Chip recognizing and learning system

Abstract

Provided is a chip recognition learning system capable of accurately recognizing chips wagered by a player. The chip recognition learning system (10) includes a game recording device (11) that records a state of chips (W) stacked on a gaming table (4) as an image using a camera (212); a chip determination device (12) that includes an artificial intelligence device (12a) that performs image analysis on the recorded image of the state of the chips (W) to determine the number and kind of chips (W) wagered by a player (C); and a teacher device (13) that, in a case where it is determined that the determination result of the chip determination device (12) is erroneous, inputs, as teacher data, the image used in the determination of the chip determination device (12) and the correct number and kind of chips (W) for the error to the artificial intelligence device (12a) and causes the artificial intelligence device (12a) to learn.

Description

TECHNICAL FIELD

[0001] The present application relates to a chip recognition learning system. BACKGROUND

[0002] In a game such as baccarat, a customer (player) stacks a plurality of chips on a table to place a bet. Therefore, it is necessary to accurately recognize the stacked chips. Further, an example of a chip for a game is disclosed in International Publication No. 2008 / 120749. SUMMARY

[0003] An object of the present application is to provide a chip recognition learning system capable of accurately recognizing chips placed by a player.

[0004] A chip recognition learning system according to one embodiment of the present application is used in a game arcade having a game table, and includes a game recording device that records a state of chips stacked on the game table as an image using a camera; a chip determination device that includes an artificial intelligence device that performs image analysis on the recorded image of the state of the chips, and determines the number and kind of chips placed by a player; and a teacher device that, in a case where it is determined that the determination result of the chip determination device is suspected to be wrong, inputs an image used in the determination of the chip determination device and the number and kind of chips correct for the wrong to the artificial intelligence device as teacher data, and causes the artificial intelligence device to learn.

[0005] According to such a mode, in a case where it is determined that the determination result of the chip determination device is suspected to be wrong, the teacher device inputs an image used in the determination of the chip determination device and the number and kind of chips correct for the wrong to the artificial intelligence device as teacher data, and causes the artificial intelligence device to learn, so the artificial intelligence device can effectively learn with respect to an image mode for which the determination precision of the chip determination device is relatively low, and can improve the determination precision of the chip determination device with emphasis on the image mode. By repeating such teacher learning, the chip determination device can accurately recognize chips placed by a player regardless of the state in which the chips are stacked.

[0006] In the chip recognition learning system according to one embodiment of the present application, the teacher device can further input an image used in the determination of the chip determination device and the number and kind of chips of the determination result to the artificial intelligence device as teacher data in a case where it is determined that the determination result of the chip determination device is correct, and cause the artificial intelligence device to learn.

[0007] According to such a manner, not only the image pattern for which the determination accuracy of the chip determination device is relatively low, but also the image pattern for which the determination accuracy is relatively high, the determination accuracy of the chip determination device can be further improved. Thus, the chip determination device can further accurately recognize the chips bet by the players.

[0008] The chip recognition learning system of one embodiment of the present application is used in a game house having a game table, and includes a game recording device that records the state of chips stacked on the game table as an image using a camera and a chip determination device that includes an artificial intelligence device that performs image analysis on the recorded image of the state of the chips, determines the number and kind of chips bet by a player, and learns by inputting, as teacher data, an image used in determination of the chip determination device and the correct number or kind of chips for an error in the determination of the chip determination device when it is determined that the determination result of the chip determination device is suspected to have an error.

[0009] According to such a manner, when it is determined that the determination result of the chip determination device is suspected to have an error, the artificial intelligence device learns by inputting, as teacher data, an image used in determination of the chip determination device and the correct number and kind of chips for an error. Thus, the artificial intelligence device can effectively learn for an image pattern for which the determination accuracy of the chip determination device is relatively low, and can improve the determination accuracy of the chip determination device for the image pattern. By repeating this, the chip determination device can accurately recognize the chips bet by the players regardless of the state of the chips.

[0010] In the chip recognition learning system of one embodiment of the present application, the control device that determines whether the determination result of the chip determination device is correct, the chip determination device that determines the kind and number of chips in a chip tray of the game table and the position, kind, and number of chips bet by each player based on an image recorded in the game recording device in a game played on the game table, and the control device that grasps the actual total amount of chips in the chip tray when the bet chips of each player are all collected, calculates the total amount of chips in the chip tray based on the determination result of the chip determination device, compares the total amount of chips in the chip tray with the actual total amount of chips in the chip tray, and determines that the determination result of the chip determination device is suspected to have an error when there is a difference between the total amount and the actual total amount, can be included. The total amount of chips in the chip tray is obtained by adding the total amount of chips in the chip tray before settlement of each game to the increase in the chip tray in the game calculated from the kind and number of chips bet by a losing player.

[0011] According to such a manner, it is possible to automatically determine by the control device whether there is a suspicion of an error in the determination result of the chip determination device.

[0012] The recognition learning system for chips of one embodiment of the present application is used in a gambling house having a gambling table, and is characterized by including: a game recording device that records a state of chips stacked on the gambling table as an image with a camera; a chip determination device that includes an artificial intelligence device that performs image analysis on the image of the recorded state of chips, determines the number and kind of chips that a player has bet; and a teacher device that, in a case where it is determined that the determination result of the chip determination device is correct, inputs the image used in the determination of the chip determination device and the number and kind of chips of the determination result as teacher data to the artificial intelligence device and causes the artificial intelligence device to learn.

[0013] The recognition learning system for chips of one embodiment of the present application is used in a gambling house having a gambling table, and is characterized by including: a game recording device that records a state of chips stacked on the gambling table as an image with a camera; and a chip determination device that includes an artificial intelligence device that performs image analysis on the image of the recorded state of chips, determines the number and kind of chips that a player has bet, and, in a case where it is determined that the determination result of the chip determination device is correct, inputs the image used in the determination of the chip determination device and the number or kind of chips of the determination result as teacher data by a teacher device and learns.

[0014] In the chip recognition learning system of one embodiment of the present application, the control device can also determine whether the determination result of the chip determination device is correct. The chip determination device can determine the kind and the number of chips in the chip tray of the gaming table and the kind, the number, and the position of the chips wagered by each player based on the images recorded in the game recording device in the game played at the gaming table. The control device can grasp the actual total amount of chips in the chip tray when the wagered chips of each player are all collected. The control device can calculate the proper total amount of chips in the chip tray based on the determination result of the chip determination device, compare the proper total amount of chips in the chip tray with the actual total amount of chips in the chip tray, and determine that the determination result of the chip determination device is correct when the proper total amount of chips and the actual total amount of chips are consistent. The proper total amount of chips in the chip tray is obtained by adding the total amount of chips in the chip tray before settlement of each game to the increase in the chip tray in the game calculated from the kind and the number of chips wagered by the losing player.

[0015] In the chip recognition learning system of one embodiment of the present application, the control device can grasp the actual total amount of chips in the chip tray with reference to the RFID provided on the chip.

[0016] According to such a mode, the control device can automatically grasp the actual total amount of chips in the chip tray using the RFID, and can improve the measurement accuracy compared to visual inspection by a staff member.

[0017] In the chip recognition learning system of one embodiment of the present application, the control device can have an artificial intelligence device for correct answer determination that is different from the artificial intelligence device of the chip determination device. The artificial intelligence device for correct answer determination can grasp the actual total amount of chips in the chip tray based on the images recorded in the game recording device.

[0018] According to such a mode, the control device can automatically grasp the actual total amount of chips in the chip tray using the artificial intelligence device for correct answer determination, and can improve the measurement accuracy compared to visual inspection by a staff member.

[0019] In the chip recognition learning system of one embodiment of the present application, the game recording device can record images obtained from a camera with an index or a time, or a tag for determining a chip collection situation or a chip payment situation, so that the game recording device can be analyzed by the chip determination device at a later time.

[0020] According to such a manner, the chip determination device can easily determine the image of the state of the chip that should be an analysis object according to the recording content of the game recording device by using the index, the time, and the label given to the image, and can shorten the time required for the determination.

[0021] In the chip recognition learning system of one embodiment of the present application, even when a part or the entire chip is hidden because a plurality of chips placed on the gaming table fall into a blind spot of the camera, the chip determination device can determine the kind, the number, and the position of the wagered chip.

[0022] According to such a manner, in particular, even when a part or the entire chip is hidden because a plurality of chips placed on the gaming table fall into a blind spot of the staff, by causing the chip determination device to determine the chip wagered by the player, the measurement accuracy can be improved compared to visual inspection by the staff.

[0023] The chip recognition learning method of one embodiment of the present application is used in a gaming establishment having a gaming table, and includes a game recording step of recording the state of chips stacked on the gaming table as an image with a camera; a chip determination step performed by an artificial intelligence device that performs image analysis on the recorded image of the state of the chips and determines the number and kind of chips wagered by a player; and a teacher step of, in a case where it is suspected that the determination result in the chip determination step is wrong, inputting the image used in the determination in the chip determination step and the correct number and kind of chips for the error as teacher data to the artificial intelligence device and causing the artificial intelligence device to learn.

[0024] According to such a manner, in a case where it is suspected that the determination result in the chip determination step is wrong, the image used in the determination in the chip determination step and the correct number and kind of chips for the error are inputted as teacher data to the artificial intelligence device in the teacher step and the artificial intelligence device is caused to learn, so the artificial intelligence device can effectively learn with respect to an image pattern for which the determination accuracy in the chip determination step is relatively low and can improve the determination accuracy with emphasis on the image pattern. Through repetition of this, in the chip determination step, chips can be recognized with high accuracy regardless of the state in which the chips are stacked.

[0025] The recognition learning method of the chip of one embodiment of the present application is used in a casino having a gaming table, and is characterized by including: a game recording step of recording a state of chips stacked on the gaming table as an image with a camera; a chip determination step performed by an artificial intelligence device that performs image analysis on the recorded image of the state of the chips and determines the number and kind of chips wagered by a player; and a teacher step of inputting, in a case where the determination result in the chip determination step is determined to be correct, the image used in the determination in the chip determination step and the number and kind of chips of the determination result as teacher data to the artificial intelligence device and causing the artificial intelligence device to perform learning. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a diagram schematically illustrating a casino having a chip recognition learning system of the first embodiment.

[0027] Figure 2 is a diagram for explaining a progress of a baccarat game.

[0028] Figure 3 is a block diagram illustrating a schematic configuration of a chip recognition learning system of the first embodiment.

[0029] Figure 4 is a flowchart for explaining a chip recognition learning method.

[0030] Figure 5 is a flowchart for explaining a modification example of a chip recognition learning method.

[0031] Figure 6 is a flowchart for explaining another modification example of a chip recognition learning method. Figure 7 is a diagram schematically illustrating a casino having a chip recognition learning system of the second embodiment.

[0032] Figure 8 is a block diagram illustrating a schematic configuration of a chip recognition learning system of the second embodiment. DETAILED DESCRIPTION

[0033] Hereinafter, embodiments of the present application will be described in detail with reference to the drawings. Note that the same portions in the drawings are denoted by the same reference numerals, and repeated description thereof will be omitted.

[0034] First, a game played in a casino having a gaming table 4 will be described. In this embodiment, a case where the gaming table 4 is a baccarat table and a baccarat game is played will be described, but the present application can be applied to other casinos or other games.

[0035] Figure 1 is a view schematically showing a gambling house having the chip recognition learning system 10 of the first embodiment. As shown in Figure 1 , a substantially semicircular gambling table 4 and a plurality of chairs 201 arranged along the arc side of the gambling table 4 in a manner facing a dealer D are arranged in the gambling house. The number of chairs 201 is arbitrary, and in the example shown in Figure 1 , six chairs 201 are arranged. Further, a bet area BA is provided on the gambling table 4 corresponding to each chair 201. That is, in the example shown in the drawing, six bet areas BA are arranged in a circular arc shape.

[0036] As shown in Figure 1 , a customer (player) C is seated on each chair 201. The customer (player) C places a bet on the result of a baccarat game by stacking chips W on the bet area BA provided in front of the seated chair 201, and bets on which of the player (PLAYER) and the banker (BANKER) wins or a tie (TIE) is achieved as the result of the baccarat game.

[0037] The chips W for betting can be of one kind or of a plurality of kinds. Alternatively, the number of chips W for betting can be arbitrarily determined by the customer (player) C. The chip recognition learning system 10 of the present embodiment can recognize the number and kind of chips W stacked and arranged.

[0038] The dealer D performs the following actions in order to end the betting of the customer (player) C: performs timing and prompts "No More Bet", moves the hand horizontally, and the like. Next, the dealer D draws cards one by one from a card dealing device S to the gambling table 4. As shown in Figure 2 , the first card is dealt to the player (PLAYER) hand, the second card is dealt to the banker (BANKER) hand, the third card is dealt to the player (PLAYER) hand, and the fourth card is dealt to the banker (BANKER) hand (hereinafter, the drawing of the first to fourth cards is referred to as "card dealing").

[0039] Further, all the cards are drawn from the card dealing device S in a state in which the back surface faces upward. Therefore, the rank (number) and suit (heart, spade, club, and diamond) of the drawn cards cannot be grasped by either the dealer D or the customer (player) C.

[0040] After the fourth card is drawn, the player C (in the case where there are multiple players betting on the player, the player C who has bet the highest amount, and in the case where there is no player betting on the player, the dealer D) who has bet on the player (player) turns the first and third cards face up. Similarly, the player C (in the case where there are multiple players betting on the banker, the player C who has bet the highest amount, and in the case where there is no player betting on the banker, the dealer D) who has bet on the banker (banker) turns the second and fourth cards face up (generally, this action of turning the cards face up is called "squeezing").

[0041] Then, based on the ranks of the first to fourth cards and the detailed rules of the baccarat game, the dealer D draws the fifth card, and further draws the sixth card, and deals these cards to the respective players (players) or the banker (banker). Similarly, the player C who has bet on the player (player) turns the cards dealt to the player (player) face up, and the player who has bet on the banker (banker) turns the cards dealt to the banker (banker) face up.

[0042] The time from when the first to fourth cards are drawn until the fifth and sixth cards are turned face up to determine the winning and losing result is the most enjoyable for the player C.

[0043] Furthermore, depending on the ranks of the cards, the winning and losing result is sometimes determined by the first to fourth cards, or sometimes not determined until the fifth card, or even the sixth card. The dealer D performs the following work: based on the ranks of the cards after the cards are turned face up, the dealer D grasps the situation where the winning and losing result has been determined, and the winning and losing result, presses the winning and losing result display button on the card dealing device S, and displays the winning and losing result on a monitor or the like in order to inform the player C of the winning and losing result.

[0044] In addition, the winning and losing result of the game is determined using the winning and losing determination section of the card dealing device S. In the case where the dealer D does not display the winning and losing result even though the winning and losing result has been determined, an error is reported. The card dealing device S detects the above error, and outputs an error signal. Finally, the dealer D settles the amount of bet of the player C during the period when the winning and losing result is displayed, pays the player C who has won, and collects the amount of bet of the player C who has lost. After the settlement is completed, the display of the winning and losing result is ended, and the betting for the next game is started.

[0045] Further, the flow of the above-described baccarat game is widely performed in a general casino, and the card dealing device S is a conventional card dealing device that adopts a structure in which a card is drawn by a hand of a dealer D, and that can read the drawn card, and further has a result display button and a result display portion, and has a function of performing win / loss determination and displaying a win / loss result. On a general casino floor, the card dealing device S and a monitor or the like are arranged with respect to each of a plurality of game tables 4, and cards used are supplied to each game table 4 or a cabinet below each game table 4 in units of a pack or a set, or even in units of a box, and are used.

[0046] The chip recognition learning system 10 of the present embodiment is a system for learning recognition of chips W stacked and arranged in a betting area BA by a customer (a player) C, and more specifically, is a system for learning recognition of the number and / or kind of chips W.

[0047] As shown in Figure 1 In the present embodiment, a monitoring camera 212 that photographs the state of chips W stacked and arranged in a betting area BA is provided outside the game table 4. In addition, an RFID (Radio Frequency Identification) is provided on each chip W, and an RFID reading device 22 that reads the RFID of the chips W in a chip tray 23 is provided on the chip tray 23 managed by the dealer D.

[0048] The chip recognition learning system 10 of the present embodiment is communicably connected to the monitoring camera 212 and the RFID reading device 22, respectively.

[0049] Figure 3 is a block diagram showing the schematic configuration of the chip recognition learning system 10 of the present embodiment.

[0050] As shown in Figure 3 The chip recognition learning system 10 has a game recording device 11, a chip determination device 12, a teacher device 13, and a control device 14. Further, at least a part of the chip recognition learning system 10 is implemented by a computer.

[0051] The game recording device 11 has a solid-state type data storage such as a hard disk. The game recording device 11 records the state of chips W stacked on the game table 4 as an image photographed by the camera 212. Further, the image can be a dynamic image, or can be a continuous still image.

[0052] The game recording device 11 can also record by assigning an index or time to the image obtained from the camera 212, or by assigning a label to determine the recycling or payment scenario of the chip W, so that the game record can be analyzed later by the chip determination device 12 described later.

[0053] The chip determination device 12 includes an artificial intelligence device 12a that uses image recognition technology, such as deep learning, to analyze images of the state of chips W recorded by the game recording device 11, thereby determining the number and type of chips W bet by the customer (player) C. The chip determination device 12 can further determine the position of the chips W bet by the customer (player) C on the betting area BA.

[0054] The chip determination device 12 can also perform image analysis on the image of the state of the chips W recorded in the game recording device 11 to determine the number and type of chips W in the chip tray 23 before each game settlement.

[0055] like Figure 3 As shown, the chip determining device 12 outputs the determining result to the output device 15. The output device 15 can output the determining result of the chip determining device 12 as text information to the monitor on the game table 4, or as audio information to the dealer D's headphones, etc.

[0056] Control device 14 is used to determine whether the determination result of chip determination device 12 is correct. When control device 14 finishes collecting all the chips W (used chips) bet by the losing guest (player) C, it knows the actual total amount V0 of chips W in chip tray 23.

[0057] In this embodiment, the control device 14 obtains the RFID information of the chips W in the chip tray 23 from the RFID reading device 22, and determines the type and number of chips W in the chip tray 23 based on the obtained RFID information, and knows the actual total amount V0 of chips W in the chip tray 23.

[0058] Additionally, the control device 14 obtains the determination result from the chip determination device 12. Based on the obtained determination result, it calculates the total amount V1 according to the type and number of chips W in the chip tray 23 before each game's settlement, and calculates the total amount of chips W bet by the losing player C (i.e., the increase in chip tray 23 for that game) V2 according to the position, type, and number of chips W bet by each player C. Then, the control device 14 adds the increase in chip tray 23 for that game to the total amount V1 of chips W in the chip tray 23 before each game's settlement, thereby calculating the proper total amount V3 (=V1+V2) of chips in the chip tray 23.

[0059] The control device 14 compares the total amount V3 of chips W that should be in the chip tray 23 with the actual total amount V0 of chips W in the chip tray 23, and when there is a difference between the total amount V3 and the actual total amount V0 (V3≠V0), determines that the determination result of the chip determination device 12 is suspected to be incorrect. On the other hand, when the total amount V3 and the actual total amount V0 coincide (V3=V0), the control device 14 determines that the determination result of the chip determination device 12 is correct.

[0060] When the chips W are collected from the losing player C is finished, the chips W are paid to the winning player C. The control device 14 calculates the total amount of chips W that the winning player C has bet based on the positions, types, and numbers of chips W that each player C has bet, and the amount of money V4 that should be paid corresponding to the total amount of chips W that the winning player C has bet. The control device 14 grasps the actual total amount of chips W in the chip tray 23 after the chips W are reduced due to the payment of chips W, determines whether the actual total amount coincides with the amount of money V4 to be paid, and displays a light indicating whether or not it coincides based on the determination result.

[0061] The control device 14 compares the total amount V5 (=V1+V2-V4) of chips W that should be in the chip tray 23 with the actual total amount of chips W in the chip tray 23 after the chips W are increased by the collected chips and reduced by the paid chips, and when there is a difference, determines that the determination result of the chip determination device 12 is suspected to be incorrect. The control device 14 determines that the determination result of the chip determination device 12 is correct when the total amount V5 and the actual total amount coincide.

[0062] For the determination of whether or not it coincides, respectively, the light can be lit, for example, in a manner that the light is lit in green if it coincides and the light is lit in red if it does not coincide.

[0063] The teacher device 13 obtains the determination of whether or not the determination result of the chip determination device 12 is correct from the control device 14. The teacher device 13 can input the image used in the determination of the chip determination device 12 (including the determination suspected to be incorrect) and the number and type of chips W that are correct for the error to the artificial intelligence device 12a of the chip determination device 12 as teacher data and cause the artificial intelligence device 12a to learn in a case where the determination result of the chip determination device 12 is determined by the control device 14 to be suspected to be incorrect. Further, the number and type of chips that are correct for the error are actually confirmed by a person who confirms the image and teaches the teacher device 13. That is, the teacher device 13 learns by a person teaching through the image at the time of the error and the device of the correct number at that time for the number and type of chips that are correct for the error.

[0064] The teacher device 13 can also input the image used in the (correct) judgment of the chip judgment device 12 and the number and type of chips W in the judgment result of the chip judgment device 12 (i.e. the correct number and type of chips W) as teacher data into the artificial intelligence device 12a of the chip judgment device 12, and enable the artificial intelligence device 12a to learn.

[0065] The teacher device 13 repeatedly inputs the aforementioned teacher data into the artificial intelligence device 12a of the chip determination device 12, enabling the AI ​​device 12a to learn such teacher actions. This improves the accuracy of the chip determination device 12 in determining the chip W. The AI ​​device 12a of the chip determination device 12 performs image analysis on the image of the state of the chip W to determine the chip W. Therefore, even if multiple chips W placed on the game table 4 are partially or completely hidden due to falling into the blind spot of the camera 212, the type, number, and position of the bet chip W can still be determined by repeatedly learning from such incomplete images.

[0066] Next, refer to Figure 4 The operation (chip recognition and learning method) of the chip recognition and learning system 10 of this embodiment will be explained.

[0067] like Figure 4 As shown, firstly, when guest (player) C stacks chips W in the betting area BA of game table 4, camera 212 captures the state of the stacked chips W as an image, and game recording device 11 records the image (step S31).

[0068] Next, the chip determination device 12 performs image analysis on the image recorded in the game recording device 11 to determine the number and type of chips W bet by the customer (player) C (step S32). Furthermore, the image analyzed by the chip determination device 12 can also be an image selected based on an index, time, or a label assigned to the image by the game recording device 11 to determine the chip W's collection or payment scenario.

[0069] In step S32, the image analysis of the state of the chips W recorded in the game recording device 11 can also be performed by the chip determination device 12 to determine not only the number and type of chips W bet by the customer (player) C, but also the position of the chips W bet by the customer (player) C on the betting area BA, and the number and type of chips W in the chip tray 23 before each game settlement.

[0070] The determination result of the chip determination device 12 is output to the output device 15. The determination result of the chip determination device 12 can be output to the monitor on the game table 4 as text information or to the earphone of the dealer D as sound information through the output device 15.

[0071] The determination result of the chip determination device 12 is also sent to the control device 14, and the control device 14 determines whether the determination result of the chip determination device 12 is correct (step S33).

[0072] In a case where it is determined by the control device 14 that the determination result of the chip determination device 12 is suspected to be incorrect (step S34: No), the image used in the determination of the chip determination device 12 (including the determination suspected to be incorrect) and the correct number and kind of chips W for the error are input to the artificial intelligence device 12a of the chip determination device 12 as teacher data from the teacher device 13, and the artificial intelligence device 12a is caused to learn (step S36).

[0073] On the other hand, in a case where it is determined by the control device 14 that the determination result of the chip determination device 12 is correct (step S34: Yes), the operation of the chip recognition learning system 10 in the game is ended.

[0074] As described above, according to the present embodiment, in a case where it is determined that the determination result of the chip determination device 12 is suspected to be incorrect, the teacher device 13 inputs the image used in the determination of the chip determination device 12 and the correct number and kind of chips W for the error to the artificial intelligence device 12a as teacher data and causes the artificial intelligence device 12a to learn, and thus the artificial intelligence device 12a can effectively learn with respect to the image pattern for which the determination accuracy of the chip determination device 12 is relatively low and can improve the determination accuracy of the chip determination device 12 with emphasis on the image pattern. By repeating such teacher learning, the chip determination device 12 can recognize the chips W bet by the player C with high accuracy regardless of the state in which the chips W are stacked.

[0075] In addition, according to the present embodiment, the control device 14 grasps the actual total amount V0 of the chips W in the chip tray 23 when the wagered chips W used by each player C are all collected, calculates the required total amount V3 of the chips W in the chip tray 23 based on the determination result of the chip determination device 12 (where the required total amount of the chips W in the chip tray 23 is the total amount V1 of the chips W in the chip tray 23 before settlement of each game plus the increase V2 of the chips W in the chip tray 23 according to the kind and the number of the chips W wagered by the losing player C in the game. V3 = V1 + V2), compares the required total amount V3 of the chips W in the chip tray 23 with the actual total amount V0 of the chips W in the chip tray 23, and when there is a difference between the required total amount V3 and the actual total amount V0 (V3 ≠ V0), determines that there is a suspicion that the determination result of the chip determination device 12 is wrong, and thus, the control device 14 can automatically determine whether there is a suspicion that the determination result of the chip determination device 12 is wrong.

[0076] In addition, according to the present embodiment, the control device 14 grasps the actual total amount V0 of the chips W in the chip tray 23 based on the RFID provided on the chips W, and thus, the control device 14 can automatically grasp the actual total amount V0 of the chips W in the chip tray 23 using the RFID, and can improve the measurement accuracy compared to the visual inspection by the staff.

[0077] In addition, according to the present embodiment, the game recording device 11 records in a manner of assigning an index or a time to the image acquired from the camera 212 or assigning a tag for determining the chip collection situation or the chip payment situation, and thus, the chip determination device 12 can easily determine the image of the state of the chips W that should be the analysis target based on the recording content of the game recording device 11 by using the index, the time, or the tag assigned to the image, and can shorten the time required for the determination.

[0078] In addition, according to the present embodiment, even when a part or the entire of the plurality of chips W placed on the gaming table 4 becomes hidden due to falling into the dead angle of the camera 212, the chip determination device 12 can determine the kind, the number, and the position of the chips W wagered, and thus, particularly in the case where a part or the entire of the plurality of chips W placed on the gaming table 4 becomes hidden due to falling into the dead angle of the staff, by causing the chip determination device 12 to determine the chips W wagered by the player C, the measurement accuracy can be improved compared to the visual inspection by the staff.

[0079] Further, various modifications can be made to the above-described embodiments. One example of a modification will be described below with reference to the drawings. In the following description and the drawings used in the following description, for portions that can be configured identically to the above-described embodiments, the same reference numerals are used as those used for corresponding portions in the above-described embodiments, and repeated description is omitted.

[0080] Figure 5 is a flowchart for explaining a modification example of the recognition learning method of the chips.

[0081] In Figure 5 In the example shown, in a case where it is determined by the control device 14 that the determination result of the chip determination device 12 is suspected to be incorrect (step S34: No), the image used in the determination (suspected to be incorrect) of the chip determination device 12 and the number and kind of the correct chips W for the error are input as teacher data from the teacher device 13 to the artificial intelligence device 12a of the chip determination device 12, and the artificial intelligence device 12a is further caused to learn (step S36).

[0082] On the other hand, in a case where it is determined by the control device 14 that the determination result of the chip determination device 12 is correct (step S34: Yes), the image used in the determination (correct) of the chip determination device 12 and the number and kind of the chips W of the determination result of the chip determination device 12 (i.e., the number and kind of the correct chips W) are further input as teacher data from the teacher device 13 to the artificial intelligence device 12a of the chip determination device 12, and the artificial intelligence device 12a is further caused to learn (step S35).

[0083] According to such a manner, not only the image mode for which the determination accuracy of the chip determination device 12 is relatively low, but also for the image mode for which the determination accuracy is relatively high, the determination accuracy can be further improved, and by so doing, the chip determination device 12 can further accurately recognize the chips W bet by the player C.

[0084] Figure 6 is a flowchart for explaining another modification example of the recognition learning method of the chips.

[0085] In Figure 6 In the example shown, in a case where it is determined by the control device 14 that the determination result of the chip determination device 12 is correct (step S34: Yes), the image used in the determination (correct) of the chip determination device 12 and the number and kind of the chips W of the determination result of the chip determination device 12 (i.e., the number and kind of the correct chips W) are further input as teacher data from the teacher device 13 to the artificial intelligence device 12a of the chip determination device 12, and the artificial intelligence device 12a is further caused to learn (step S35).

[0086] On the other hand, in a case where it is determined by the control device 14 that the determination result of the chip determination device 12 is suspected to be wrong (Step S34: No), the operation of the chip recognition learning system 10 in the game is ended.

[0087] According to such a manner, since the image used in the determination of the chip determination device 12 and the number and kind of the (correct) chips W of the determination result are inputted to the artificial intelligence device 12a as the teacher data by the teacher device 13 in a case where it is determined that the determination result of the chip determination device 12 is correct, and the artificial intelligence device 12a is caused to learn, the artificial intelligence device 12a can effectively learn with respect to the image pattern for which the determination accuracy of the chip determination device 12 is relatively high, and can improve the determination accuracy of the chip determination device 12 with emphasis on the image pattern. By repeatedly performing such teacher learning, the chip determination device 12 can recognize the chips W bet by the player C with high accuracy.

[0088] Figure 7 is a diagram schematically showing a game arcade having the chip recognition learning system 100 of the second embodiment. Figure 8 is a block diagram showing the schematic configuration of the chip recognition learning system 100 of the second embodiment.

[0089] As shown in Figure 7 In the second embodiment, in addition to the monitoring camera 212 for photographing the state of the chips W stacked and arranged in the betting area BA, a chip tray monitoring camera 24 for photographing the state of the chips W in the chip tray 23 managed by the dealer D is provided outside the game table 4.

[0090] The chip recognition learning system 100 of the second embodiment is communicably connected to the monitoring camera 212 and the chip tray monitoring camera 24, respectively.

[0091] As shown in Figure 8 The game recording device 11 records the state of the chips W in the chip tray 23 as an image photographed by the chip tray monitoring camera 24. Further, the image can be a dynamic image, or a continuous still image.

[0092] The control device 14 has an artificial intelligence device 14a (correct answer determination artificial intelligence device) different from the artificial intelligence device 12a of the chip determination device 12, which performs image analysis on the image of the state of the chips W in the chip tray 23 recorded by the game recording device 11, determines the number and kind of the chips W in the chip tray 23, and grasps the actual total amount V0 of the chips W in the chip tray 23, which performs image recognition using, for example, deep learning technology.

[0093] According to the second embodiment, the control device 14 is able to automatically grasp the actual total amount of the chips W in the chip tray 23 using the correct answer determination artificial intelligence device 14a, and is able to improve the measurement accuracy compared to the visual inspection by the staff.

[0094] Further, there is a technique called teacherless data learning, but this teacherless data learning is a technique that teaches the artificial intelligence whether the result determined by the artificial intelligence is correct or incorrect, which is also called teacher data learning in the present application. The above-described embodiments are described with the aim of enabling a person having ordinary knowledge in the technical field to which the present application pertains to implement the present application. As long as various modifications of the above-described embodiments are of course able to be implemented by a person skilled in the art, the technical idea of the present application is also able to be applied to other embodiments. Therefore, the present application is not limited to the described embodiments, but should be the widest range based on the technical idea defined by the scope of the claims.

Claims

1. A chip recognition learning system for a casino having a game table, the chip recognition learning system characterized by comprising: a camera configured to take an image including a plurality of stacks of chips bet by a player on a plurality of betting objects of the gaming table arranged in a depth direction from the camera; at least one processor, the at least one processor configured to, execute an artificial intelligence program to analyze the images of the chips in the plurality of stacks of chips stacked on the game table to determine the number and kind of the chips, and use one or more images related to an error in the determination of the number or kind of the chips as new teacher data for the artificial intelligence program to learn, thereby increasing teacher data learned by the artificial intelligence program, the new teacher data further including a correct number and a correct kind of chips related to the one or more images.

2. The chip recognition learning system according to claim 1, characterized in that, in a case where it is determined that a determination result is correct, the at least one processor is further configured to input one or more images used for the determination and the determined number and kind of chips to an artificial intelligence program as teacher data to allow the artificial intelligence program to learn.

3. The chip recognition learning system according to claim 1, characterized in that, the at least one processor is configured to, determine the kind and number of chips in a chip tray included in the game table using the images, and perform the determination of the kind and number of the chips wagered by each of a plurality of players playing a game on the game table and the recognition of the position of the chips wagered by the player from the images for each of the plurality of players playing the game on the game table, determine an actual total amount of the chips in the chip tray when all of the used chips wagered by the player are collected, calculate an appropriate total amount of the chips in the chip tray based on the recognition based on the images by adding the total amount of the chips in the chip tray before clearing of each game to an increase in the chip tray in the game calculated from the kind and number of the chips wagered by a player who lost the game, compare the appropriate total amount of the chips in the chip tray with the actual total amount of the chips in the chip tray with each other, and determine that there is an error when there is a difference between the appropriate total amount and the actual total amount.

4. The chip recognition learning system according to claim 3, characterized in that, the at least one processor is configured to determine the actual total amount of the chips in the chip tray based on radio frequency identification provided on the chips.

5. The chip recognition learning system according to claim 3, characterized in that, the at least one processor is configured to use an artificial intelligence different from the artificial intelligence program used to determine the number and kind of the chips wagered by the player for the calculation of the appropriate total amount including the actual total amount of the chips in the chip tray determined from the images. ​ ​ 6. The chip identification learning system according to claim 3, wherein the at least one processor is configured to record the images taken from the camera after the images are given an index, a time, or a label for determining a payout situation or a collection situation of the chips, whereby the recording of the game can be analyzed.

7. The chip identification learning system according to claim 1, wherein the kind, the number, and the position of the chips wagered are determined even if some of the plurality of chips placed on the gaming table are partially or entirely hidden due to a blind spot of the camera.

8. A chip identification learning method for use in a gaming house having a gaming table, the chip identification learning method comprising: executing an artificial intelligence program to analyze images of chips stacked on the gaming table taken by a camera configured to take images including a plurality of stacks of chips wagered on a plurality of wagering objects arranged in a depth direction on the gaming table from a view of the camera, to determine a number and a kind of chips wagered by a player; and using one or more images related to an error in the determination of the number or the kind of chips as new teacher data for the artificial intelligence program to learn, thereby increasing teacher data learned by the artificial intelligence program, the new teacher data further including a correct number and a correct kind of chips related to the one or more images.

9. A chip identification learning system for use in a gaming house having a gaming table, the chip identification learning system comprising: a camera configured to take an image including a plurality of stacks of the chips bet on a plurality of bet objects arranged in a depth direction from the camera on the game table; at least one processor, the at least one processor is configured to execute an artificial intelligence program to analyze images of a state of chips stacked on the gaming table taken by the camera, to identify a number and a kind of chips wagered by a player; in response to determining that the identification is correct, allow the artificial intelligence program to learn by inputting the images and the identified number and kind of chips as teacher data to the artificial intelligence program.

10. The chip identification learning system according to claim 9, wherein the at least one processor is configured to identify a kind and a number of chips in a chip tray included in the gaming table using the taken images, and perform the identification of the kind and the number of chips wagered by each of a plurality of players playing a game on the gaming table and the identification of a position of the chips wagered by the player separately for each of the players based on the taken images; determine an actual total amount of the chips in the chip tray when wagered and used chips wagered by the player are all collected. ​ calculating a proper total amount of the chips in the chip tray based on the recognition based on the image, by adding the total amount of the chips in the chip tray before each game is settled to an increase in the chip tray in the game calculated from the kind and the number of the chips bet by a loser of one of the games; comparing the proper total amount of the chips in the chip tray with the actual total amount of the chips in the chip tray with each other; and determining that the recognition is correct when the proper total amount and the actual total amount coincide.

11. The chip recognition learning system according to claim 10, wherein the at least one processor is configured to determine the actual total amount of the chips in the chip tray based on radio frequency identification provided on the chips.

12. The chip recognition learning system according to claim 10, wherein the at least one processor is configured to use an artificial intelligence different from the artificial intelligence used to perform the determination of the number and the kind of the chips bet by the players for the calculation of the proper total amount including the determination of the actual total amount of the chips in the chip tray based on the image.

13. The chip recognition learning system according to claim 10, wherein the at least one processor is configured to record the image taken from the camera after the image is given an index, a time, or a label for determining a chip collection situation or a chip payout situation, whereby the recording of the game can be analyzed.

14. The chip recognition learning system according to claim 9, wherein the kind, the number, and the position of the chips bet are determined even if some of a plurality of chips placed on the gaming table are partially or entirely hidden due to a blind spot of the camera.

15. A chip recognition learning system for use in a gaming house having a gaming table, the chip recognition learning system comprising: a camera configured to take an image including a plurality of stacks of chips bet on a plurality of bet objects arranged in a depth direction on the gaming table from the camera; and at least one processor, the at least one processor being configured to execute an artificial intelligence program to analyze images of the chips in the plurality of stacks of chips stacked on the gaming table during a period in which a game is played to determine the number and the kind of the chips, the artificial intelligence program receiving the images and the determined number and kind of the chips as teacher data for the determination and learning based on the received images and the number and the kind of the chips in a case where the determination of the number and the kind of the chips is determined to be correct.

16. A chip recognition learning method for use in a gaming house having a gaming table, the chip recognition learning method comprising: ​ ​ ​ ​ ​ ​ ​ An artificial intelligence program is executed to analyze an image of chips stacked on the game table during a game played on the game table, the image being acquired by a camera configured to capture an image including a plurality of stacks of the chips wagered on a plurality of wagering objects arranged in a depth direction on the game table as viewed from the camera, to determine the number and kind of the chips; and In response to determining that the determined number and kind of the chips are correct, the image and the determined number and kind of the chips are input as teacher data to the artificial intelligence program to allow the artificial intelligence program to learn.

Images