table in a memory or program. The code reader 108 is configured to determine the number (card rank) and type (suit) of cards 1 based on the predetermined lookup table (not shown) by determining each code 52. Furthermore, the 256 codes can be freely associated with the 52 card types and stored in the lookup table. This allows for complex combinations, and the combination of the 256 codes and the 52 card types can vary depending on time and location. The code is preferably printed using a paint that becomes visible upon exposure to ultraviolet light, and is printed at a position that does not overlap with the card type indication and index 103 .

Furthermore, in the above-described embodiment, the image analysis device 12 and the control device 14 are devices that utilize artificial intelligence or deep learning architectures. However, specifically, the image analysis device 12 and the control device 14 may perform image analysis and the various controls described above using the Scale-Invariant Feature Transform (SIFT) algorithm, a convolutional neural network (CNN), deep learning, machine learning, or similar methods. These technologies perform image recognition on captured images to identify objects contained within them. In recent years, deep learning techniques that utilize multi-layered neural networks have been gaining popularity for achieving high-precision object recognition. This deep learning technique generally achieves high-precision object recognition by overlapping multiple layers between the input and output layers of the neural network. Among these deep learning techniques, convolutional neural networks, in particular, have attracted attention due to their superior performance compared to conventional methods for identifying objects based on image features.

Convolutional neural networks train on labeled target images and identify the primary objects contained in them. When multiple primary objects exist within the training image, they are designated using rectangular regions, and the images corresponding to these regions are labeled for training. Furthermore, convolutional neural networks can also determine the primary objects within an image and their locations.

Further explaining convolutional neural networks, the object recognition process involves extracting candidate regions based on local features, such as edge extraction, from the target image. These regions are then fed into the convolutional neural network, where they are classified based on feature vectors. The most reliable candidate regions are then classified as the recognition result. Confidence is a measure of how much the similarity between a particular image region and the subject of the image learned along with the label is higher than other similarity levels.

In addition, devices with artificial intelligence effective utilization or deep learning structures are described in U.S. Patent No. 9,361,577, U.S. Patent Publication No. 2016-171336, U.S. Patent Publication No. 2015-036920, Japanese Patent Publication No. 2016-110232, etc., and these records are introduced into this specification by reference.

Various embodiments of the present invention have been described above. However, the above embodiments can of course be modified by those skilled in the art within the scope of the present invention. The device of this embodiment can be appropriately modified according to the needs of the game to which it is applied.

Explanation of symbols

1 playing cards

1s Multiple shuffled playing cards

2 surveillance cameras

3 Card distribution device

4 Game Tables

5 Dealer

6 Guests (game participants/players)

7 chairs

8 Placement Area

9. Substitute currency for gaming

10 regions

10P Player Area

10B Banker Area

11 Game Recording Device

12 Image Analysis Device

13 Result display light

14 Control Device

14C Card Issuance Detection Device

15 Output (abnormal determination results, etc.)

16 Abnormal display light

30 Issuance of restraining devices

33 Slit

34 Locking parts

35 drive unit

36 Locking parts

37 Drive unit

40 Issuance of restraining devices

102 Card storage unit

103 Index

105 Card Guidance Department

106 opening

107 Card Guide

109 Control Department

112 Side Monitor

Claims (7)

1. A management system for tabletop games, specifically for tabletop games in an amusement park with multiple game tables, comprising: At least one game recording device records the state of the game being played at the game table as video using a camera; The image analysis device performs image analysis on the recorded game state images; and The control device uses the image analysis device to identify the location, type, and quantity of game tokens placed by each player. The control device is capable of identifying individual players at each play position on the game table, wherein each player is assigned an identification number for identification, and has the function of associating the game tokens with the identification number of the player who placed the game tokens. The control device also has the following functions: it can use artificial intelligence or machine learning technology to identify the location, type, and number of game tokens placed by the players in each game; identify the amount of game tokens placed by the players in each game; and identify the total amount and historical record of game tokens placed by each player at a predetermined time point, and record the identified (1) amount of game tokens placed by each player on the game table and/or (2) total amount or historical record of game tokens placed by each player at a predetermined time point. The camera is positioned diagonally above the game table, enabling it to capture images of multiple placement areas. The camera captures images of these placement areas positioned in the shooting direction of the camera from diagonally above, generating images that allow the control device to identify the location, type, and number of game tokens placed in these multiple placement areas based on the images. 2. The management system for the desktop game as described in claim 1, wherein, The control device also has the following configuration: it is able to identify the increase or decrease of the player's game currency in the game calculated based on the game's win or loss results, and it is able to identify the historical records in the game, and it is able to identify the win or loss history and income/expenditure amounts of each player at each point in time after a specified number of games have passed. 3. The management system for desktop games according to claim 2, wherein, It also includes a win/loss determination device for determining the win/loss results of each game at the game table. The outcome of the game is determined by the outcome determination device. 4. The management system for a desktop game according to any one of claims 1 to 3, wherein, The control device is a structure capable of identifying individual players whose placement amount, winning amount, and/or win/loss history are unusual compared to statistical data, and has the function of notifying the existence of the identified players. 5. The management system for a desktop game as described in claim 4, wherein, The abnormal condition is at least one of the following: 1) When the winnings exceed the specified amount; 2) The situation where the amount of substitute currency used in the game was small when losing but large when winning persisted for many games, which is an abnormal situation compared with the statistical data. 3) An anomaly occurs when the actual winnings of each player are compared to the average winnings that can be statistically expected based on the historical data of the placed amounts; 4) The positions of the game tokens placed by each player and the amount of game tokens obtained are abnormal compared to the statistical data; 5) When the amount placed exceeds the specified amount, the win rate is abnormal compared to the statistical data. 6. The management system for a desktop game according to any one of claims 1 to 3, wherein, The control device determines the player's function by using artificial intelligence or machine learning techniques to extract an image of the player's face for identification. 7. The management system for a tabletop game according to any one of claims 1 to 3, comprising two game recording devices.