Securities processing system, settlement system, and securities processing method

Abstract

To improve the accuracy of security identification using a simple method. [Solution] The securities processing device 10 includes an identification unit 154 that determines whether or not to accept a gift certificate GT by identifying the gift certificate GT based on an image IIMG of the gift certificate GT. The identification unit 154 performs an identification process that includes a gift certificate type identification process that identifies the gift certificate type of the gift certificate GT based on a plurality of pre-registered reference images RIMG and an image IIMG of the gift certificate GT, and identifies the gift certificate GT based on the result of the identification process. The gift certificate type identification process includes a first process that uses AI to calculate the similarity between the image IIMG and the reference image RIMG to search for candidate images from a plurality of reference images RIMG that have a similarity to the image IIMG of the gift certificate GT that is equal to or greater than a standard value, and a second process that identifies the gift certificate type of the gift certificate GT based on the degree of agreement of the candidate image with the image IIMG of one or more predetermined feature parts FT.

Description

Technical Field

[0001] The present invention relates to a securities processing apparatus, a settlement processing system, and a securities processing method.

Background Art

[0002] In retail stores such as supermarkets, securities such as gift certificates may be handled in addition to currency. For example, Patent Document 1 discloses a mode in which a coin changer and a POS register are communicatively connected to each other, and a currency processing unit in which the POS register and a gift certificate processor are also communicatively connected to each other is installed at a cash desk in the front area of a store. The system disclosed in Patent Document 1 allows employees in a retail store or the like to identify gift certificates received from customers and perform settlement processing for purchased goods.

[0003] In recent years, there has been a growing consideration of using AI (Artificial Intelligence) for currency identification to accurately identify currency. For example, Patent Document 2 discloses comparing an infrared reflection image of an imaged banknote with infrared reflection image data based on a genuine banknote and performing authenticity determination of the banknote using machine learning.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] Incidentally, the variety of gift certificates is far greater than, for example, banknotes. Furthermore, with gift certificates, there are cases where only the issuing company or sponsoring company's information differs, the design remains the same, the design itself differs slightly depending on the issuance date, and the color changes slightly due to fading, etc. Therefore, it is extremely difficult to accurately distinguish between these types of gift certificates. In recent years, the use of AI processing to determine whether a gift certificate is genuine or counterfeit has been considered, as seen in the machine learning-based authenticity determination described in Patent Document 2. However, to accurately perform AI processing to distinguish between gift certificate types, which are far more numerous than banknotes, it is practically difficult to do so without preparing a vast amount of training data and continuously improving the accuracy of machine learning. For this reason, even if AI processing is used for gift certificate type discrimination, there is a risk that many gift certificates will end up being unidentifiable. If there are many unidentifiable gift certificates, supermarkets and other stores will have to call an employee each time, which will not only reduce customer turnover but also create mental health issues by accumulating customer stress. Furthermore, many customers will either choose to use a face-to-face checkout counter or give up on using their gift certificates. [Means for solving the problem]

[0006] A securities processing device according to a preferred embodiment of the present invention is a securities processing device that determines whether or not to accept a security that has been inserted, comprising: an insertion unit into which the security is inserted; a transport unit for transporting the security inserted into the insertion unit; an image reading unit that generates first image information showing a first image which is an image of the security; an identification unit that determines whether or not to accept the security by identifying the security based on the first image information; an operation control unit that controls the transport operation of the security by the transport unit according to the determination result of the identification unit; and a storage unit for storing the acceptable security, wherein the identification unit generates first image information showing a first image which is an image of each of a plurality of reference securities that have been registered in advance. An identification process is performed which includes a security type identification process that identifies the type of security based on a second image information showing an image and the first image information, and the security is identified based on the result of the identification process, and the security type identification process includes a first process which uses AI to calculate the similarity between the first image and the second image to search for candidate images from among a plurality of second images corresponding to a plurality of reference securities, where the similarity with the first image is equal to or greater than a standard value, as candidate images of the second image showing the reference securities corresponding to the security, and a second process which identifies the type of security based on the degree of agreement of one or more predetermined feature portions of the candidate image with respect to the first image.

[0007] Furthermore, a settlement processing system according to a preferred embodiment of the present invention is a settlement processing system having at least a POS register, a settlement processing device, and a securities processing device, wherein the securities processing device comprises: an input unit into which securities are inserted; a transport unit for transporting the securities inserted into the input unit; an image reading unit for generating first image information showing a first image which is an image of the securities; an identification unit for determining whether or not to accept the securities by identifying the securities based on the first image information; an operation control unit for controlling the transport operation of the securities by the transport unit according to the determination result of the identification unit; and a storage unit for storing the acceptable securities, wherein the identification unit performs an identification process including a securities type identification process for identifying the securities type based on a second image showing a second image which is an image of each of a plurality of reference securities that have been registered in advance and the first image information, identifies the securities based on the result of the identification process, and the securities type identification process includes the first image and the second image The operation control unit includes a first process that uses AI to calculate similarity to a plurality of reference securities to search for candidate images from a plurality of second images corresponding to the plurality of reference securities, where the similarity to the first image is equal to or greater than a threshold value, as candidates for the second image representing the reference securities corresponding to the securities; and a second process that identifies the type of securities based on the degree of agreement of one or more predetermined feature portions of the candidate images with respect to the first image. The operation control unit stores the securities in the storage unit by causing the transport unit to perform the transport operation to transport the securities to the storage unit when the identification unit determines that the securities are acceptable. The operation control unit discharges the securities to the outside when the identification unit determines that the securities are unacceptable. The POS register and the settlement processing device perform settlement processing for the products the customer wishes to purchase, according to the identification result in the identification unit, when the identification unit determines that the securities are acceptable.

[0008] Furthermore, a securities processing method according to a preferred embodiment of the present invention is a securities processing method executed by a securities processing device comprising: an input unit into which securities are inserted; a transport unit for transporting the securities inserted into the input unit; an image reading unit for generating first image information showing a first image which is an image of the securities; an identification unit for determining whether or not to accept the securities by identifying the securities based on the first image information; an operation control unit for controlling the transport operation of the securities by the transport unit according to the determination result of the identification unit; and a storage unit for storing the acceptable securities, wherein the identification unit shows a second image which is an image of each of a plurality of reference securities that have been registered in advance. An identification process is performed which includes a security type identification process that identifies the type of security based on the second image information and the first image information, and the security is identified based on the result of the identification process, and the security type identification process includes a first process which uses AI to calculate the similarity between the first image and the second image to search for candidate images from among a plurality of second images corresponding to the plurality of reference securitys, which have a similarity to the first image that is equal to or greater than a standard value, as candidate second images that represent the reference security corresponding to the security, and a second process which identifies the type of security based on the degree of agreement of one or more predetermined feature portions of the candidate image with respect to the first image. [Effects of the Invention]

[0009] According to the present invention, the accuracy of security identification can be improved in a simple manner. [Brief explanation of the drawing]

[0010] [Figure 1] This is a conceptual diagram illustrating the outline of the settlement processing system according to the first embodiment. [Figure 2] This is a schematic diagram illustrating the internal structure of a securities processing unit. [Figure 3] This is a diagram showing the configuration of a securities processing unit. [Figure 4] Figure 3 is an explanatory diagram illustrating an example of a ticket type table. [Figure 5] This is an explanatory diagram illustrating the overview of the ticket type identification process. [Figure 6] This is an explanatory diagram illustrating an example of a confirmation screen used to verify the identification result of a gift certificate inserted into the input section of a securities processing device. [Figure 7] This flowchart shows an example of the operation of a securities processing device when performing settlement processing using gift certificates. [Figure 8] This is a flowchart showing an example of the identification process. [Figure 9] This is a flowchart showing an example of ticket type identification processing. [Modes for carrying out the invention]

[0011] Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. Note that the dimensions and scale of each part in each drawing have been appropriately changed from the actual ones. Furthermore, the embodiments described below are preferred specific examples of the present invention and are subject to various technically preferred limitations, but the scope of the present invention is not limited to these embodiments unless otherwise stated in the following description.

[0012] [1. Embodiments] First, with reference to Figure 1, an example of the outline of the settlement processing system 1 according to the first embodiment will be described.

[0013] [First Embodiment] Figure 1 is a conceptual diagram illustrating the outline of the settlement processing system 1 according to the first embodiment. In Figure 1, the settlement processing system 1 includes a securities processing device 10 for processing securities such as gift certificates GT and service vouchers, a POS register 20 (including self-checkout and semi-self-checkout registers), and a change dispenser 30 which is a settlement processing device.

[0014] The securities processing device 10 is communicably connected to the POS register 20, and the POS register 20 is communicably connected to the change dispenser 30. The connection between the securities processing device 10 and the POS register 20, and the connection between the POS register 20 and the change dispenser 30 may be a connection using either wired or wireless means, or a connection using both wired and wireless means. Also, the connection between a plurality of elements including the securities processing device 10 and the like only needs to be a connection that enables communication between the plurality of elements, and may be a connection via a network. The network may include the Internet and a local area network. In the present embodiment, for example, it is assumed that the securities processing device 10, the POS register 20, and the change dispenser 30 are installed in a retail store such as a supermarket that handles gift tickets GT.

[0015] Details will be described later in FIG. 3, but the securities processing device 10 has, for example, an insertion unit 110, a temporary storage unit outlet 131, a storage unit outlet 141, and the like. In the present embodiment, the insertion unit 110 serves as both an insertion port into which the gift ticket GT is inserted and a discharge (reject) port for discharging the gift ticket GT inserted into the securities processing device 10. Whether to discharge the gift ticket GT inserted into the securities processing device 10 is determined by an identification process for identifying the gift ticket GT based on, for example, an image IIMG of the gift ticket GT. Also, the securities inserted into the securities processing device 10 can be taken out from the temporary storage unit outlet 131 or the storage unit outlet 141 as necessary.

[0016] The POS register 20 has, for example, a display operation unit 21 and is used when performing settlement processing for the goods purchased by a customer in a store such as a supermarket. The display operation unit 21 is an output device such as a display and also functions as an operation panel which is an input device of the POS register 20. Thus, in the present embodiment, it is assumed that the display operation unit 21 is configured as an integrated operation device (for example, a touch panel) that receives operations by an administrator or the like of the settlement processing system 1.

[0017] In addition, in this embodiment, the POS register 20 is communicably linked with the securities processing device 10 and the change dispenser 30, and it is assumed that the securities processing device 10 identifies securities such as commodity vouchers GT that a customer wishes to use, and determines whether the securities can be used at the store, and the POS register 20 and the change dispenser 30 perform settlement processing. Note that the securities processed by the settlement processing system 1 in the present invention are not limited to commodity vouchers GT, and may be valuable securities such as regional revitalization vouchers, or non-valuable securities such as service vouchers, coupon vouchers, and discount vouchers. Further, the settlement processing device is not limited to the change dispenser 30, and may be a settlement processing device other than the change dispenser 30.

[0018] FIG. 2 is a schematic diagram schematically showing the internal structure of the securities processing device 10. In FIG. 2, for ease of explanation, a number (any one of 1, 2, 3, and 4) is attached to the end of the reference numeral of the commodity voucher GT. However, the commodity vouchers GT1, GT2, GT3, and GT4 are also referred to as commodity vouchers GT without particular distinction. Further, in FIG. 2, the internal structure of the securities processing device 10 is described by taking as an example the case where the securities processing device 10 is installed on a plane (horizontal plane) perpendicular to the gravitational direction.

[0019] Hereinafter, for convenience of explanation, the X-axis, Y-axis, and Z-axis orthogonal to each other shown in FIG. 2 will be appropriately used for explanation. Further, the direction pointed by the arrow of the X-axis is referred to as the +X direction, and the opposite direction of the +X direction is referred to as the -X direction. Similarly, the direction pointed by the arrow of the Y-axis is referred to as the +Y direction, and the opposite direction of the +Y direction is referred to as the -Y direction. And the direction pointed by the arrow of the Z-axis is referred to as the +Z direction, and the opposite direction of the +Z direction is referred to as the -Z direction. For example, FIG. 2 schematically shows the internal structure of the securities processing device 10 when viewed from the +X direction to the -X direction. In this embodiment, the plane grasped when the securities processing device 10 is viewed from the +X direction to the -X direction corresponds to the front. Also, in this embodiment, the -Z direction corresponds to the gravitational direction. Hereinafter, the +Z direction may be referred to as upward, and the -Z direction may be referred to as downward.

[0020] The securities processing device 10 includes an input section 110 into which gift certificates GT are inserted, an image reading section 120 that reads an image IIMG of the gift certificates GT inserted from the input section 110, a temporary storage section 130 that temporarily stores the gift certificates GT after the image IIMG has been read, a storage section 140 that stores the gift certificates GT after settlement processing is complete, and a processing device 150. Furthermore, the securities processing device 10 includes a transport section 160 that transports the gift certificates GT inserted into the input section 110. The transport section 160 has, for example, a plurality of transport rollers ROL driven by a motor (not shown). Hereinafter, the rotation of the transport rollers ROL when transporting the gift certificates GT in the direction from the input section 110 to the temporary storage section 130 (from top to bottom in Figure 2) will be referred to as forward rotation, and the rotation in the opposite direction to forward rotation will be referred to as reverse rotation. Also, although not clearly shown in Figure 2, the transport path from the image reading section 120 to the storage section 140 is also part of the transport section 160.

[0021] For example, an employee or customer of a retail store where the securities processing device 10 is installed will insert a gift certificate GT1 into the input section 110. The input section 110 then inserts the inserted gift certificate GT1 into the securities processing device 10. In other words, the securities processing device 10 can take in the gift certificate GT1 inserted into the input section 110. If the identification process determines that the gift certificate GT1 inserted into the input section 110 is unacceptable, the gift certificate GT1 is returned from the input section 110 to the employee or customer. In other words, the input section 110 serves as both an input and output port for gift certificates GT.

[0022] The image reading unit 120 is, for example, an optical device such as a camera and an image scanner. In this embodiment, the image reading unit 120 has an end ED1 closer to the input unit 110 than the storage unit 140, and an end ED2 closer to the storage unit 140 than the input unit 110. For example, a gift certificate GT1 inserted from the input unit 110 passes through the end ED1 of the image reading unit 120 by a transport roller ROL. The image reading unit 120 then generates image information representing the image IIMG by reading one or both of the front and back surfaces of the gift certificate GT that has passed through the end ED1. Reading the image IIMG of the gift certificate GT is achieved, for example, by imaging the gift certificate GT. For example, the image reading unit 120 shown in Figure 2 can read the image IIMG of the front and back surfaces of the gift certificate GT by imaging the gift certificate GT from the +Y direction and the -Y direction, respectively.

[0023] Based on the image information generated by the image reading unit 120, an identification process for identifying the gift certificate GT is performed by the processing unit 150 (more specifically, the identification unit 154 shown in Figure 3), which will be described later. For example, a gift certificate GT inserted into the input unit 110 is the target of the identification process. Hereinafter, the gift certificate GT inserted into the input unit 110 may be referred to as the gift certificate GT to be identified.

[0024] In this embodiment, the transport unit 160 waits while the identification unit 154 determines whether or not to accept the gift certificate GT, holding the gift certificate GT so that it can move in reverse from the image reading unit 120 to the input unit 110. The state in which the gift certificate GT can move in reverse from the image reading unit 120 to the input unit 110 is, for example, the state before the gift certificate GT passes the transport roller ROL closest to the end ED2 of the image reading unit 120, among the multiple transport roller ROLs located within the image reading unit 120. For example, a portion of the gift certificate GT has passed the end ED2 of the image reading unit 120, and the area of ​​the gift certificate GT that has not passed the end ED2 is held by the transport roller ROL closest to the end ED2 of the gift certificate GT (state of gift certificate GT2 in Figure 2). That is, the transport unit 160 waits until the result of the identification process is known, holding the gift certificate GT so that it extends from the end ED2 toward the storage unit 140 (state of gift certificate GT2 in Figure 2).

[0025] If the result of the identification process is determined and the gift certificate GT2 is to be accepted, the transport roller ROL rotates forward, releasing the gift certificate GT2 from its position at the end ED2 of the image reading unit 120, and it is stored in the temporary storage unit 130. If the gift certificate GT2 is not to be accepted, the transport roller ROL rotates backward, causing the gift certificate GT2 to move in reverse from its position at the end ED2 of the image reading unit 120 and be discharged (rejected) from the input unit 110. Reverse movement of the gift certificate GT means, for example, transporting the gift certificate GT in the opposite direction to the transport direction when it is being transported into the securities processing unit 10, in the transport path for loading the gift certificate GT into the securities processing unit 10.

[0026] Thus, in this embodiment, the transport path from the input section 110 to the end ED2 of the image reading section 120 is also used as a transport path for discharging the gift certificate GT2 to the outside. Therefore, in this embodiment, it is not necessary to provide a separate transport path for discharging the gift certificate GT2 to the outside from the transport path for inputting the gift certificate GT into the securities processing device 10, thus preventing the transport path for the gift certificate GT from becoming complicated. As a result, in this embodiment, the configuration of the securities processing device 10 can be kept from becoming complicated. This prevents an increase in the cost of the securities processing device 10 in this embodiment.

[0027] Furthermore, in this embodiment, since unacceptable gift certificates GT are returned to the input unit 110, it is possible to prevent employees or customers from forgetting to take their gift certificates GT. Also, in this embodiment, since there is no need to provide a separate discharge port from the input unit 110, as described above, the transport path of gift certificates GT inside the securities processing device 10 is simplified, so it is possible to suppress the time it takes to remove unacceptable gift certificates GT from the machine. Moreover, in this embodiment, the gift certificates GT are held in a waiting state so that a portion of them is located between end ED1 and end ED2, so the time it takes to remove unacceptable gift certificates GT from the machine can be shortened. From the viewpoint of shortening the time it takes to remove unacceptable gift certificates GT from the machine, it is preferable that the image reading unit 120 be located close to the input unit 110.

[0028] The temporary storage unit 130 includes a temporary storage unit outlet 131 provided on the side of the temporary storage unit 130, an opening / closing plate 132, a support part 133 that rotatably supports the opening / closing plate 132, and a stopper 134. The temporary storage unit outlet 131 can be opened and closed, for example, using an axis along the direction of gravity (-Z direction) as the axis of rotation. The opening / closing plate 132 can be rotated around the support part 133 by a motor (not shown), and can be switched between an inclined state shown by a solid line and a vertical state shown by a dashed line in Figure 2.

[0029] The inclined state shown by the solid line in Figure 2 is the closed state, where the transport path from the temporary storage section 130 to the storage section 140 is closed, for example, when the opening / closing plate 132 is inclined with respect to the direction of gravity (-Z direction). On the inclined opening / closing plate 132, the gift certificate GT2 transported from the image reading section 120 is temporarily stored as gift certificate GT3. The stopper 134 is a stopper for securing the gift certificate GT3 on the inclined opening / closing plate 132. The vertical state shown by the dashed line in Figure 2 is the open state, where the transport path from the temporary storage section 130 to the storage section 140 is opened. When the opening / closing plate 132 transitions from the inclined state to the vertical state, the gift certificate GT3 temporarily stored on the opening / closing plate 132 is transported to the storage section 140.

[0030] For example, the gift certificate GT transported from the image reading unit 120 is temporarily stored on the inclined opening / closing plate 132. Once the settlement process by the POS register 20 is complete, the opening / closing plate 132 transitions from an inclined state to a vertical state, and the gift certificate GT is transported to the storage unit 140. If an employee or customer inserts multiple gift certificates GT, the multiple gift certificates GT are stored in the temporary storage unit 130, and once the settlement process is complete, all of the gift certificates GT are transported to the storage unit 140. Furthermore, if any problem occurs during the settlement process, the temporary storage unit outlet 131 can be opened, and the gift certificates GT GT stored in the temporary storage unit 130 can be removed from the securities processing device 10.

[0031] Thus, in this embodiment, a temporary storage unit 130 is provided in the middle of the transport path from the image reading unit 120 to the storage unit 140. Therefore, the temporary storage unit 130 can also be considered as part of the transport unit 160.

[0032] The storage unit 140 has a storage unit outlet 141 provided on its side and a collection box 142 for storing the gift certificates GT3 transported from the temporary storage unit 130 as gift certificates GT4. The storage unit outlet 141 can be opened and closed, for example, using an axis along the direction of gravity (-Z direction) as the axis of rotation. By opening the storage unit outlet 141, the collection box 142 can be removed from the securities processing device 10. That is, the gift certificates GT4 stored in the storage unit 140 can be removed from the securities processing device 10 through the storage unit outlet 141. For example, an employee or security officer can unlock the storage unit outlet 141 of the storage unit 140 and retrieve the gift certificates GT4 stored in the storage unit 140.

[0033] Furthermore, the present invention is not limited to a configuration in which, if the gift certificate GT2 is not accepted, the transport roller ROL rotates in reverse, causing the gift certificate GT2 to move in reverse from a state in which it is held to extend from the end ED2 toward the storage section 140, and to be discharged (rejected) to the outside from the input section 110. For example, the gift certificate GT may not be held to extend from the end ED2 toward the storage section 140, but all may be transported to the temporary storage section 130, and if the gift certificate GT2 is not accepted, the temporary storage section outlet 131 may be opened and the gift certificate GT2 may be removed to the outside. Alternatively, the temporary storage section 130 may not be provided, and the gift certificates may be transported directly from the image reading section 120 to the storage section 140. This makes it possible to miniaturize the securities processing device 10.

[0034] The processing unit 150 controls the identification process using image information indicating the image read by the image reading unit 120, and the overall operation of the securities processing unit 10.

[0035] Figure 3 is a block diagram showing the configuration of the securities processing device 10.

[0036] The securities processing device 10 includes, for example, the elements described in Figure 2 (the input unit 110, the image reading unit 120, the temporary storage unit 130, the storage unit 140, and the processing device 150), as well as a storage device 156 for storing various information and a communication device 170. The securities processing device 10 also has a plurality of sensors (not shown) corresponding to the image reading unit 120, the temporary storage unit 130, the storage unit 140, and the transport unit 160. For example, each sensor outputs a signal to the processing device 150 for detecting the state of the element corresponding to that sensor (any of the image reading unit 120, the temporary storage unit 130, the storage unit 140, and the transport unit 160).

[0037] The storage device 156 includes, for example, one or both of the following: a volatile memory such as RAM (Random Access Memory) that functions as a work area for the processing unit 150, and a non-volatile memory such as EEPROM (Electrically Erasable Programmable Read-Only Memory) that stores various information such as the control program PG1. The storage device 156 may be detachable from the securities processing unit 10. Specifically, the storage device 156 may be a storage medium such as a memory card that is detachable from the securities processing unit 10. Furthermore, the storage device 156 may be a storage device (e.g., online storage) that is communicably connected to the securities processing unit 10 via a network or the like.

[0038] In this embodiment, the storage device 156 stores not only the control program PG1 executed in the identification process, but also the ticket type table TBL which is referenced during the identification process. The ticket type table TBL will be explained in Figure 4, which will be described later.

[0039] The communication device 170 is hardware for communicating with external devices such as a POS register 20 located outside the securities processing device 10. For example, the communication device 170 can send and receive signals with the POS register 20 located outside the securities processing device 10. The communication device 170 is not limited to the POS register 20; it can also be connected to communicate with a change dispenser 30, a server device, a cloud device, and the like.

[0040] The processing unit 150 is a processor that controls the entire securities processing unit 10 and is configured, for example, to include one or more CPUs (Central Processing Units). For example, the processing unit 150 functions as a display control unit 151, an operation control unit 152, and an identification unit 154 by executing a control program PG1 stored in the storage device 156. The control program PG1 may be transmitted from another device via a network.

[0041] Based on the identification result by the securities processing device 10, the POS register 20 processes the settlement of the customer's purchased goods. For example, when an employee or customer settles using a gift certificate GT, the processing device 150 causes the image reading unit 120 to read the gift certificate GT (the gift certificate GT to be identified) inserted into the input unit 110. The image reading unit 120 generates image information showing the image IIMG (image IIMG of the gift certificate GT to be identified) of the read gift certificate GT and transmits the generated image information to the processing device 150. The image reading unit 120 may also store the image information in the storage device 156. In this case, the processing device 150 may read the image information from the storage device 156.

[0042] Furthermore, for example, if the processing unit 150 is configured to include multiple CPUs, some or all of the functions of the processing unit 150 may be realized by these multiple CPUs cooperating and operating according to a program such as a control program PG1. Also, the processing unit 150 may be configured to include hardware such as a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), or an FPGA (Field Programmable Gate Array) in addition to one or more CPUs, or in place of some or all of the one or more CPUs. In this case, some or all of the functions of the processing unit 150 may be realized by hardware such as a DSP.

[0043] Thus, in this embodiment, the control program PG1 includes an application program for causing the processing unit 150 to function as a display control unit 151, an operation control unit 152, and an identification unit 154. The control program PG1 may also include, for example, an operating system program for the processing unit 150 to control each part of the securities processing unit 10.

[0044] The display control unit 151 can control the display operation unit 21 of the POS register 20. For example, the display control unit 151 can display a confirmation screen SC, as shown in Figure 6 (described later), on the display operation unit 21.

[0045] The motion control unit 152 controls the operation of each of the input unit 110, image reading unit 120, temporary storage unit 130, storage unit 140, and transport unit 160 based on signals from sensors and the like that are provided corresponding to each element. For example, the motion control unit 152 controls the timing and speed of the transport operation of the transport unit 160, the opening and closing operation of the opening / closing plate 132 of the temporary storage unit 130, the opening and closing operation of the temporary storage unit outlet 131, and the opening and closing operation of the storage unit outlet 141. The transport operation of the transport unit 160 may include, for example, the operation of rotating the transport roller ROL forward or backward. In addition, the opening and closing operation of the opening / closing plate 132 of the temporary storage unit 130 may be included in the transport operation of the transport unit 160.

[0046] Furthermore, the identification unit 154 obtains image information from the image reading unit 120 that shows the image of the gift certificate GT (the gift certificate GT to be identified) inserted into the input unit 110. Then, the identification unit 154 performs an identification process to determine whether or not to accept the gift certificate GT to be identified based on the image information obtained from the image reading unit 120 (image information showing image IIMG of the gift certificate GT to be identified). Hereafter, the image information showing image IIMG of the gift certificate GT may be referred to as the image information of the gift certificate GT. Note that if the image reading unit 120 stores the image information of the gift certificate GT to be identified in the storage device 156, the identification unit 154 may obtain the image information of the gift certificate GT to be identified from the storage device 156. Image IIMG of the gift certificate GT to be identified is an example of the "first image," and the image information showing image IIMG of the gift certificate GT to be identified is an example of the "first image information."

[0047] The identification process performed by the identification unit 154 will be explained in Figure 8 below, but the identification process includes at least image processing and OCR (Optical Character Recognition) processing. Therefore, the identification unit 154 may be considered to include at least an image processing unit and an OCR processing unit.

[0048] The configuration of the settlement processing system 1 is not limited to the example shown in Figure 3. For example, the processing unit 150 and storage device 156 of the securities processing device 10 may be installed in a separate device from the main body of the securities processing device 10 and may be connected to the image reading unit 120 etc. via a network such as a LAN (Local Area Network). In this case, the device including the processing unit 150 and storage device 156 (a device installed in a separate device from the main body of the securities processing device 10) may function as a server for identifying gift certificates GT. For example, the image may be sent to an API (Application Programming Interface) on the cloud such as AWS (Amazon Web Services) to obtain the determination result. Note that AWS is a registered trademark. Amazon is a registered trademark.

[0049] Next, we will explain an example of a ticket type table (TBL) with reference to Figure 4.

[0050] Figure 4 is an explanatory diagram illustrating an example of the voucher type table TBL shown in Figure 3. The voucher type table TBL has, for example, one or more records. In the example shown in Figure 4, the voucher type table TBL has multiple records that correspond one-to-one with multiple types of gift certificates GT registered in the voucher type table TBL.

[0051] For example, each record in the voucher type table TBL stores voucher type information, including identification information, information indicating the voucher type name, and information indicating the face value. The reference image RIMG stores image information indicating the reference image RIMG and information regarding the feature portion FT. Furthermore, each record in the voucher type table TBL stores information indicating the position of the expiration date area NAR on the voucher GT, where the expiration date of the voucher GT is written, as the expiration date area NAR. The color analysis results for the image IIMG of the voucher GT are stored as the color analysis results.

[0052] The identification information is information used to identify each type of gift certificate GT from among several types. The certificate type name is the name of the type of gift certificate GT that corresponds to the identification information. For example, the certificate type name is displayed on the confirmation screen SC shown in Figure 6, which will be described later, so that employees of a retail store where a POS register 20 is installed can easily recognize the type of gift certificate GT inserted into the input section 110 of the securities processing device 10. The face value is the price of the gift certificate GT as a monetary voucher, as written on it.

[0053] Here, the types of gift certificates (GT) are distinguished by combinations of, for example, the design, face value, and issuer of the gift certificate. For example, three gift certificates (GT) corresponding to three different types of identification information, "001," "002," and "003," have the same design and issuer, but are stored as different types because their face values ​​are different. Similarly, three gift certificates (GT) corresponding to three different types of identification information, "003," "006," and "022," have the same face value, but are stored as different types because their design and issuer are different. Furthermore, even if the issuer and face value are the same, gift certificates (GT) with different designs are stored as different types.

[0054] Furthermore, the reference image RIMG is, for example, an image of the front of an acceptable gift certificate GT. The reference image RIMG is an example of a "second image," and the image information indicating the reference image RIMG is an example of "second image information." Note that the reference image RIMG may also be images of the front and back of an acceptable gift certificate GT. In addition, the storage location of the image information indicating the reference image RIMG may be stored in each record of the gift certificate type table TBL. The storage location of the image information is, for example, an address indicating the storage area where the image information is stored. Note that the image information may be stored in the storage device 156 where the gift certificate type table TBL is stored, or it may be stored in a storage device different from the storage device 156. Hereinafter, the image information indicating the reference image RIMG may be referred to as the image information of the reference image RIMG.

[0055] Furthermore, the reference image RIMG has feature portions FT corresponding to the type of gift certificate GT. It is preferable that multiple feature portions FT are defined for a single reference image RIMG. Feature portions FT are used, for example, in template matching performed in the identification process (more specifically, the gift certificate type identification process described in Figure 5). Template matching is a process that detects the location of an image that matches a template image within the image subject to template matching. "Match" includes not only exact matches but also cases where the images are similar enough to be considered a match. In this embodiment, the image IIMG of the gift certificate GT to be identified is the image subject to template matching, and the image of the feature portion FT corresponds to the template image. For example, examples of feature portion FT images include an image showing the issuing company of the gift certificate GT and an image showing the logo marks of sponsoring companies of the gift certificate GT. The image of the feature portion FT and its location differ depending on the type of gift certificate GT, for example. Therefore, the information regarding feature portions FT stored in each record of the gift certificate type table TBL includes, for example, image information showing the image of the feature portion FT and information showing the location of the feature portion FT in the gift certificate GT. Furthermore, the position coordinates of the feature portion FT may have a slight range to account for printing misalignment, etc. Additionally, the template images for each reference image RIMG can be automatically extracted and stored by the processing unit 150.

[0056] The location of the expiration date area NAR varies depending on the type of gift certificate GT. For example, in the gift certificate type table TBL shown in Figure 4, the expiration date area NAR for gift certificate GTs with identification information "001" to "006" is located in a portion of the back of the gift certificate GT, while the expiration date area NAR for gift certificate GTs with identification information "022" is located in a portion of the front of the gift certificate GT.

[0057] The color analysis results are, for example, the results of a color analysis of the surface of the gift certificate GT against the reference image RIMG. Color analysis involves, for example, analysis using the hue, lightness, and saturation of the image. Known color analysis methods can be employed.

[0058] Thus, in the voucher type table (TBL), the type of voucher GT, image, expiration date area (NAR), and color analysis results are associated with each other. Voucher GTs registered in the voucher type table (TBL) as acceptable voucher GTs are examples of "reference securities." Hereafter, voucher GTs registered in the voucher type table (TBL) may be referred to as reference voucher GTs.

[0059] Note that the configuration of the voucher type table TBL is not limited to the example shown in Figure 4. For example, the card company that issued the voucher GT may store information in each record of the voucher type table TBL indicating the location of the area where the voucher number (serial number) for managing the voucher GT is written. Also, for example, in the case of a voucher GT without an expiration date, information indicating that no expiration date is set may be stored instead of information indicating the location of the expiration date area NAR. Furthermore, for example, the color analysis results do not necessarily have to be stored in the voucher type table TBL.

[0060] The voucher type table TBL is referenced, for example, by the processing unit 150 of the securities processing unit 10 when the settlement processing system 1 performs settlement processing using vouchers GT. For example, the processing unit 150 functions as an identification unit 154 and performs voucher type identification processing to identify the voucher type of the voucher GT to be identified based on the image information of the reference image RIMG and the image information of the voucher GT to be identified, IIMG. The voucher type identification processing searches for a reference image RIMG that matches the image IIMG of the voucher GT to be identified from among multiple reference image RIMGs stored in the voucher type table TBL. Note that the voucher type identification processing is included in the identification processing to determine whether or not to accept the voucher GT to be identified. The program that performs the identification processing, including the voucher type identification processing, is included, for example, in the control program PG1 stored in the storage device 156. When the settlement processing system 1 performs settlement processing using vouchers GT, for example, the confirmation screen SC shown in Figure 6 is displayed on the display operation unit 21 of the POS register 20.

[0061] Next, we will explain the overview of the ticket type identification process, referring to Figure 5.

[0062] Figure 5 is an explanatory diagram illustrating the overview of the ticket type identification process.

[0063] The image IIMGf shown in Figure 5 is the image of the surface of the gift certificate GT to be identified, read by the image reading unit 120, and used as the image IIMG of the gift certificate GT to be identified in the gift certificate type identification process. For example, image IIMGf is an image represented by the image information generated by the image reading unit 120. In this embodiment, we assume that the image IIMGf of the gift certificate GT to be identified is used in the gift certificate type identification process. In the example shown in Figure 5, the gift certificate GT to be identified has some scribbles NZ1 and a stamp impression NZ2.

[0064] Furthermore, the reference images RIMG1, RIMG2, and RIMG3 shown in Figure 5 are reference images RIMG that are searched using AI (Artificial Intelligence) as candidates for reference image RIMG that match image IIMGf. In the example shown in Figure 5, in order to facilitate understanding of the voucher type identification process, two feature parts FT are predetermined for each of the reference images RIMG1, RIMG2, and RIMG3, and it is assumed that image IIMGf has two feature parts FT. The images of the two feature parts FT (template images) are, for example, an image showing the company that issued the gift certificate GT, and an image showing the logo marks of the sponsoring companies of the gift certificate GT.

[0065] In Figure 5, for the sake of clarity, a two-character alphanumeric code (either ia, ib, 1a, 1b, 2a, 2b, 3a, or 3b) is appended to the end of the feature portion FT code. However, below, feature portions FTia, FTib, FT1a, FT1b, FT2a, FT2b, FT3a, and FT3b will also be referred to simply as feature portions FT without any particular distinction. Note that feature portions FTia and FT1b are not pre-registered as feature portions FT; rather, images that match the reference image RIMG through the template processing described later are conveniently treated as feature portions. Similarly, below, reference images RIMG1, RIMG2, and RIMG3 will also be referred to simply as reference images RIMG without any particular distinction. Furthermore, below, a candidate reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified may be referred to as a matching candidate reference image RIMG. A matching candidate reference image RIMG is an example of a "candidate image".

[0066] In the ticket type identification process, first, a first process is performed that uses AI to search for a reference image RIMG of potential matches. Then, a second process is performed to identify the type of the gift certificate GT to be identified based on the degree of match between the multiple feature portion FTs of the reference image RIMG and the image IIMGf. In the second process, template matching is performed using the image of the feature portion FT as a template image. The degree of match is an index that represents the degree of match between the feature portion FTs of the reference image RIMG and the feature portion FTs of the image IIMGf. For example, the degree of match is calculated based on how much of the feature corresponding portion of the image IIMGf corresponds to the feature portion FTs of the reference image RIMG. For example, the degree of match is calculated based on whether or not there is an image in the image IIMGf that can be considered to match the feature portion FTs of the reference image RIMG, and if so, whether or not the position of that image can be considered to be equal to the position of the feature portion FTs of the reference image RIMG. In other words, the ticket type identification process includes a first process using AI and a second process using template matching.

[0067] First, let me explain the first process using AI.

[0068] In the first process, the identification unit 154 generates a feature vector for image IIMGf by performing a process to generate a feature vector representing the features of the image on the image IIMGf of the gift certificate GT to be identified. The process of generating the feature vector includes, for example, multiple convolutional layers and pooling layers in a convolutional neural network. For example, the identification unit 154 generates the feature vector for image IIMGf using an AI model that has been trained (metric learning) to reduce the distance between feature vectors of multiple images of the same gift certificate type and increase the distance between feature vectors of multiple images of different gift certificate types. Then, the identification unit 154 calculates the similarity between image IIMGf and each reference image RIMG based on the inter-vector distance, which is the distance between the feature vector of image IIMGf and the feature vector of each reference image RIMG. The feature information representing the feature vector of the reference image RIMG is stored in the gift certificate type table TBL, for example, associated with the reference image RIMG.

[0069] Thus, in this embodiment, an AI model generated by distance learning specifically for gift certificates GT and coupons is used. However, while the above-mentioned AI model is very effective for gift certificates GT and coupons, which have many types and many similar tickets, the AI ​​model in the present invention is not limited to the AI ​​model generated by distance learning described above, and it is possible to use a generally known AI model.

[0070] For example, the similarity calculated using AI is calculated as a numerical value such as 0.000 to 1.000. For example, the closer the similarity of two images is to 1.000, the more similar the two images are to each other. If the images are exactly the same, the similarity will be 1.000. However, in cases where there are many types and many similar images, such as with gift certificates (GT), it is often difficult to accurately determine a single reference image (RIMG) that matches the image (IIMGf) of the gift certificate (GT) to be identified, based solely on the similarity calculated by AI. For example, gift certificates (GT) often have scribbles, stains, and wrinkles, and if the AI ​​accuracy is increased too much, the detection accuracy decreases. In other words, there are many cases where the highest similarity score calculated by AI does not correspond to the correct gift certificate (GT). Therefore, in this embodiment, all reference images (RIMG) whose similarity to the image (IIMGf) of the gift certificate (GT) to be identified is above a certain threshold value are extracted as candidate reference images (RIMG). Furthermore, if there is no reference image RIMG whose similarity to the image IIMGf of the gift certificate GT to be identified is equal to or greater than the threshold value, it may be determined that there is no reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified. In this embodiment, it is assumed that the threshold value is 0.900.

[0071] In the example shown in Figure 5, reference images RIMG1, RIMG2, and RIMG3, whose similarity to the image IIMGf of the gift certificate GT to be identified is 0.9 or higher, are searched from the gift certificate type table TBL as candidate reference images RIMG. For example, the similarity between image IIMGf and reference image RIMG1 is 0.941, the similarity between image IIMGf and reference image RIMG2 is 0.935, and the similarity between image IIMGf and reference image RIMG3 is 0.925. From the candidate reference images RIMG1, RIMG2, and RIMG3, the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified is identified by the second process. Note that if there are no reference images RIMG with a similarity of 0.9 or higher, the identification unit 154 may determine that there are no reference images RIMG that match the image IIMGf of the gift certificate GT to be identified without performing the second process. In other words, if there is no reference image RIMG with a similarity score equal to or greater than the standard value, the identification unit 154 may determine that the type of gift certificate GT to be identified is unknown.

[0072] Next, we will explain the second process, which utilizes template matching.

[0073] In the second process, the identification unit 154 determines whether the image IIMGf of the gift certificate GT to be identified has a feature corresponding portion corresponding to the feature portion FT of each of the candidate reference images RIMG1, RIMG2, and RIMG3. Specifically, for example, the identification unit 154 performs template matching to detect the location in the image IIMGf of the gift certificate GT to be identified where an image matching the image of the feature portion FT (template image) exists. The identification unit 154 then calculates the degree of matching for each of the candidate reference images RIMG1, RIMG2, and RIMG3 with respect to the image IIMGf of the feature portion FT, based on the results of the template matching. The identification unit 154 also identifies the type of gift certificate GT to be identified based on the degree of matching for each of the candidate reference images RIMG1, RIMG2, and RIMG3.

[0074] For example, the matching candidate reference image RIMG1 has a feature portion FT1a that shows the logo of a sponsoring company and a feature portion FT1b that shows the issuing company. That is, the image of feature portion FT1a is an image that shows the logo of a sponsoring company, and the image of feature portion FT1b is an image that shows the issuing company. Furthermore, the region of feature portion FT1a is the region enclosed by four vertices P11, P12, P13 and P14, and the region of feature portion FT1b is the region enclosed by four vertices P15, P16, P17 and P18.

[0075] The identification unit 154 uses the image of feature portion FT1a of the reference image RIMG1 as a template image and performs template matching on image IIMGf. This identifies the position of the image of feature portion FTia that matches the image of feature portion FT1a. The region of feature portion FTia is the region enclosed by four vertices P1, P2, P3, and P4. Figure 5 assumes that the position coordinates of the region enclosed by the four vertices P1, P2, P3, and P4 (i.e., the position of feature portion FTia) coincide with the position coordinates of the region enclosed by the four vertices P11, P12, P13, and P14 (i.e., the position of feature portion FT1a). Therefore, the identification unit 154 identifies the feature portion FTia of image IIMGf as the feature corresponding portion to the feature portion FT1a of the reference image RIMG1.

[0076] Furthermore, the identification unit 154 uses the image of feature portion FT1b of the reference image RIMG1 as a template image to perform template matching on image IIMGf. In this case, since there is no portion in image IIMGf that matches the image of feature portion FT1b, the identification unit 154 identifies that there is no feature corresponding portion in image IIMGf that corresponds to feature portion FT1b of the reference image RIMG1. Thus, in the reference image RIMG1, a feature corresponding portion that matches one of the two feature portions FT is detected in image IIMGf. In this case, for example, in a method that calculates the degree of match of the reference image RIMG1 based on the number of feature portions FT that correspond to the feature corresponding portion detected in image IIMGf, the degree of match of the reference image RIMG1 will be 1.

[0077] The method for calculating the degree of matching of reference image RIMG1 is not limited to the method described above. For example, the degree of matching of reference image RIMG1 may be calculated based on the matching ratio of feature segments FT. In the example shown in Figure 5, the number of feature segments FT in reference image RIMG1 is two, and the number of feature segments FT corresponding to the feature-matching portion detected from image IIMGf is one. In this case, using the method of calculating the degree of matching of reference image RIMG1 based on the matching ratio of feature segments FT, the degree of matching of reference image RIMG1 would be 0.5 (=1 / 2). Alternatively, the degree of matching of reference image RIMG of matching candidates such as reference image RIMG1 may be quantified using AI or the like.

[0078] Furthermore, for example, the matching candidate reference image RIMG2 has a feature portion FT2a that shows the logo of a sponsoring company and a feature portion FT2b that shows the issuing company. That is, the image of feature portion FT2a is an image that shows the logo of a sponsoring company, and the image of feature portion FT2b is an image that shows the issuing company. Also, the region of feature portion FT2a is the region enclosed by four vertices P21, P22, P23 and P24, and the region of feature portion FT2b is the region enclosed by four vertices P25, P26, P27 and P28.

[0079] The identification unit 154 uses the image of feature portion FT2a of the reference image RIMG2 as a template image and performs template matching on image IIMGf. This identifies the position of the image of feature portion FTia that matches the image of feature portion FT2a. Figure 5 assumes a case where the position coordinates of the region enclosed by the four vertices P1, P2, P3, and P4 (i.e., the position of feature portion FTia) match the position coordinates of the region enclosed by the four vertices P21, P22, P23, and P24 (i.e., the position of feature portion FT2a). Therefore, the identification unit 154 identifies the feature portion FTia of image IIMGf as the feature corresponding portion to the feature portion FT2a of the reference image RIMG2.

[0080] Furthermore, the identification unit 154 uses the image of feature portion FT2b of the reference image RIMG2 as a template image and performs template matching on image IIMGf. This identifies the position coordinates of the image of feature portion FTib that matches the image of feature portion FT2b. The region of feature portion FTib is the region enclosed by four vertices P5, P6, P7, and P8. Figure 5 assumes that the position coordinates of the region enclosed by the four vertices P5, P6, P7, and P8 (i.e., the position of feature portion FTib) match the position coordinates of the region enclosed by the four vertices P25, P26, P27, and P28 (i.e., the position of feature portion FT2b). Therefore, the identification unit 154 identifies the feature portion FTib of image IIMGf as the feature corresponding portion to feature portion FT2b of the reference image RIMG2. In this way, two feature corresponding portions that match the two feature portions FT in the reference image RIMG2 are detected from image IIMGf. In this case, for example, if the degree of matching of the reference image RIMG2 is calculated based on the number of feature segments FT corresponding to the feature-matching portion detected in image IIMGf, the degree of matching of the reference image RIMG2 will be 2. However, if the degree of matching of the reference image RIMG2 is calculated based on the matching ratio of feature segments FT, the degree of matching of the reference image RIMG2 will be 1 (=2 / 2).

[0081] Furthermore, for example, the matching candidate reference image RIMG3 has a feature portion FT3a that shows the logo of a sponsoring company and a feature portion FT3b that shows the issuing company. That is, the image of feature portion FT3a is an image that shows the logo of a sponsoring company, and the image of feature portion FT3b is an image that shows the issuing company. Also, the region of feature portion FT3a is the region enclosed by four vertices P31, P32, P33 and P34, and the region of feature portion FT3b is the region enclosed by four vertices P35, P36, P37 and P38.

[0082] The identification unit 154 uses the image of feature portion FT3a of the reference image RIMG3 as a template image and performs template matching on image IIMGf. This identifies the position coordinates of the image of feature portion FTia that matches the image of feature portion FT3a. Figure 5 assumes a case where the position coordinates of the region enclosed by the four vertices P1, P2, P3 and P4 (i.e., the position of feature portion FTia) are different from the position coordinates of the region enclosed by the four vertices P31, P32, P33 and P34 (i.e., the position of feature portion FT3a). Therefore, the identification unit 154 identifies that there is no feature corresponding to feature portion FT3a of the reference image RIMG3 in image IIMGf.

[0083] Furthermore, the identification unit 154 uses the image of feature portion FT3b of the reference image RIMG3 as a template image and performs template matching on image IIMGf. This identifies the position coordinates of the image of feature portion FTib that matches the image of feature portion FT3b. Figure 5 assumes a case where the position coordinates of the region enclosed by the four vertices P5, P6, P7, and P8 (i.e., the position of feature portion FTib) match the position coordinates of the region enclosed by the four vertices P35, P36, P37, and P38 (i.e., the position of feature portion FT3b). Therefore, the identification unit 154 identifies the feature portion FTib of image IIMGf as the feature corresponding portion corresponding to feature portion FT3b of the reference image RIMG3. In this way, in the reference image RIMG3, a feature corresponding portion that matches one of the two feature portions FT is detected from image IIMGf. In this case, for example, in a method that calculates the degree of match of the reference image RIMG3 based on the number of feature portions FT corresponding to the feature corresponding portion detected from image IIMGf, the degree of match of the reference image RIMG3 will be 1. Furthermore, in the method of calculating the degree of matching of the reference image RIMG3 based on the matching ratio of the feature portion FT, the degree of matching of the reference image RIMG3 is 0.5 (=1 / 2).

[0084] The identification unit 154 identifies reference image RIMG2, which has the highest degree of match calculated based on the feature portion FT, as the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified, from among the candidate reference images RIMG1, RIMG2, and RIMG3. Generally, the likelihood that a candidate reference image RIMG matches image IIMGf increases as the degree of match between the feature portion FT of the reference image RIMG and image IIMGf increases. Furthermore, the accuracy of the degree of match improves when multiple feature portion FTs are defined for a reference image RIMG compared to when only one feature portion FT is defined for a reference image RIMG. For this reason, it is preferable to define multiple feature portion FTs for a single reference image RIMG.

[0085] Here, for example, in a proportional approach where the second processing is not performed, the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified is identified based solely on the similarity calculated using AI. Therefore, in the example shown in Figure 5, in a proportional approach, the reference image RIMG1, which has the highest similarity, is incorrectly identified as the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified. In contrast, in this embodiment, in addition to the similarity calculated using AI, the degree of match with the image IIMGf of the feature portion FT is also considered. Therefore, in this embodiment, it is possible to suppress the incorrect identification of the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified.

[0086] Thus, in this embodiment, even with gift certificates such as GT, which have many types and many similar certificates, the type of gift certificate can be identified with high accuracy. Furthermore, in this embodiment, the processing that is black-boxed by AI is used in the first process to calculate the similarity of the matching candidate reference image RIMG. The final identification of the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified is performed in the second process, which has a clear processing algorithm. For this reason, in this embodiment, even if it becomes necessary to change the judgment criteria for the gift certificate type identification process due to some factor, the judgment criteria can be easily changed.

[0087] Note that the ticket type identification process is not limited to the example shown in Figure 5. For example, if there is no matching candidate reference image RIMG with a matching degree of a predetermined value or higher, the identification unit 154 may determine that there is no reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified. In other words, if there is no matching candidate reference image RIMG with a matching degree of a predetermined value or higher, the identification unit 154 may determine that the type of gift certificate GT to be identified is unknown.

[0088] Furthermore, for example, if there are multiple reference image RIMGs with the same degree of similarity among multiple candidate reference image RIMGs, the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified may be identified as follows. For example, the identification unit 154 may identify the reference image RIMG with the highest similarity among multiple reference image RIMGs with the same degree of similarity as the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified. Alternatively, if the degree of similarity is calculated based on the number of feature portion FTs corresponding to the feature corresponding portion detected from the image IIMGf, the identification unit 154 may identify the reference image RIMG with the highest matching ratio of feature portion FTs among multiple reference image RIMGs with the same degree of similarity as the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified.

[0089] Furthermore, in calculating the degree of match, the method for determining the number of feature segment FTs corresponding to the feature-matching portion detected from image IIMGf is not limited to the example described above (where both the image and position coordinates of the feature segment FTs match). For example, in image IIMGf, if there is an image that matches the image of a feature segment FT, but the position coordinates of the image that matches the image of the feature segment FT are different from the position coordinates of the feature segment FT, the number of feature segment FTs corresponding to the feature-matching portion detected from image IIMGf may be set to 0.5. In other words, the degree of match may be quantified for each feature segment FT.

[0090] Furthermore, the second process is not limited to template matching; any process is acceptable as long as it can identify one reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified from the candidate reference image RIMGs. Specifically, for example, in the second process, instead of detecting the position coordinates of an image that matches the image of the feature portion FT, it may be determined whether the image at the position of the feature portion FT in the reference image RIMG in image IIMGf matches the image of the feature portion FT. Alternatively, in the second process, a single reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified may be identified by checking whether a predetermined number of symbols or numbers are written in a specific area using character recognition processing such as OCR processing.

[0091] Furthermore, while Figure 5 assumes that the number of feature segments (FTs) defined in the reference image RIMG is the same (2) in multiple candidate reference image RIMGs, the number of feature segments (FTs) may differ in multiple reference image RIMGs. If the number of feature segments (FTs) differs in multiple candidate reference image RIMGs, the degree of match is preferably calculated based, for example, on the matching ratio of the feature segments (FTs).

[0092] Next, referring to Figure 6, we will explain the confirmation screen SC for checking the identification result of the gift certificate GT inserted into the input section 110 of the securities processing device 10.

[0093] Figure 6 is an explanatory diagram illustrating an example of a confirmation screen SC for confirming the identification result of a gift certificate GT inserted into the input section 110 of the securities processing device 10. The confirmation screen SC shown in Figure 6 is displayed on the display operation unit 21 of the POS register 20 when a store employee confirms the gift certificate GT that the customer wishes to use, when the POS register 20 is a regular counter-type register or a semi-self-service register. If the POS register 20 is a self-service register, the confirmation screen SC is not displayed on the display operation unit 21. The display operation unit 21 displays the confirmation screen SC in response to control from the processing device 150 (more specifically, the display control unit 151) when the securities processing device 10 performs the operation shown in Figure 7, which will be described later. The confirmation screen SC shown in Figure 6 displays the identification result of one gift certificate GT.

[0094] The confirmation screen SC displays, for example, the reading order, ticket type name, face value, reference image RIMG, read image, and multiple buttons BT for the GUI (Graphical User Interface).

[0095] The reading order is, for example, the reading order in which the image reading unit 120 reads multiple gift certificates GT to be identified. The certificate type name is the name of the type of gift certificate GT to be identified. For example, among the multiple certificate types registered in the certificate type table TBL, the certificate type indicated by the reference image RIMG that is determined to match the image IIMGf of the gift certificate GT to be identified is identified as the certificate type name of the gift certificate GT to be identified.

[0096] The face value is the face value of the gift certificate GT being identified. For example, among the face values ​​of multiple gift certificate types registered in the gift certificate type table TBL, the face value of the same gift certificate type as the gift certificate GT being identified is identified as the face value of the gift certificate GT being identified. The reference image RIMG is the image of the front of the gift certificate GT corresponding to the same gift certificate type as the gift certificate GT being identified, among the multiple gift certificate types registered in the gift certificate type table TBL. In other words, the reference image RIMG displayed on the confirmation screen SC is the reference image RIMG that has been determined to match the image IIMGf of the gift certificate GT being identified, among the multiple reference images RIMG registered in the gift certificate type table TBL.

[0097] The image to be read is the image IIMG of the gift certificate GT to be identified (the image IIMG of the gift certificate GT read by the image reading unit 120). In the example shown in Figure 6, the image IIMGf of the front of the gift certificate GT to be identified and the image IIMGb of the back of the gift certificate GT to be identified are displayed on the confirmation screen SC. Hereafter, images IIMGf and IIMGb will be referred to as image IIMG without any particular distinction.

[0098] Store employees who identify gift certificates GT can easily confirm whether the gift certificate GT has been correctly identified by checking the confirmation screen SC. For example, if an employee confirms on the confirmation screen SC that the gift certificate GT has been correctly identified, they may press the "Start Settlement" button BT on the confirmation screen SC to have the settlement system 1 execute the settlement process using the gift certificate GT. For example, if an employee has another gift certificate GT to insert into the input section 110 of the securities processing device 10 (a second gift certificate GT in the example shown in Figure 6), they may press the "Next" button BT on the confirmation screen SC. Also, for example, if an employee wants to cancel the settlement process currently being processed, they may press the "Cancel" button BT on the confirmation screen SC. If the settlement process currently being processed is canceled, for example, the gift certificate GT with reading order "1" is stored in the temporary storage section 130, and can be retrieved by unlocking the temporary storage section outlet 131 of the temporary storage section 130.

[0099] The confirmation screen SC is not limited to the example shown in Figure 6. For example, if the gift certificate GT to be identified is determined to be unacceptable, only the reading order and the read image may be displayed on the confirmation screen SC, out of the reading order, certificate type name, face value, reference image RIMG, and read image. Also, if the gift certificate GT to be identified is determined to be unacceptable, the reason for unacceptability (e.g., unknown certificate type name, expired, or counterfeit) may be displayed in the certificate type name field. Furthermore, for example, a button BT for continuing the settlement process using a gift certificate GT other than the one determined to be unacceptable may be displayed on the confirmation screen SC. The gift certificate GT determined to be unacceptable is held at the end ED2 of the image reading unit 120, then moves in the reverse direction towards the input unit 110, and is discharged from the input unit 110 to the outside.

[0100] Next, with reference to Figure 7, we will describe the general operation of the securities processing device 10 when an employee or customer processes a settlement for purchased goods using gift certificates GT.

[0101] Figure 7 is a flowchart showing an example of the operation of the securities processing device 10 when performing settlement processing using gift certificates GT. The process in step S100 is executed, for example, when an employee or customer inserts gift certificates GT into the input unit 110. In this embodiment, when settlement processing for purchased goods is performed using multiple gift certificates GT, it is assumed that the gift certificates GT are inserted into the input unit 110 one by one.

[0102] First, in step S100, the processing unit 150 functions as an operation control unit 152 and controls the transport unit 160 to feed the gift certificate GT inserted into the input unit 110 into the securities processing unit 10 (more specifically, the image reading unit 120). As a result, the gift certificate GT inserted into the input unit 110 passes through the image reading unit 120 in step S120.

[0103] In step S120, the image reading unit 120 reads the image IIMG of the gift certificate GT as it passes through the image reading unit 120. As a result, the image information of the gift certificate GT is generated by the image reading unit 120.

[0104] Next, in step S140, the processing unit 150 functions as an identification unit 154 and determines whether or not it can accept the gift certificate GT inserted into the input unit 110. The process in step S140 is the identification process of the gift certificate GT. As will be described in detail in Figure 8, in this embodiment, the identification process of the gift certificate GT (i.e., the process in step S140) includes multiple identification processes. Until the result of the determination by the process in step S140 (the result of the identification process) is known, the gift certificate GT is maintained in a state where a part of the gift certificate GT has passed the end ED2 of the image reading unit 120, as explained in Figure 2.

[0105] If the result of the determination in step S140 is negative, that is, if it is determined that the gift certificate GT cannot be accepted, the processing device 150 functions as the operation control unit 152 in step S162 and returns the gift certificate GT to the input unit 110. For example, in step S162, the operation control unit 152 reverses the transport roller ROL to transport the gift certificate GT held at the end ED2 of the image reading unit 120 to the input unit 110 and discharges it to the outside from the input unit 110. After executing the process in step S162, the processing device 150 proceeds to step S180.

[0106] On the other hand, if the result of the determination in step S140 is positive, that is, if it is determined that the gift certificate GT can be accepted, the processing device 150 proceeds to step S160.

[0107] In step S160, the processing unit 150 functions as an operation control unit 152 and controls the transport unit 160 to transport the gift certificate GT held at the end ED2 of the image reading unit 120 to the temporary storage unit 130. For example, the operation control unit 152 rotates the transport roller ROL in the forward direction to release the gift certificate GT held at the end ED2 of the image reading unit 120 and transport it to the temporary storage unit 130. As a result, the gift certificate GT determined to be acceptable is stored in the temporary storage unit 130. After executing the process in step S160, the processing unit 150 proceeds to step S180.

[0108] In step S180, the processing unit 150 functions as an operation control unit 152 and determines whether or not there is another gift certificate GT (a gift certificate GT to be inserted into the input unit 110).

[0109] For example, if the POS register 20 is a self-checkout register, the processing unit 150 may function as a display control unit 151 and display a selection screen on the display operation unit 21 of the POS register 20 to select whether or not there is a gift certificate GT to be inserted next. If it is selected via the selection screen that there is a gift certificate GT to be inserted next, the operation control unit 152 may determine that there is another gift certificate GT. In this case, the display control unit 151 may display an input screen on the display operation unit 21 instructing the insertion of the next gift certificate GT. Alternatively, if it is selected via the selection screen that there is no gift certificate GT to be inserted next, the operation control unit 152 may determine that there is no next gift certificate GT. In this case, the display control unit 151 may display an input screen on the display operation unit 21 instructing the execution of the settlement process.

[0110] Furthermore, for example, if the POS register 20 is a regular counter-type register or a semi-self-service register, the confirmation screen SC is displayed on the display operation unit 21 of the POS register 20, as explained in Figure 6. The operation control unit 152 may, for example, determine that there is a next gift certificate GT when the "Next" button BT on the confirmation screen SC is pressed, and determine that there is no next gift certificate GT when the "Start Payment" button BT on the confirmation screen SC is pressed.

[0111] If the result of the determination in step S180 is positive, that is, if there is another gift certificate GT to be inserted, the processing unit 150 returns to step S100. For example, if an employee or customer inserts the next gift certificate GT into the input unit 110, the series of processes from step S100 to step S140 are executed on the gift certificate GT inserted into the input unit 110 (the next gift certificate GT). Then, the series of processes in steps S160 and S180, or the series of processes in steps S162 and S180 are executed.

[0112] In this manner, when an employee or customer inserts n gift certificates GT consecutively (where n is a natural number greater than or equal to 2), after the mth gift certificate GT (where m is a natural number greater than or equal to 1 and less than n) is determined to be acceptable, in step S180, the presence or absence of the next gift certificate GT to be inserted is selected, and the (m+1)th gift certificate GT is inserted into the insertion unit 110. For example, if the first to third gift certificates GT were accepted, but the fourth gift certificate GT was determined to be unacceptable, a different gift certificate GT from the one returned (the fourth gift certificate GT) may be inserted into the insertion unit 110.

[0113] On the other hand, if the result of the determination in step S180 is negative, that is, if there are no more gift certificates GT to insert, the processing unit 150 proceeds to step S200. Also, if the result of the determination in step S180 is negative, the employee or customer performs the settlement process via the POS register 20. If the POS register 20 is a self-checkout register, as described above, an input screen indicating that the settlement process will be performed may be displayed on the display operation unit 21 of the POS register 20. Here, for example, if the first to third gift certificates GT were accepted, but the fourth gift certificate GT was determined to be unacceptable, the settlement process using the first to third gift certificates GT may be performed.

[0114] In step S200, the processing unit 150 functions as an operation control unit 152 and determines whether the settlement process has been completed. For example, if an employee or customer has completed the settlement process, the operation control unit 152 receives a notification from the POS register 20 via the communication device 170 indicating that the settlement process has been completed. Therefore, if the operation control unit 152 receives a notification from the POS register 20 via the communication device 170 indicating that the settlement process has been completed, it determines that the settlement process has been completed and proceeds to step S220. On the other hand, if the operation control unit 152 has not received a notification from the POS register 20 via the communication device 170 indicating that the settlement process has been completed, it determines that the settlement process has not been completed and returns to step S200.

[0115] In this way, if the result of the determination in step S200 is positive, the operation control unit 152 proceeds to step S220, and if the result of the determination in step S200 is negative, it returns to step S200. In other words, the operation control unit 152 waits to execute the process in step S220 until the settlement process is completed.

[0116] In step S220, the processing unit 150 functions as an operation control unit 152 and controls the opening / closing plate 132 of the temporary storage unit 130 to transport the gift certificates GT stored in the temporary storage unit 130 to the storage unit 140. For example, the operation control unit 152 transitions the state of the opening / closing plate 132 of the temporary storage unit 130 from an inclined state to a vertical state, thereby opening the transport path from the temporary storage unit 130 to the storage unit 140 and transporting the gift certificates GT stored on the opening / closing plate 132 to the storage unit 140. As a result, the gift certificates GT (gift certificates GT used in the settlement process) temporarily stored in the temporary storage unit 130 are stored in the storage unit 140.

[0117] The operation shown in Figure 7 is completed when the process in step S220 is executed.

[0118] Note that the operation of the securities processing device 10 is not limited to the example shown in Figure 7. For example, when the process in step S162 is executed (when the gift certificate GT is returned), it may be possible to choose whether to continue or cancel the settlement process using the gift certificate GT. For example, if the first to third gift certificates GT were accepted, but the fourth gift certificate GT was determined to be unacceptable, it is conceivable that the user would wish to cancel the use of the first to third gift certificates GT. If the user chooses to cancel the use of the first to third gift certificates GT, the first to third gift certificates GT are stored in the temporary storage unit 130 and can be retrieved by unlocking the temporary storage unit outlet 131 of the temporary storage unit 130.

[0119] Alternatively, for example, the processing device 150 may terminate the operation shown in Figure 7 after executing the process in step S162. In this case as well, the gift certificates GT stored in the temporary storage unit 130 can be retrieved by unlocking the temporary storage unit outlet 131 of the temporary storage unit 130.

[0120] Next, an example of the identification process will be explained with reference to Figure 8.

[0121] Figure 8 is a flowchart showing an example of the identification process. The identification process shown in Figure 8 is executed as the process in step S140 of Figure 7 after the process in step S120 of Figure 7 has been executed. Therefore, the series of processes (identification process) from step S141 to step S148 is executed by the processing device 150, which functions as the identification unit 154.

[0122] First, in step S141, the identification unit 154 determines whether a reference image RIMG (image IIMG of an acceptable gift certificate GT) that matches the image of the gift certificate GT (gift certificate GT to be identified) inserted into the input unit 110 is stored in the gift certificate type table TBL. For example, the identification unit 154 compares each of the multiple reference image RIMGs stored in the gift certificate type table TBL with the image IIMG of the gift certificate GT to be identified. Then, based on the comparison result, the identification unit 154 determines whether there is a reference image RIMG among the multiple reference image RIMGs stored in the gift certificate type table TBL that matches the image IIMG of the gift certificate GT to be identified. In addition, in the gift certificate type determination process in step S141, the image reading unit 120 may obtain image information of both sides of the gift certificate GT and perform a determination of whether the gift certificate GT is front or back.

[0123] Furthermore, "match" as described above includes not only cases of exact matching, but also cases where images are similar enough to be considered identical. For example, if the similarity between two images is above a certain threshold, the two images may be considered identical. Known methods can be used to calculate the similarity between two images.

[0124] Step S141 is a voucher type identification process that identifies the type of voucher GT to be identified. The voucher type identification process is explained in Figure 5, so a detailed explanation is omitted here. However, the flow of the voucher type identification process (step S141) will be described later in Figure 9.

[0125] If the result of the determination in step S141 is negative, that is, if it is determined that there is no reference image RIMG that matches the image IIMG of the gift certificate GT to be identified, the identification unit 154 determines that the gift certificate GT cannot be accepted and proceeds to step S162 in Figure 7. Therefore, if there is no reference image RIMG that matches the image IIMG of the gift certificate GT to be identified, the gift certificate GT held at the end ED2 of the image reading unit 120 is returned to the input unit 110 in step S162.

[0126] On the other hand, if the result of the determination in step S141 is positive, that is, if it is determined that there is a reference image RIMG that matches the image IIMG of the gift certificate GT to be identified, the identification unit 154 proceeds to step S146.

[0127] In step S146, the identification unit 154 performs character recognition processing, such as OCR processing, on the image IIMG of the gift certificate GT to be identified. For example, the identification unit 154 performs OCR processing on the expiration date area NAR within the image IIMG of the gift certificate GT to be identified. The information indicating the location of the expiration date area NAR within the image IIMG of the gift certificate GT is stored in the gift certificate type table TBL, as described in Figure 4, and is associated with the type of gift certificate GT. For example, the identification unit 154 refers to the gift certificate type table TBL and identifies the location of the expiration date area NAR of the gift certificate type corresponding to the reference image RIMG that matches the image IIMG of the gift certificate GT to be identified as the location of the expiration date area NAR within the image IIMG of the gift certificate GT to be identified. Then, the identification unit 154 extracts the characters contained in the expiration date area NAR within the image IIMG of the gift certificate GT by performing known OCR processing on the expiration date area NAR identified by referring to the gift certificate type table TBL within the image IIMG of the gift certificate GT to be identified. For example, the identification unit 154 extracts the string printed in the expiration date area NAR of the gift certificate GT to be identified as text data. This identifies the expiration date of the gift certificate GT to be identified.

[0128] The type of data extracted from the expiration date area NAR can be determined by the matching data used to compare the extracted data with the NAR data. For example, if the data is date data such as an expiration date, it is sufficient to match the numbers with "year", "month", "day", " / ", and ".".

[0129] After executing the process in step S146, the identification unit 154 proceeds to step S147.

[0130] In step S147, the identification unit 154 determines whether the gift certificate GT to be identified is a gift certificate GT that is within its validity period. For example, the identification unit 154 determines whether the gift certificate GT to be identified is a gift certificate GT that is within its validity period by comparing the date of the expiration date (the expiration date of the gift certificate GT to be identified) identified in the character recognition process in step S146 with today's date on which the settlement process is executed. The identification unit 154 may also determine that the gift certificate GT to be identified is not a gift certificate GT that is within its validity period if the expiration date is not printed in the expiration date area NAR identified by referring to the gift certificate type table TBL in the image IIMG of the gift certificate GT to be identified. In other words, the identification unit 154 may determine that the gift certificate GT to be identified is not a gift certificate GT that is within its validity period if the expiration date is not identified in the character recognition process in step S146.

[0131] If the result of the determination in step S147 is negative, that is, if it is determined that the gift certificate GT to be identified is not a gift certificate GT within its validity period, the identification unit 154 determines that the gift certificate GT cannot be accepted and proceeds to step S162 in Figure 7. Therefore, if the gift certificate GT to be identified is not a gift certificate GT within its validity period, the gift certificate GT held at the end ED2 of the image reading unit 120 is returned to the input unit 110 in step S162.

[0132] On the other hand, if the result of the determination in step S147 is affirmative, that is, if the gift certificate GT to be identified is determined to be a gift certificate GT within its validity period, the identification unit 154 proceeds to step S148. In this way, the characters printed in the validity period area NAR of the gift certificate GT are identified through the series of processes in steps S146 and S147. Hereafter, the series of processes in steps S146 and S147 will also be referred to as the character information identification process. The character information identification process (the series of processes in steps S146 and S147) corresponds to one of several identification processes.

[0133] In step S148, the identification unit 154 determines whether the gift certificate GT to be identified is a genuine gift certificate. A genuine gift certificate is a legitimate gift certificate GT that is not counterfeit. Hereafter, gift certificates GT that have been counterfeited by copying, etc., may be referred to as counterfeit. For example, the identification unit 154 performs a counterfeit identification process to determine whether the gift certificate GT to be identified is a counterfeit (a counterfeit gift certificate GT). That is, in step S148, the identification unit 154 determines that a gift certificate GT that has been determined not to be counterfeit by the counterfeit identification process is a genuine gift certificate.

[0134] For example, the identification unit 154 performs color analysis on the image IIMG of the gift certificate GT to be identified and determines whether the gift certificate GT is counterfeit or not. Here, for example, the color analysis results of each reference image RIMG are stored in the gift certificate type table TBL, corresponding to the type of gift certificate GT, as explained in Figure 4. For example, the identification unit 154 obtains the color analysis result of the reference image RIMG that matches the image IIMG of the gift certificate GT to be identified from the gift certificate type table TBL. By comparing the color analysis result of the reference image RIMG that matches the image IIMG of the gift certificate GT to be identified with the color analysis result of the image IIMG of the gift certificate GT to be identified, the identification unit 154 can determine whether the gift certificate GT to be identified is counterfeit or not. For example, by performing color analysis using the hue, brightness, and saturation of the image, it is possible to distinguish between simple fading and discoloration due to copying. Furthermore, by using AI, more accurate analysis is possible.

[0135] Furthermore, in the counterfeit identification process, in addition to judgment based on color, it is also possible to determine whether the target gift certificate GT is counterfeit by comparing a specific part within the image IIMG of the gift certificate GT and identifying the degradation of image resolution due to copying.

[0136] If the result of the determination in step S148 is negative, that is, if the gift certificate GT to be identified is determined not to be a genuine gift certificate, the identification unit 154 determines that the gift certificate GT cannot be accepted and proceeds to step S162 in Figure 7. Therefore, if the gift certificate GT to be identified is determined not to be a genuine gift certificate, that is, if the gift certificate GT to be identified is determined to be a counterfeit gift certificate, the gift certificate GT held at the end ED2 of the image reading unit 120 is returned to the input unit 110 in step S162.

[0137] On the other hand, if the result of the determination in step S148 is positive, that is, if the gift certificate GT to be identified is determined to be genuine, the identification unit 154 proceeds to step S160 in Figure 7. Therefore, if the gift certificate GT to be identified is determined to be genuine, that is, if the gift certificate GT to be identified is determined not to be counterfeit, the gift certificate GT held at the end ED2 of the image reading unit 120 is transported to the temporary storage unit 130 in step S160.

[0138] Here, the series of processes in steps S146 and S147 (character information identification process) and the process in step S148 (counterfeit identification process) are validity verification processes that confirm the validity of the gift certificate GT identified in the process in step S141 (certificate type identification process). Note that the validity verification process is not limited to the character information identification process and the counterfeit identification process. For example, as a validity verification process, a barcode process may be performed in which the barcode containing the encrypted JAN (Japanese Article Number) code printed on the gift certificate GT is read, and the validity of the gift certificate GT identified in the certificate type identification process is confirmed based on the information read from the barcode. For example, in the barcode process, if the certificate type identified based on the JAN code matches the certificate type of the gift certificate GT identified in the certificate type identification process, the gift certificate GT identified in the certificate type identification process is determined to be valid. The barcode process may be performed in addition to the character information identification process and the counterfeit identification process, or in place of one or both of the character information identification process and the counterfeit identification process.

[0139] As described above, in the identification process shown in Figure 8, multiple identification processes (voucher type identification process, character information identification process, and counterfeit voucher identification process) determine whether or not the voucher GT to be identified is acceptable, thus preventing an unacceptable voucher GT from being mistakenly identified as acceptable. Furthermore, in the identification process shown in Figure 8, each of the multiple identification processes is executed sequentially, and as soon as it is determined that the voucher GT is unacceptable, the process proceeds to step S162 in Figure 7. In other words, when the identification process shown in Figure 8 is executed, each identification process is executed sequentially, and as soon as it is determined that the voucher GT is unacceptable, the operation control unit 152 causes the transport unit 160 to perform a transport operation to move the voucher GT in reverse and transport it to the input unit 110, thereby discharging the voucher GT from the input unit 110 to the outside. For this reason, in this embodiment, compared to a configuration in which an unacceptable voucher GT is returned after all of the multiple identification processes have been completed, it is possible to suppress the time it takes to remove an unacceptable voucher GT from the machine.

[0140] Furthermore, in this embodiment, in addition to the ticket type identification process, character information identification and counterfeit ticket identification processes are also performed, so it is possible to identify expired gift certificates GT and counterfeit tickets. In other words, in this embodiment, for example, the probability that a gift certificate GT that cannot be used at a retail store where a POS register 20 is installed will be deemed acceptable can be reduced. As a result, in this embodiment, the number of times staff are called to a retail store such as a supermarket can be reduced, and the customer turnover rate can be improved. In addition, in this embodiment, since the probability that a gift certificate GT that cannot be used at a retail store will be deemed acceptable is reduced, it is also effective in reducing customer stress.

[0141] Note that the identification process is not limited to the example shown in Figure 8. For example, at least two of the processes in step S141, the series of processes in steps S146 and S147, and the process in step S148 may be executed in parallel. Specifically, for example, the series of processes in steps S146 and S147 and the process in step S148 may be executed in parallel. Executing multiple processes in parallel includes not only switching between processes but also executing multiple processes simultaneously. Note that, for example, the series of processes in steps S146 and S147 may be executed after the process in step S148.

[0142] Furthermore, for example, at least a part of the processing in step S148 (for example, color analysis of the image IIMG of the gift certificate GT to be identified) may be executed in parallel with the processing in step S141. Also, for example, the processing in step S141, the series of processing in steps S146 and S147, and the processing in step S148 may be executed in parallel. In this case, for example, the character recognition processing (for example, OCR processing) that extracts text data indicating the expiration date from the image IIMG of the gift certificate GT may be performed on the entire image IIMG of the gift certificate GT.

[0143] Even in the configuration where multiple identification processes are executed in parallel, when any of the multiple identification processes determines that the gift certificate GT is unacceptable, the operation control unit 152 causes the transport unit 160 to execute a transport operation to move the gift certificate GT in reverse and transport it to the input unit 110, thereby ejecting the gift certificate GT from the input unit 110 to the outside. Therefore, even in the configuration where multiple identification processes are executed in parallel, it is possible to suppress the time required to remove the unacceptable gift certificate GT from the machine. Furthermore, in the configuration where multiple identification processes are executed in parallel, if it is determined that none of the multiple identification processes have determined that the gift certificate GT is unacceptable, in step S160 of Figure 7, the gift certificate GT held at the end ED2 of the image reading unit 120 is transported to the temporary storage unit 130.

[0144] Furthermore, Figure 8 illustrates a case where three identification processes are performed as part of the identification process: the process in step S141 (voucher type identification process), the series of processes in steps S146 and S147 (character information identification process), and the process in step S148 (counterfeit voucher identification process). However, the present invention is not limited to this embodiment. For example, in addition to the three identification processes described above (voucher type identification process, character information identification process, and counterfeit voucher identification process), one or more of the following may be performed: an identification process based on the degree of notch and scale of the voucher GT identified by image processing, an identification process based on the thickness of the voucher GT, and an identification process using magnetism. Alternatively, one or more of the following may be performed instead of one or both of the character information identification process and the counterfeit voucher identification process: an identification process based on the degree of notch and scale of the voucher GT identified by image processing, an identification process based on the thickness of the voucher GT, and an identification process using magnetism. Alternatively, only the voucher type identification process may be performed from the three identification processes described above, or only two identification processes, including the voucher type identification process, may be performed.

[0145] Next, an example of ticket type identification processing will be explained with reference to Figure 9.

[0146] Figure 9 is a flowchart showing an example of the ticket type identification process. The ticket type identification process shown in Figure 9 is executed as the process in step S141 of Figure 8, after the process in step S120 of Figure 7 has been executed. Therefore, the series of processes from step S142 to step S145 (ticket type identification process) is executed by the processing device 150, which functions as the identification unit 154. The process in step S142 corresponds to the first process described in Figure 5, and the series of processes from step S143 to step S145 corresponds to the second process described in Figure 5.

[0147] First, in step S142, the identification unit 154 determines whether a reference image RIMG of a matching candidate whose similarity to the image of the gift certificate GT (gift certificate GT to be identified) inserted into the input unit 110 is equal to or greater than a certain threshold is stored in the gift certificate table TBL. For example, the identification unit 154 uses AI to calculate the similarity between each of the multiple reference image RIMGs stored in the gift certificate table TBL and the image IIMG of the gift certificate GT to be identified. Then, based on the similarity calculated using AI, the identification unit 154 determines whether there is a reference image RIMG of a matching candidate among the multiple reference image RIMGs stored in the gift certificate table TBL.

[0148] If the result of the determination in step S142 is negative, that is, if it is determined that there is no matching candidate reference image RIMG whose similarity to the image IIMG of the gift certificate GT to be identified is equal to or greater than the standard value, the identification unit 154 determines that the gift certificate GT cannot be accepted and proceeds to step S162 in Figure 7. Therefore, if there is no reference image RIMG that matches the image IIMG of the gift certificate GT to be identified, the gift certificate GT held at the end ED2 of the image reading unit 120 is returned to the input unit 110 in step S162.

[0149] On the other hand, if the result of the determination in step S141 is positive, that is, if it is determined that there is a matching candidate reference image RIMG whose similarity to the image IIMG of the gift certificate GT to be identified is equal to or greater than the standard value, the identification unit 154 proceeds to step S143.

[0150] In step S143, the identification unit 154 reads information about the feature portion FT (information showing the image and position coordinates of the feature portion FT) of the candidate reference image RIMG from the ticket type table TBL. If multiple candidate reference image RIMGs are identified as candidate reference image RIMGs, information about the feature portion FT of all candidate reference image RIMGs is read from the ticket type table TBL.

[0151] Next, in step S144, the identification unit 154 calculates the degree of matching between the feature portion FT of the candidate reference image RIMG and the image IIMG of the gift certificate GT to be identified, for each candidate reference image RIMG. For example, as explained in Figure 5, the identification unit 154 calculates the degree of matching between the feature portion FT of the candidate reference image RIMG by performing template matching using the image of the feature portion FT as a template image.

[0152] Next, in step S145, the identification unit 154 identifies the reference image RIMG with the highest matching degree calculated in step S144 as the reference image RIMG that matches the image IIMG of the gift certificate GT to be identified. After executing the process in step S145, the identification unit 154 proceeds to step S146 in Figure 8.

[0153] Note that the ticket type identification process is not limited to the example shown in Figure 9. For example, if there are no matching candidate reference image RIMGs with a matching degree of a predetermined value or higher calculated in step S144, the identification unit 154 may determine that there are no reference image RIMGs that match the image IIMGf of the gift certificate GT to be identified, and proceed to step S162 in Figure 7. In other words, if there are no matching candidate reference image RIMGs with a matching degree of a predetermined value or higher, the identification unit 154 may determine that the gift certificate GT to be identified is unacceptable.

[0154] In this embodiment, the settlement processing system 1 includes at least a POS register 20, a change dispenser 30, and a security processing device 10. The security processing device 10 determines whether or not to accept the inserted gift certificate GT. The security processing device 10 includes an input unit 110 into which the gift certificate GT is inserted, a transport unit 160 for transporting the gift certificate GT inserted into the input unit 110, an image reading unit 120 for generating first image information showing image IIMG, which is an image of the gift certificate GT, an identification unit 154 for determining whether or not to accept the gift certificate GT by identifying the gift certificate GT based on the first image information, an operation control unit 152 for controlling the transport operation of the gift certificate GT by the transport unit 160 according to the determination result of the identification unit 154, and a storage unit 140 for storing the acceptable gift certificate GT. The identification unit 154 performs an identification process that includes a gift certificate type identification process to identify the gift certificate type of the gift certificate GT based on second image information and first image information, which represent reference image RIMGs, each of the images of a pre-registered set of reference gift certificate GTs. Based on the result of the identification process, the unit identifies the gift certificate GT. The gift certificate type identification process includes a first process that uses AI to calculate the similarity between image IIMG and reference image RIMG to search for candidate images (hereinafter referred to as matching candidate reference image RIMGs) from among a set of reference image RIMGs corresponding to multiple reference gift certificate GTs, where the similarity with image IIMG is equal to or greater than a standard value, as candidates for the reference gift certificate GT corresponding to the gift certificate GT; and a second process that identifies the gift certificate type of the gift certificate GT based on the degree of agreement between the matching candidate reference image RIMG and image IIMG, which is predetermined for one or more feature parts FTs.

[0155] Furthermore, if the identification unit 154 determines that the gift certificate GT can be accepted, the operation control unit 152 causes the transport unit 160 to perform a transport operation to transport the gift certificate GT to the storage unit 140, thereby storing the gift certificate GT in the storage unit 140. If the identification unit 154 determines that the gift certificate GT cannot be accepted, the operation control unit 152 controls the transport operation by the transport unit 160 to discharge the gift certificate GT to the outside. If the identification unit 154 determines that the gift certificate GT can be accepted, the POS register 20 and the change dispenser 30 perform the settlement process for the products that the customer wishes to purchase, according to the identification result of the identification unit 154.

[0156] Thus, in this embodiment, the AI ​​processing is used in the first process to calculate the similarity of candidate reference images RIMG. Then, in the second process, the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified is identified based on the degree of match of one or more feature portion FTs defined in the reference image RIMG to the image IIMG. For example, gift certificates GT often have scribbles, stains, and wrinkles, and if the AI ​​accuracy is increased too much, the detection accuracy will decrease. Therefore, there are cases in which the highest similarity score calculated by the AI ​​does not correspond to the correct gift certificate GT. In this embodiment, the final identification of the reference image RIMG that matches the image IIMGf of the gift certificate GT to be identified is performed by the second process, which uses the degree of match of the feature portion FTs of the reference image RIMG to the image IIMG. As a result, in this embodiment, even with gift certificates GT that have many types and many similar certificates, the type of gift certificate GT can be identified with high accuracy. Furthermore, the second process can utilize known image processing with a clear processing algorithm (e.g., template matching). Therefore, in this embodiment, the identification accuracy of gift certificates (GT) can be improved in a simpler way compared to adjusting the judgment results of processing by an AI that is a black box.

[0157] Furthermore, in this embodiment, each of the multiple reference image RIMGs may have a predetermined number of feature portion FTs. The second process may include a process to determine whether the image IIMG has a feature corresponding portion in each of the multiple feature portion FTs of the candidate reference image RIMG, and a process to calculate the degree of matching of the candidate reference image RIMG based on the number of feature portion FTs corresponding to the feature corresponding portion detected from the image IIMG. Thus, in this embodiment, the degree of matching of the candidate reference image RIMG is calculated based on the number of feature portion FTs corresponding to the feature corresponding portion detected from the image IIMG. For this reason, in this embodiment, the degree of matching of the candidate reference image RIMG can be easily calculated.

[0158] Furthermore, in this embodiment, the second process may include a process to determine whether image IIMG has a feature corresponding portion corresponding to each of the multiple feature portion FTs of the candidate reference image RIMG, and a process to calculate the degree of matching of the candidate reference image RIMG based on the number of feature portion FTs corresponding to the feature corresponding portion detected from image IIMG relative to the number of feature portion FTs of the candidate reference image RIMG. Thus, in this embodiment, the degree of matching of the candidate reference image RIMG is calculated based on the matching ratio of feature portion FTs. For this reason, in this embodiment, the degree of matching of the candidate reference image RIMG can be calculated simply. Even if the number of feature portion FTs differs among multiple reference image RIMGs, the scale of the degree of matching of the candidate reference image RIMG can be unified across the multiple reference image RIMGs. As a result, in this embodiment, even if the number of feature portion FTs differs among multiple reference image RIMGs, the type of gift certificate GT can be identified with high accuracy.

[0159] Furthermore, in this embodiment, the identification process may include a character recognition process that recognizes characters contained in the image IIMG, a character information identification process that determines whether or not to accept the gift certificate GT based on the information identified by the multiple characters recognized by the character recognition process, and a counterfeit certificate identification process that determines whether or not the gift certificate GT is counterfeit based on the first image information. Thus, in this embodiment, in addition to the gift certificate type identification process, a character information identification process and a counterfeit certificate identification process are performed, so it is possible to identify invalid gift certificates GT (for example, gift certificates GT that have expired or counterfeit certificates). For this reason, in this embodiment, for example, the probability that a gift certificate GT that cannot be used at a retail store where a POS register 20 is installed will be determined to be acceptable can be reduced. As a result, in this embodiment, the number of times staff are called at retail stores such as supermarkets can be reduced, and the customer turnover rate can be improved. In addition, in this embodiment, since the probability that a gift certificate GT that cannot be used at a retail store will be determined to be acceptable is reduced, it is also effective in reducing customer stress.

[0160] [2. Variant] The present invention is not limited to the embodiments illustrated above. Specific variations are illustrated below. Two or more embodiments arbitrarily selected from the following examples may be combined.

[0161] [First variation] In the embodiment described above, for gift certificates GT that do not have an expiration date set, the series of processes (character information identification process) in steps S146 and S147 shown in Figure 8 may not be performed. For example, among the multiple gift certificates GT stored in the gift certificate table TBL, for gift certificates GT that do not have an expiration date set, information indicating that no expiration date is set may be stored instead of information indicating the position of the gift certificate GT in the expiration date area NAR. In this case, for example, the identification unit 154 can determine whether or not an expiration date is set for the gift certificate GT to be identified by referring to the field in the expiration date area NAR that corresponds to the reference image RIMG that matches the image IIMG of the gift certificate GT to be identified in the gift certificate table TBL. Thus, the same effects as in the embodiment described above can be obtained in this modified example as well. Furthermore, in this modified example, settlement processing using gift certificates GT that do not have an expiration date set can be easily performed.

[0162] [Second variation] In the embodiments described above, the input section 110 is provided on the upper surface of the securities processing device 10 (the surface that is grasped when viewing the securities processing device 10 from the +Z direction), but the present invention is not limited to this embodiment. For example, in the embodiments and modifications described above, the input section 110 may be provided on the front surface of the securities processing device 10 (the surface that is grasped when viewing the securities processing device 10 from the +X direction). In this modification as well, the same effects as in the embodiments and modifications described above can be obtained.

[0163] [Third variation] In the embodiments described above, the input section 110 is shown as serving as both an input port for inserting gift certificates GT and an output port for discharging the gift certificates GT inserted into the securities processing device 10. However, the present invention is not limited to this embodiment. For example, in the embodiments and modifications described above, the output port for discharging the gift certificates GT inserted into the securities processing device 10 may be provided separately from the input section 110. In this modified form as well, the same effects as in the embodiments and modifications described above can be obtained.

[0164] [Fourth variation] In the embodiments and modifications described above, multiple reference image RIMGs for multiple gift certificate GTs may be associated with a single gift certificate type information. In this case, for example, the expiration date area NAR and color analysis results are stored in the gift certificate type table TBL, associated with each of the multiple reference image RIMGs for multiple gift certificate GTs. In this modification, even if there are multiple gift certificate GTs with different designs for the same gift certificate type due to design changes, for example, it is possible to identify each gift certificate GT by type while suppressing the complexity of managing the reference gift certificate GTs. Thus, the same effects as in the embodiments and modifications described above can be obtained in this modification as well. [Explanation of Symbols]

[0165] 1...Settlement processing system, 10...Securities processing device, 20...POS register, 21...Display operation unit, 30...Change dispenser, 110...Input unit, 120...Image reading unit, 130...Temporary storage unit, 140...Storage unit, 150...Processing device, 151...Display control unit, 152...Operation control unit, 154...Identification unit, 156...Storage device, 160...Transport unit, 170...Communication device, GT...Gift certificate.

Claims

1. In a securities processing system that determines whether or not to accept the securities that have been submitted, The input section into which the aforementioned securities are inserted, A transport unit for transporting the securities that have been inserted into the input unit, An image reading unit that generates first image information showing a first image which is an image of the security, An identification unit that identifies the security based on the first image information and determines whether or not to accept the security, An operation control unit controls the transport operation of the securities by the transport unit according to the determination result of the identification unit, A storage compartment for storing the aforementioned securities that can be accepted, Equipped with, The aforementioned identification unit is An identification process is performed which includes a security type identification process that identifies the type of security based on a second image information which is an image of each of a plurality of pre-registered reference securities and the first image information, and the security is identified based on the result of the identification process. The aforementioned ticket type identification process is: A first process involves using Artificial Intelligence (AI) to calculate the similarity between the first image and the second image, and searching among a plurality of second images corresponding to the plurality of reference securities for a candidate second image representing the reference securities corresponding to the securities, wherein the similarity to the first image is equal to or greater than a threshold value. A second process for identifying the type of security based on the degree of agreement between one or more predetermined feature portions of the candidate image and the first image, including, A securities processing device characterized by the following:

2. Each of the aforementioned plurality of second images has a predetermined set of feature parts. The second process described above is: A process to determine whether the first image has a feature corresponding portion in each of the plurality of feature portions of the candidate image, A process to calculate the degree of match of the candidate image based on the number of feature portions corresponding to the feature portions detected from the first image, including, The securities processing apparatus according to claim 1.

3. Each of the aforementioned plurality of second images has a predetermined set of feature parts. The second process described above is: A process to determine whether the first image has a feature corresponding portion in each of the plurality of feature portions of the candidate image, A process to calculate the degree of match of the candidate image based on the number of feature portions corresponding to the feature-corresponding portions detected from the first image relative to the number of feature portions of the candidate image, including, The securities processing apparatus according to claim 1.

4. The aforementioned identification process is, A character recognition process is performed to recognize characters contained in the first image, and a character information identification process is performed to determine whether or not to accept the securities based on the information identified by the multiple characters recognized by the character recognition process. A counterfeit certificate identification process that determines whether the security is counterfeit or not based on the first image information, including, A securities processing apparatus according to any one of claims 1 to 3.

5. A settlement processing system having at least a POS register, a settlement processing device, and a securities processing device, The securities processing device is The input section into which securities are inserted, A transport unit for transporting the securities that have been inserted into the input unit, An image reading unit that generates first image information showing a first image which is an image of the security, An identification unit that identifies the security based on the first image information and determines whether or not to accept the security, An operation control unit controls the transport operation of the securities by the transport unit according to the determination result of the identification unit, A storage compartment for storing the aforementioned securities that can be accepted, Equipped with, The aforementioned identification unit is An identification process is performed which includes a security type identification process that identifies the type of security based on a second image information which is an image of each of a plurality of pre-registered reference securities and the first image information, and the security is identified based on the result of the identification process. The aforementioned ticket type identification process is: A first process involves using AI to calculate the similarity between the first image and the second image to search for candidate images from among a plurality of second images corresponding to the plurality of reference securities, where the similarity to the first image is equal to or greater than a threshold value, as candidate second images representing the reference securities corresponding to the securities. A second process for identifying the type of security based on the degree of agreement between one or more predetermined feature portions of the candidate image and the first image, Includes, The aforementioned operation control unit, If the identification unit determines that the security can be accepted, the transport unit is instructed to perform the transport operation to transport the security to the storage unit, thereby storing the security in the storage unit. If the identification unit determines that the security cannot be accepted, it controls the transport operation by the transport unit to discharge the security to the outside. The POS register and the settlement processing device are If the identification unit determines that the security is acceptable, it will execute the settlement process for the product the customer wishes to purchase, according to the identification result in the identification unit. A settlement processing system characterized by the following:

6. Each of the aforementioned plurality of second images has a predetermined set of feature parts. The second process described above is: A process to determine whether the first image has a feature corresponding portion in each of the plurality of feature portions of the candidate image, A process to calculate the degree of match of the candidate image based on the number of feature portions corresponding to the feature portions detected from the first image, including, The settlement processing system according to feature 5.

7. Each of the aforementioned plurality of second images has a predetermined set of feature parts. The second process described above is: A process to determine whether the first image has a feature corresponding portion in each of the plurality of feature portions of the candidate image, A process to calculate the degree of match of the candidate image based on the number of feature portions corresponding to the feature-corresponding portions detected from the first image relative to the number of feature portions of the candidate image, including, The settlement processing system according to feature 5.

8. The aforementioned identification process is, A character recognition process is performed to recognize characters contained in the first image, and a character information identification process is performed to determine whether or not to accept the securities based on the information identified by the multiple characters recognized by the character recognition process. A counterfeit certificate identification process that determines whether the security is counterfeit or not based on the first image information, including, The settlement processing system according to any one of claims 5 to 7.

9. A securities processing method executed by a securities processing apparatus comprising: an input unit into which securities are inserted; a transport unit for transporting the securities inserted into the input unit; an image reading unit for generating first image information showing a first image which is an image of the securities; an identification unit for determining whether or not to accept the securities by identifying the securities based on the first image information; an operation control unit for controlling the transport operation of the securities by the transport unit according to the determination result of the identification unit; and a storage unit for storing the securities that are acceptable, wherein the securities processing apparatus is performed by the securities processing apparatus, The aforementioned identification unit is An identification process is performed which includes a security type identification process that identifies the type of security based on a second image information which is an image of each of a plurality of pre-registered reference securities and the first image information, and the security is identified based on the result of the identification process. The aforementioned ticket type identification process is: A first process involves using AI to calculate the similarity between the first image and the second image to search for candidate images from among a plurality of second images corresponding to the plurality of reference securities, where the similarity to the first image is equal to or greater than a threshold value, as candidate second images representing the reference securities corresponding to the securities. A second process for identifying the type of security based on the degree of agreement between one or more predetermined feature portions of the candidate image and the first image, including, A method for processing securities, characterized by the following:

Images