accounting system

Abstract

To provide an accounting system that can both improve the accuracy of commodity identification and reduce costs for automatic accounting processing without depending on the type of commodities to be handled.SOLUTION: An accounting system 1 includes: a first commodity identification unit 12 that acquires price information and the number of first commodities 31 specified based on image information of a commodity group 30 captured by a camera 3; and a second commodity identification unit 13 that acquires price information and the number of second commodities 32 to which an RFID tag 33 is affixed based on information read by an RFID reader 4. The commodity group 30 includes the first commodities 31 and the second commodities 32 that are classified based on a predetermined criterion. The accounting system 1 includes an accounting processing unit 15 that calculates the total price of the first commodities 31 and the second commodities 32 based on the price information and the number of commodities acquired by the first commodity identification unit 12 and the second commodity identification unit 13.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to accounting systems. [Background technology]

[0002] In recent years, various methods have been proposed to improve the efficiency of checkout procedures when purchasing goods at stores and the like.

[0003] For example, Patent Document 1 discloses a configuration that automatically reads information from RFID tags attached to each of a group of products placed in a shopping cart, and presents the customer with a list of the products and the total price based on the read information.

[0004] Furthermore, Patent Document 2 discloses a configuration in which a camera installed in a store acquires image information of the group of products that a customer has placed in a shopping cart, identifies the group of products to be paid for based on the image information, and automatically processes the payment. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2018-190255 [Patent Document 2] Japanese Patent Application Publication No. 2019-086832 Summary of the Invention [Problem to be solved by the invention]

[0006] The method using RFID tags as described in Patent Document 1 has the problem that the unit price of RFID tags is relatively high, resulting in high costs and poor cost-effectiveness when managing inexpensive products such as food. On the other hand, the method using image processing as described in Patent Document 2 may result in a situation where the accuracy of product identification decreases when managing products that are easily deformed, such as clothing. As such, the suitability of the method using RFID tags and the method using image processing differs depending on the type of product being handled.

[0007] The present disclosure aims to provide an accounting system that can achieve both improved accuracy in identifying products in automatic accounting processes and cost reduction, regardless of the type of products handled. [Means for solving the problem]

[0008] A checkout system according to one aspect of an embodiment of the present invention comprises an imaging unit that captures images of a group of products moving relatively within an imaging area, and an RFID reader that reads information from RFID tags attached to the group of products moving relatively within a reception range, the group of products including first products and second products that are classified based on predetermined criteria, a first product identification unit that acquires item count and price information of the first product identified based on image information of the group of products captured by the imaging unit, a second product identification unit that acquires item count and price information of the second product to which the RFID tag is attached based on the information read by the RFID reader, a checkout processing unit that calculates the total price of the first product and the second product based on the item count and price information acquired by the first product identification unit and the second product identification unit, and a product counting unit that counts the total number of items of the group of products.

[0009] According to this aspect, for example, for inexpensive products (first products) such as food, product identification is performed using image processing, making it possible to manage products without using RFID tags, which have a relatively high unit price, thereby reducing product management costs. Furthermore, for products (second products) that are easily deformed, such as clothing, product identification using image processing may result in a decrease in product identification accuracy, but product identification using RFID tags improves identification accuracy. In this way, the accounting system of this aspect selectively applies a method suited to the product management suitability of the products handled, thereby achieving both improved product identification accuracy in automated accounting processing and cost reduction, regardless of the type of products handled.

[0010] An accounting system according to another aspect of an embodiment of the present invention may include a product counting unit that counts the total number of items in the group of products, and an error detection unit that detects an error when the number of items obtained by the first product identification unit and the second product identification unit differs from the total number of items counted by the product counting unit.

[0011] As described above, when two product identification methods are used in combination in checkout processing, problems such as duplicate detection of the same product can occur. According to this aspect, by comparing the number of products identified by image processing, the number of products identified by RFID tags, and the total number counted by the product counting unit and detecting an error if there is a discrepancy in the numbers, it is possible to prevent problems from occurring when two identification methods are used in combination and further improve the accuracy of product identification.

[0012] In a checkout system relating to another aspect of an embodiment of the present invention, the error detection unit may perform at least some of the following: detecting overlap of the group of products that are moving relative to each other; detecting double reading of the same item by the first product identification unit and the second product identification unit; or detecting fraudulent behavior by a purchaser of the group of products.

[0013] According to this aspect, it becomes possible to detect various errors that may occur in automated checkout processing, thereby improving the accuracy and reliability of automated checkout processing.

[0014] A checkout system according to another aspect of the present invention may include a transport unit that transports the group of products, and the imaging unit and the RFID reader may read information from the group of products transported by the transport unit.

[0015] According to this embodiment, customers simply place the items they wish to purchase on the conveying unit, which automatically passes the items through the imaging area of ​​the imaging unit and the reception range of the RFID reader, allowing the checkout process to be carried out automatically, significantly reducing the burden on customers.

[0016] An accounting system according to another aspect of an embodiment of the present invention comprises a product information storage unit that stores individual weight information for the group of products, a weighing unit that measures the weight of the group of products being transported on the transport unit, and an error detection unit, wherein the error detection unit acquires the weight information from the product information storage unit for the group of products read by the first product identification unit or the second product identification unit, and may detect an error of multiple products overlapping if the acquired weight information differs from the weight measured by the weighing unit.

[0017] According to this aspect, even in a situation where the first product is hidden under the second product and cannot be identified by the imaging unit, an overlap error can be detected based on the change in weight of the identified product, thereby further reducing false detection of products and further improving the accuracy of product identification.

[0018] A checkout system according to another aspect of an embodiment of the present invention includes a product information storage unit that stores individual height information for the group of products, a height measurement unit that measures the height of the group of products being transported on the transport unit, and an error detection unit, wherein the error detection unit acquires the height information from the product information storage unit for the group of products read by the first product identification unit or the second product identification unit, and may detect an error of multiple products overlapping if the acquired height information differs from the height measured by the height measurement unit.

[0019] According to this aspect, even in a situation where the first product is hidden under the second product and cannot be identified by the imaging unit, an overlap error can be detected based on the change in height of the identified product, thereby further reducing false detection of products and further improving the accuracy of product identification.

[0020] A checkout system according to another aspect of an embodiment of the present invention includes multiple checkout terminals that display the total price calculated by the checkout processing unit and through which purchasers of the group of products process payments, the transport unit having a branching unit that transports the group of products to one of the multiple checkout terminals downstream of the imaging area of ​​the imaging unit and the reception range of the RFID reader, and the checkout processing unit may display the total price on one of the checkout terminals selected by the branching unit.

[0021] According to this aspect, even if a problem occurs at one of the multiple payment terminals, the branching section can appropriately guide the customer to another terminal to carry out the payment procedure for the next customer. This prevents products from being stuck on the upstream conveyor section even if a problem occurs downstream in the conveying direction, and allows processes such as product identification to be carried out without delay, resulting in faster payment processing.

[0022] A checkout system according to another aspect of an embodiment of the present invention may provide a passing area that overlaps the imaging area of ​​the imaging unit and the receiving range of the RFID reader, and the imaging unit and the RFID reader may read information from the group of products that are passed through the passing area by a purchaser of the group of products.

[0023] According to this aspect, it is possible to apply this type of accounting system to existing configurations that do not have a conveying unit for conveying products, such as existing systems in which customers manually scan products to make payments, so-called self-checkouts, thereby improving versatility.

[0024] A checkout system according to another aspect of an embodiment of the present invention includes a first placement unit on which the group of products is placed before passing through the passing area and which measures the weight of the placed group of products, a second placement unit on which the group of products is placed after passing through the passing area and which measures the weight of the placed group of products, and an error detection unit, wherein the error detection unit may detect fraudulent behavior by a purchaser of the group of products when the amount of weight decrease measured by the first placement unit differs from the amount of weight increase measured by the second placement unit when products are moved from the first placement unit to the second placement unit.

[0025] According to this aspect, even in existing configurations that do not have a transport unit for transporting products, such as existing systems in which customers manually scan products to process payments, so-called self-checkout registers, it is possible to detect fraud such as customers concealing products while processing payments, thereby improving the accuracy of payment processing. [Effects of the Invention]

[0026] According to the present disclosure, it is possible to provide a checkout system that can achieve both improved accuracy in identifying products in automatic checkout processes and cost reduction, regardless of the types of products handled. [Brief explanation of the drawings]

[0027] [Figure 1] 1 is a schematic diagram showing the overall configuration of a transaction system 1 according to a first embodiment. [Figure 2] FIG. 2 is a functional block diagram of a control unit 9 according to the first embodiment. [Figure 3] FIG. 10 is a diagram illustrating an example of product identification in the present embodiment. [Figure 4] 1 is a flowchart of a transaction process performed by the transaction system 1 of the first embodiment. [Figure 5] FIG. 10 is a diagram showing an example of a display screen 81 of the accounting terminal 8 at the time of accounting. [Figure 6] FIG. 10 is a schematic diagram showing the overall configuration of a transaction system 1A according to a second embodiment. [Figure 7] FIG. 10 is a schematic diagram showing the overall configuration of a transaction system 1B according to a third embodiment. [Figure 8] FIG. 10 is a functional block diagram of a control unit 20 according to a third embodiment. [Figure 9] 11 is a flowchart of a transaction process by a transaction system 1B according to a third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0028] Hereinafter, embodiments will be described with reference to the accompanying drawings. To facilitate understanding of the description, the same components in the drawings will be denoted by the same reference numerals as much as possible, and duplicated descriptions will be omitted.

[0029] [First embodiment] The first embodiment will be described with reference to Figures 1 to 5. Figure 1 is a schematic diagram showing the overall configuration of a transaction system 1 according to the first embodiment.

[0030] As shown in Figure 1, the accounting system 1 of the first embodiment is applied to a system in a store such as a supermarket where a customer places a group of products 30 in a shopping cart 40, manually arranges the group of products 30 on a conveyor 2, and performs accounting for the group of products 30. The accounting system 1 of the first embodiment automatically acquires information about the group of products 30 arranged on the conveyor 2, and based on the acquired information, calculates and presents the total price to the customer.

[0031] The accounting system 1 includes a conveyor 2 (transport unit), a camera 3 (imaging unit), an RFID reader 4, an overlap detection unit 5 (error detection unit), an accounting terminal 8, and a control unit 9.

[0032] The conveyor 2 is an example of a conveying unit that conveys the placed product group 30 in a predetermined conveying direction A. A customer places the product group 30 on the conveyor 2 on the upstream side of the conveying direction. A checkout terminal 8 is located downstream in the conveying direction, and after placing the product group 30, the customer moves to the checkout terminal to complete the checkout procedure.

[0033] In the example of FIG. 1, a conveyor 2, specifically a belt conveyor, is illustrated as an example of a conveying unit. In a belt conveyor, a group of products 30 placed on an endless conveyor belt is conveyed by the rotation of pulleys. The conveyor 2 is not limited to the belt conveyor shown in FIG. 1, and other types of conveyors such as a drive roller conveyor or a chain conveyor may also be used. Furthermore, the conveying unit may be a device other than the conveyor 2 as long as it can convey the group of products 30 along the conveying direction A.

[0034] The camera 3 is an example of an imaging unit that captures images of the group of products 30 that move relatively through the imaging area B. In the first embodiment, the camera 3 is installed above the conveyor 2, and is installed so that a part of the conveyor 2 directly below it is the imaging area B. In the first embodiment, the camera 3 captures images of the group of products 30 that are transported on the conveyor 2 and pass through the imaging area B to obtain image information.

[0035] Any type of camera can be used as the camera 3, such as a digital camera capable of capturing still images or videos, an RGB camera that captures RGB images, a depth camera that captures depth images, or an RGB-D camera that can capture both RGB images and depth images. While the camera 3 is illustrated in Fig. 1 as an example of the imaging unit, the imaging unit may be a device other than the camera 3, such as an image sensor, as long as it can capture an image of the product group 30 that moves relatively within the imaging area B and acquire image information of the product group 30.

[0036] The RFID reader 4 reads information from RFID tags 33 affixed to a group of products 30 that move relatively within a reception range C. The RFID tags 33 store various information about the products, such as the type, product name, price, weight, and dimensions. In the first embodiment, the RFID reader 4 is installed, for example, inside the conveyor 2, so that a part of the conveyor 2 directly above it becomes the reception range C. In the first embodiment, information is obtained from the RFID tags 33 affixed to products that are transported on the conveyor 2 and pass through the reception range C.

[0037] In the following embodiment, a case will be described in which RFID tags 33 are attached to at least some of the products in the product group 30, but this embodiment can be implemented even if RFID tags 33 are attached to all of the products in the product group 30, or even if RFID tags 33 are not attached to all of the products.

[0038] The product group 30 handled in a store to which the accounting system 1 is applied can be broadly divided into products 31, such as snacks, bottled drinks, and fruit and vegetables, which are relatively inexpensive and therefore difficult to attach RFID tags 33 to, and products 32, such as clothing and books, which are relatively expensive and can be attached with RFID tags 33. The products 31 are also referred to as "first products 31," whose price information is acquired by product identification based on image information captured by the camera 3. The products 32 are also referred to as "second products 32," whose price information is acquired by reading product information from RFID tags 33 using the RFID reader 4. The definition of the product group 30 described above can also be expressed as "the product group 30 includes first products 31 and second products 32 classified based on a predetermined criterion." Examples of the "predetermined criterion" include evaluation criteria such as whether it is easy to affix RFID tags 33 to the surface of the products and whether the price of the products is relatively low compared to the unit price of the RFID tags 33, as described above.

[0039] The camera 3 and RFID reader 4 are preferably installed at approximately the same position in the conveying direction A, as shown in Figure 1, for example. In other words, it is preferable that the imaging area B of the camera 3 and the receiving range C of the RFID reader 4 overlap. This allows the range for reading information from the first product 31 and the second product 32 to be the same, thereby shortening the dimension of the conveyor 2 in the conveying direction. However, the imaging area B of the camera 3 and the receiving range C of the RFID reader 4 may also be configured to be arranged in series along the conveying direction A.

[0040] The overlap detection unit 5 detects overlaps between the group of products 30 being transported on the conveyor 2. In the first embodiment, two detection methods are applied, and the unit has a weighing scale 6 (weighing unit) and a photoelectric sensor 7 (height measurement unit). The weighing scale 6 is installed inside the conveyor 2 and measures the weight of each item in the group of products 30 passing over the weighing scale 6. The photoelectric sensor 7 has a pair of light-emitting and light-receiving units on both sides of the width direction, which is perpendicular to the conveyance direction of the conveyor 2, and is installed so that the light-receiving unit can receive light emitted from the light-emitting unit. For example, the light-emitting unit is installed so that it can emit light over a predetermined range in the height direction. As the group of products 30 passes between the light-emitting unit and the light-receiving unit, part of the light is blocked depending on the height of each item, so the height of the passing items can be detected depending on the reception status of the light-receiving unit.

[0041] That is, in this embodiment, the overlap detection unit 5 acquires information on the weight of the passing product measured by the weighing scale 6 and the height of the passing product measured by the photoelectric sensor 7. This weight and height information is used by the control unit 9, which will be described later, to detect overlap errors.

[0042] The overlap detection unit 5 may be configured to include only one of the weighing scale 6 and the photoelectric sensor 7, or may be configured to use other elements. In the example of Fig. 1, the overlap detection unit 5 is configured to be arranged downstream of the imaging area B of the camera 3 and the receiving range C of the RFID reader 4 in the conveying direction A, but it may also be configured to be arranged upstream of the imaging area B and the receiving range C in the conveying direction A.

[0043] The accounting terminal 8 is a device that displays information such as a list of products 30 and the total price read by the camera 3 and RFID reader 4 on a screen 81, and allows the customer to complete payment procedures. The accounting terminal 8 is communicatively connected to the control unit 9, and presents accounting information to the customer and processes the transaction based on the information received from the control unit 9.

[0044] The control unit 9 controls the operation of each part of the accounting system 1. The control unit 9 operates the conveyor 2 to transport a group of products, and uses the camera 3 and RFID reader 4 to acquire information from the group of products being transported and identify the products. The control unit 9 then acquires price information for the identified group of products, calculates the total amount, and outputs it to the accounting terminal 8.

[0045] 2 is a functional block diagram of the control unit 9 of the first embodiment. The control unit 9 has the following functions related to the above-mentioned series of processes: a product counting unit 11, a first product identification unit 12, a second product identification unit 13, an error detection unit 14, an accounting processing unit 15, a product information storage unit 16, and a conveyor control unit 17.

[0046] The product counting unit 11 counts the total number of products in the product group 30. For example, the product counting unit 11 uses image information from the camera 3 to count the total number of products in the product group 30 for a specific customer being transported on the conveyor 2. The product counting unit 11 outputs information on the counted total number of products in the product group 30 to the error detection unit 14.

[0047] When counting items using image information, the item counting unit 11 can use a well-known object detection method, such as semantic segmentation using deep learning, to extract areas of individual items from the image information and calculate the number of items based on the number of extracted areas. Alternatively, AR markers or two-dimensional codes can be attached to all items in the group of items 30, and these markers can be detected from the image information using a well-known image recognition method, and the number of items can be calculated based on the number of detected markers. The accounting system 1 may also be configured to include an imaging device separate from the camera 3 to provide image information to the item counting unit 11.

[0048] The counting method of the commodity counting unit 11 may be one that uses information other than image information, and may be configured to count the total number of commodities based on the number of commodity groups 30 that pass the photoelectric sensor 7 shown in FIG.

[0049] The first product identification unit 12 acquires the number of items and price information of the first product 31 identified based on the image information of the product group 30 captured by the camera 3. The first product identification unit 12 can identify the type of product in question by comparing the image information of the product group 30 with individual sample images of the product group 30 stored in the product information storage unit 16 using a well-known search method, for example, pattern matching. The first product identification unit 12 counts the number of items in question based on the identified product type, and also acquires price information stored in the product information storage unit 16 in association with the type information.

[0050] It is preferable that the commodity information storage unit 16 stores a large number of sample images, such as images captured from multiple directions for each commodity, and the first commodity identification unit 12 identifies the type of commodity using these sample images. This improves the accuracy of commodity identification by the first commodity identification unit 12.

[0051] The first product identification unit 12 may also use a machine learning technique such as a neural network to identify the type of product from the image information of the product group 30. In this case, the first product identification unit 12 has a trained model in which an input / output relationship is acquired in advance through machine learning, with the product appearance image as input and the product type as output. By inputting image information from the camera 3 into the trained model, the first product identification unit 12 can identify the type of product based on the output of the trained model. Unlike pattern matching, a configuration using machine learning involves training the model in advance, allowing for a larger number of sample images to be used as training data for machine learning compared to pattern matching, which requires real-time performance. This further improves the accuracy of product identification based on image information by the first product identification unit 12.

[0052] The second product identification unit 13 acquires the number and price information of the second product 32 to which the RFID tag 33 is attached based on the information of the RFID tag 33 read by the RFID reader 4. The second product identification unit 13 acquires information such as the type and price of the product from the information read from the RFID tag 33, for example. The second product identification unit 13 also counts the number of products based on the identified type of product.

[0053] The first product identification unit 12 and the second product identification unit 13 output the acquired information on the number and price of the product group 30 to the accounting processing unit 15. They also output the acquired information on the number, weight, dimensions, etc. of the product group 30 to the error detection unit 14.

[0054] The error detection unit 14 detects the occurrence of various errors during a series of processes in the accounting system 1. The error detection unit 14 detects an error when the number of items in the product group 30 acquired by the first product identification unit 12 and the second product identification unit 13 differs from the total number of items counted by the product counting unit 11. The error detection unit 14 may also be configured to detect overlapping of the product group 30 moving on the conveyor 2, or to detect duplicate reading of the same item by the first product identification unit 12 and the second product identification unit 13.

[0055] The accounting processing unit 15 calculates the total price of the first product 31 and the second product 32 based on the item number and price information of the product group 30 acquired by the first product identification unit 12 and the second product identification unit 13. The accounting processing unit 15 outputs information such as the product name, item number, and price of each item in the product group 30, as well as information on the total price, to the accounting terminal 8.

[0056] The product information storage unit 16 stores various information about each product in a product group 30 sold in the store where the checkout system 1 is installed. As described above, the product group 30 is classified into first products 31 and second products 32. The product information storage unit 16 stores information about the first products 31, such as a sample image used for image recognition, the product type, product name, price, weight, and dimensions, linked to each product. The product information storage unit 16 may also be configured to store at least some of the various information recorded on the RFID tag 33 described above as information about the second products 32.

[0057] The conveyor control unit 17 controls the operation of the conveyor 2 .

[0058] The control unit 9 can be physically configured as a computer system including a CPU (Central Processing Unit), RAM (Random Access Memory) and ROM (Read Only Memory) as main storage devices, a communication module, an auxiliary storage device, etc. Each function of the control unit 9 described with reference to Fig. 2 is realized by loading predetermined computer software into the CPU, RAM, etc., thereby operating various hardware under the control of the CPU and reading and writing data from and to the RAM.

[0059] 3A and 3B are diagrams showing an example of product identification in this embodiment. As shown in Fig. 3A, in the case of first products 31 such as plastic bottles and sweets, the product is identified by the first product identification unit 12 based on image information from the camera 3. Furthermore, since no RFID tag 33 is attached to these first products 31, the RFID reader 4 does not read information from these first products 31, and the second product identification unit 13 does not identify the product.

[0060] As shown in Figure 3(B), in the case of second products 32 such as books or clothes, the RFID reader 4 reads information from the RFID tag 33 attached to each product, and the second product identification unit 13 identifies the product based on the information read from the RFID tag 33. Furthermore, since no sample images of these second products 32 are stored in the product information storage unit 16, the first product identification unit 12 is generally unable to identify the product. However, there may be cases where the first product identification unit 12 mistakenly detects a different product, in which case the error detection unit 14 will detect a reading error, as will be described later.

[0061] As shown in Figure 3(C), a situation may occur in which the first product 31 is transported overlapping the second product 32. In this case, the image information from the camera 3 captures only the upper first product 31, so the first product identification unit 12 can identify the first product 31 based on the image information. On the other hand, the RFID reader 4 can read the information from the RFID tag 33 of the lower second product 32, so the second product identification unit 13 can identify the second product 32 based on the information read from the RFID tag 33.

[0062] As shown in Figure 3(D), a situation may occur in which a second product 32 is transported overlapping a first product 31. In this case, the image information from camera 3 captures only the upper second product 32, so the first product identification unit 12 cannot identify the first product 31 based on the image information. On the other hand, the second product identification unit 13 can identify the second product 32 based on the information read from the RFID tag 33. Therefore, in this case, the first product hidden by the second product 32 will not be identified or counted. To prevent this situation from being overlooked, in this embodiment, the overlap detection unit 5 determines whether multiple products are overlapping based on weight and height information, and performs error detection.

[0063] FIG. 4 is a flowchart of the accounting process performed by the accounting system 1 of the first embodiment.

[0064] In step S101, when a customer begins placing a group of products 30 on the conveyor 2, the camera 3 and RFID reader 4 begin reading. As shown in FIG. 1, a marker such as a rod-shaped bar 34 can be used as a trigger to start reading. Before lining up the group of products 30 to be purchased on the conveyor 2, the customer places the bar 34 just before it on the downstream side in the conveying direction A. When the accounting system 1 detects this bar 34 from an image captured by the camera 3, for example, it begins reading.

[0065] In step S102, the product counting unit 11 starts counting the number of products in the product group 30. The product counting unit 11 can count the total number of products passing through the imaging area B of the camera 3 on the conveyor 2, for example, by extracting products from the image information of the camera 3.

[0066] In step S103, the first product identification unit 12 identifies the first product 31 using the image information from the camera 3. The first product identification unit 12 can identify the type of the relevant product by using a well-known search method, such as pattern matching, to compare the image information of the product group 30 with individual sample images of the product group 30 stored in the product information storage unit 16. Based on the identified product type, the first product identification unit 12 counts the number of items of the relevant product, and also obtains price information that is stored in the product information storage unit 16 in association with the type information.

[0067] In step S104, the second product identification unit 13 identifies the second product 32 using the information read by the RFID reader 4 from the RFID tag 33. The second product identification unit 13 acquires information such as the type and price of the product from the information read from the RFID tag 33. The second product identification unit 13 also counts the number of points of the product based on the identified type of product.

[0068] The processing of steps S103 and S104 does not necessarily have to be performed in this order, but may be performed in the reverse order or simultaneously in parallel.

[0069] In step S105, the error detection unit 14 determines whether double detection, in which the first product 31 and the second product 32 are simultaneously identified, has occurred in steps S103 and S104. The error detection unit 14 can determine whether double detection has occurred, for example, based on the timing of information input from the first product identification unit 12 and the second product identification unit 13. Possible cases in which double detection occurs include when the first product identification unit 12 erroneously detects the second product 32 as the first product 31, as described with reference to FIG. 3(B), or when the first product 31 is transported overlapping the second product 32, as shown in FIG. 3(C).

[0070] If it is determined that double detection has occurred (YES in S105), a reading error is detected by the error detection unit 14 in step S106, and the process returns to step S102. Possible actions to be taken after a reading error is detected include, for example, notifying the customer of the occurrence of the error by voice or text information, having the relevant products rearranged upstream on the conveyor 2 so that they can be read again, or stopping the conveyor 2 from transporting the group of products 30.

[0071] On the other hand, if it is not determined that double detection has occurred (NO in S105), in step S107, the error detection unit 14 acquires information on the weight and dimensions of the product identified in step S103 or step S104 from the product information storage unit 16. This information is used as reference weight information and reference height information in a later stage.

[0072] In step S108, the overlap detection unit 5 measures the weight and height of the product identified in step S103 or step S104 downstream of the imaging area B and detection range C in the conveying direction A. The weight scale 6 of the overlap detection unit 5 measures the weight of the product, and the photoelectric sensor 7 measures the height of the product. The overlap detection unit 5 outputs the measured weight and height information to the error detection unit 14.

[0073] In step S109, it is determined whether the weight or height of the corresponding product measured in step S108 is equal to or greater than the reference weight or height acquired in step S107. If the weight or height of the corresponding product is equal to or greater than the reference value (YES in S109), the error detection unit 14 detects an overlap error in step S110, and the process returns to step S102. An overlap error can occur, for example, when a second product 32 is transported overlapping a first product 31, as shown in FIG. 3(D). Unlike the situation in FIG. 3(C), the overlap error cannot be detected in the situation in FIG. 3(D). Therefore, overlapping of the products is detected by detecting a weight that is heavier than the reference value or a height that is higher than the reference value. Possible actions after detecting an overlap error include, for example, notifying the customer of the error via voice or text, having the corresponding product rearranged upstream on the conveyor 2 so that it can be scanned again, or stopping the conveyor 2 from transporting the group of products 30.

[0074] On the other hand, if the weight or height of the corresponding product is below the reference value (NO in S109), it is determined that there are no overlapping products, and the transaction information is updated by the transaction processing unit 15 in step S111. The transaction processing unit 15 stores information such as the type, quantity, and price of the products identified from the start of reading in step S101 as transaction information, and adds information about products identified in step S103 or step S104 that passed the judgments in step S105 and step S109 to this transaction information.

[0075] In step S112, it is determined whether the camera 3 and RFID reader 4 have finished reading the group of products 30. As shown in FIG. 1, a marker such as a rod-shaped bar 35 can be used to trigger the completion of reading, just as at the start. After the customer has finished arranging the group of products 30 they wish to purchase on the conveyor 2, they place the bar 35 immediately upstream in the conveying direction A. The accounting system 1 completes reading when it detects this bar 35 from the image captured by the camera 3 or the like.

[0076] If it is determined that reading of the group of products 30 has not been completed (NO in S112), the process returns to step S102 and the next product is read.

[0077] On the other hand, if it is determined that the reading of the product group 30 has been completed (YES in S112), in step S113, the error detection unit 14 determines whether the "read number" of products recorded in the transaction information by the transaction processing unit 15 in step S111 is the same as the "counted number" of the product group 30 counted by the product counting unit 11 in step S102.

[0078] If the number of items read and the number of items counted are not the same (NO in S113), it is possible that some error was overlooked when reading the group of products 30, such as overlapping of products as shown in FIG. 3(D), erroneous detection of the first product 31 or the second product 32, or failure to detect the first product 31 based on the image information by the first product identification unit 12. For this reason, a reading error is detected by the error detection unit 14 in step S106, and the process returns to step S102. Possible actions to be taken after a reading error is detected include, for example, notifying the customer of the occurrence of the error by voice or text, having the relevant products rearranged upstream on the conveyor 2 so that they can be read again, or stopping the conveyor 2 from transporting the group of products 30.

[0079] Furthermore, for example, if a personal item is placed on the conveyor 2 by mistake, or if a product that should have an RFID tag 33 is not affixed, and the product cannot be identified from the captured image information, the information may not be read by either the first product identification unit 12 or the second product identification unit 13. In such cases, it may be possible to guide the customer to an area where a store clerk or the like is located.

[0080] On the other hand, if the read number and counted number are the same (YES in S113), the accounting processor 15 outputs the accounting information for the product group 30 recorded in step S111 to the accounting terminal 8 in step S114.

[0081] Figure 5 is a diagram showing an example of the display screen 81 at the time of checkout on the checkout terminal 8. As shown in Figure 5, the display screen 81 displays a list of information about the product group 30 that has been read by, for example, the camera 3 or the RFID reader 4 and identified by the first product identification unit 12 or the second product identification unit 13 in step S103 or step S104.

[0082] 5 shows an example of a list display of three types of products, with each product list 82, 83, and 84 displayed on a separate line. Each product list 82, 83, and 84 displays information such as the product name, price, and number of items. As shown in FIG. 5, each product list 82, 83, and 84 may also display a thumbnail of the product, making it easier for customers to intuitively understand the product they wish to purchase.

[0083] Also, on the display screen 81, below the lists 82, 83, and 84 of each product, a payment amount display area 85 is displayed which displays the total number of products purchased and the total amount.

[0084] 4, when display screen 81 shown in FIG. 5 is displayed on accounting terminal 8 in step S114, the customer checks the contents of display screen 81 and performs payment procedures using accounting terminal 8. When the processing of step S114 is completed, this control flow ends.

[0085] The effects of the transaction system 1 according to the first embodiment will be described.

[0086] The checkout system 1 of the first embodiment includes a camera 3 that captures images of a group of products 30 that move relatively within an imaging area B, and an RFID reader 4 that reads information from RFID tags 33 affixed to at least some of the group of products 30 that move relatively within a reception range C. The group of products 30 includes a first product 31 and a second product 32 that are classified based on predetermined criteria. The checkout system 1 also includes a first product identification unit 12 that acquires the number and price information of the first product 31 identified based on the image information of the group of products 30 captured by the camera 3, a second product identification unit 13 that acquires the number and price information of the second product 32 to which the RFID tag 33 is attached based on the information read by the RFID reader 4, and a checkout processing unit 15 that calculates the total price of the first product 31 and the second product 32 based on the number and price information acquired by the first product identification unit 12 and the second product identification unit 13.

[0087] With this configuration, for example, for inexpensive products such as food (first products 31), product identification is performed using image processing, making it possible to manage products without using RFID tags 33, which have a relatively high unit price, thereby reducing product management costs. Furthermore, for easily deformed products such as clothing (second products 32), product identification using image processing may result in a decrease in product identification accuracy, but product identification using RFID tags 33 improves identification accuracy. In this way, the checkout system 1 of the first embodiment selectively applies a method suited to the product management suitability of the products handled, thereby improving the product identification accuracy of automated checkout processing and reducing costs regardless of the type of product handled.

[0088] In addition, the accounting system 1 of the first embodiment includes a product counting unit 11 that counts the total number of items in the product group 30, and an error detection unit 14 that detects an error when the number of items obtained by the first product identification unit 12 and the second product identification unit 13 differs from the total number of items counted by the product counting unit 11.

[0089] As described above, when two product identification methods are used in combination in checkout processing, problems such as duplicate detection of the same product can occur. This configuration compares the number of products identified by image processing, the number of products identified by RFID tag 33, and the total number counted by product counting unit 11, and detects an error if there is a discrepancy in the numbers. This reduces the occurrence of problems that can occur when two identification methods are used in combination, and further improves the accuracy of product identification.

[0090] The checkout system 1 of the first embodiment also includes a conveyor 2 that transports a group of products 30. A camera 3 and an RFID reader 4 read information from the group of products 30 transported on the conveyor 2.

[0091] With this configuration, customers can simply place the group of products 30 they wish to purchase on the conveyor 2, which will automatically pass the group of products 30 through the imaging area B of the camera 3 and the receiving range C of the RFID reader 4, thereby automatically processing the transaction, significantly reducing the burden on customers.

[0092] The accounting system 1 of the first embodiment also includes a product information storage unit 16 that stores weight information for each item in the product group 30, and a weighing scale 6 that measures the weight of the product group 30 being transported on the conveyor 2. The error detection unit 14 acquires weight information from the product information storage unit 16 for the product group 30 read by the first product identification unit 12 or the second product identification unit 13, and detects an error of multiple items overlapping if the acquired weight information differs from the weight measured by the weighing scale 6.

[0093] Similarly, the checkout system 1 of the first embodiment includes a product information storage unit 16 that stores height information for each item in the product group 30, and a photoelectric sensor 7 that measures the height of the product group 30 being transported on the conveyor 2. The error detection unit 14 obtains height information from the product information storage unit 16 for the product group 30 read by the first product identification unit 12 or the second product identification unit 13, and if the obtained height information differs from the height measured by the photoelectric sensor 7, detects an error in which multiple items are overlapping.

[0094] With these configurations, even in a situation where the first product 31 is hidden under the second product 32 and cannot be identified by the camera 3, as illustrated in Figure 3(D), overlap errors can be detected based on changes in the weight and height of the identified product, thereby further reducing false detection of products and further improving the accuracy of product identification.

[0095] In the first embodiment, the overlap detection unit 5 from which the error detection unit 14 acquires information to detect an overlap error in the group of products 30 includes a weighing scale 6 that measures product weight information and a photoelectric sensor 7 that measures product height information. However, the overlap detection unit 5 may be configured to acquire information other than weight and height, as long as it acquires information that allows the error detection unit 14 to detect an overlap error. For example, the overlap detection unit 5 may be configured to acquire a transmission image of the group of products 30 transported on the conveyor 2 using an optical method such as an infrared camera or X-ray camera, and the error detection unit 14 may use the transmission image to determine whether or not the products are overlapping.

[0096] Furthermore, when products overlap, it is believed that characteristics specific to the overlapping state, such as an unnatural bulge in the appearance of the upper product, will occur. The error detection unit 14 may be configured to use a machine learning method such as a neural network to extract characteristics specific to such an overlapping state and detect an overlap error. In this case, the overlap detection unit 5 may capture an image of the product's appearance from above, similar to the camera 3, and provide the image to the error detection unit 14.

[0097] Furthermore, the first product 31, from which price information is acquired by performing product identification based on image information captured by the camera 3, does not have an RFID tag 33 attached. Therefore, if the first product identification unit 12 cannot detect the first product 31 from the image information, the entire system may not be able to recognize the product itself. To avoid this situation, for example, a mark or other indicator may be attached to the surface of the first product 31 to which the RFID tag 33 is not attached, and the first product identification unit 12 may use this mark as part of the information for product identification. This improves the accuracy with which the first product identification unit 12 can identify the first product 31.

[0098] [Second embodiment] The second embodiment will be described with reference to Fig. 6. Fig. 6 is a schematic diagram showing the overall configuration of a transaction system 1A according to the second embodiment.

[0099] Similar to the first embodiment, the checkout system 1A according to the second embodiment is used in stores such as supermarkets, where customers manually arrange a group of products 30 placed in a shopping cart 40 on a conveyor 2 and perform checkout processing for these group of products 30. Similar to the first embodiment, the checkout system 1A according to the second embodiment includes a conveyor 2, a camera 3, an RFID reader 4, an overlap detection unit 5, and a control unit 9, although the overlap detection unit 5 is not shown in FIG. 6.

[0100] As shown in FIG. 6, the checkout system 1A according to the second embodiment displays checkout information, such as the total price calculated by the checkout processor 15, and includes multiple checkout terminals 8A, 8B, and 8C where purchasers of a group of products 30 process payments. The conveyor 2 also includes a branching section 2D, located downstream in the conveyance direction A from the imaging area B of the camera 3 and the reception range C of the RFID reader 4, that transports the group of products 30 to one of the multiple checkout terminals 8A, 8B, and 8C. Downstream in the conveyance direction A from the branching section 2D, the conveyor 2 has multiple conveyance paths 2A, 2B, and 2C (three in the example shown in FIG. 6) corresponding to the number of checkout terminals 8A, 8B, and 8C. Each checkout terminal 8A, 8B, and 8C is located at the downstream end of each conveyance path 2A, 2B, and 2C.

[0101] The structure of the branching section 2D can be that of an existing conveyor branching device, such as a structure in which a pusher pushes products in the direction of the desired conveying path 2A, 2B, or 2C, or a structure in which a movable guide rail guides products in the direction of the desired conveying path 2A, 2B, or 2C.

[0102] In the accounting system 1A of the second embodiment, the accounting processing unit 15 displays the total price on one of the multiple accounting terminals 8A, 8B, 8C that is located on the conveying path selected by the branching section 2D of the conveyor 2.

[0103] In a configuration with a single accounting terminal 8, as in the first embodiment, problems can occur downstream in conveyance direction A, such as a customer struggling to operate the accounting terminal 8 and taking extra time to complete the payment procedure. In this case, products may accumulate upstream on the conveyor 2, potentially delaying processes such as product identification. In contrast, accounting system 1A in the second embodiment is configured with multiple accounting terminals 8A, 8B, and 8C, and displays the total price on a single accounting terminal located on a conveyor path selected by branching section 2D of conveyor 2.

[0104] With this configuration, even if a problem occurs at one of the multiple accounting terminals 8A, 8B, or 8C, the branching section 2D can appropriately guide the customer to another terminal to carry out the payment procedure for the next customer. This prevents products from being stuck on the conveyor 2 upstream even if a problem occurs downstream in the conveying direction A, and allows processes such as product identification to be carried out without delay, resulting in faster accounting processing.

[0105] Additionally, the accounting system 1A of the second embodiment may be provided with a notification unit near the branching point 2D of the conveyor 2 that notifies the customer which of the multiple accounting terminals 8A, 8B, and 8C to use to make the payment. For example, as shown in Figure 6, the notification unit may be configured with illuminated arrow-shaped signs 18A, 18B, and 18C for each of the transport paths 2A, 2B, and 2C. In the example of Figure 6, sign 18A for transport path 2A lights up to guide the customer to accounting terminal 8A.

[0106] By providing such a notification unit, customers can be more clearly guided to the appropriate payment terminal 8A, making the payment process smoother. Note that the notification method used by the notification unit is not limited to the visual method shown in Figure 6, and other methods such as audio or text may also be used.

[0107] [Third embodiment] The third embodiment will be described with reference to Figures 7 to 9. Figure 7 is a schematic diagram showing the overall configuration of a transaction system 1B according to the third embodiment.

[0108] As shown in Figure 7, the checkout system 1B according to the third embodiment is used in a supermarket or other store where a customer places a group of products 30 in a shopping cart 40 and passes each item through a designated passage area 21, performing checkout processing for the group of products 30. The checkout system 1B of the third embodiment automatically acquires information about the group of products 30 as they pass through the passage area 21, and then calculates and presents the total price to the customer based on the acquired information.

[0109] Similar to the first embodiment, the checkout system 1B includes a camera 3, an RFID reader 4, a checkout terminal 8, and a control unit 20. The checkout system 1B also has a passage area 21 where the imaging area B of the camera 3 and the reception range C of the RFID reader 4 overlap, and the camera 3 and RFID reader 4 read information from the group of products 30 as they are passed through the passage area 21 by a purchaser of the group of products 30. In other words, in the third embodiment, the group of products 30 moves relatively between the imaging area B of the camera and the reception range C of the RFID reader 4 as the purchaser manually moves the group of products 30.

[0110] The passage area 21 is formed, for example, by a base having a flat surface on the top as shown in Fig. 7. The camera 3 is installed above the base so that a part of the flat surface of the top of the base directly below it becomes the imaging area B. The RFID reader 4 is installed, for example, inside the base so that a part of the flat surface directly above it becomes the receiving area C. In other words, the space between the camera 3 and the RFID reader 4 becomes the passage area 21.

[0111] The checkout system 1B also includes a first placement unit 22 on which the group of products 30 is placed before passing through the passing area 21 and which measures the weight of the placed group of products 30, and a second placement unit 23 on which the group of products 30 is placed after passing through the passing area 21 and which measures the weight of the placed group of products 30. The first placement unit 22 and the second placement unit 23 are, for example, housings with a flat surface on the top as shown in FIG. 7, and are equipped with a weighing scale inside so that the weight of an object placed on the flat upper surface can be measured.

[0112] When processing the transaction, the customer places a shopping basket 40 containing a group of products 30 on the first placement section 22, moves the products taken out of the shopping basket 40 so that they pass through the passing area 21, and transfers them into bags 41 placed on the second placement section 23.

[0113] The upper planes of first placement section 22 and second placement section 23 are preferably positioned below the upper plane of the base of passage area 21. This allows a customer to easily place items in bags 41 by lifting items from shopping basket 40 and passing them horizontally through passage area 21 to above bags 41 during the checkout process, making it easier to move group of items 30 from first placement section 22 to second placement section 23.

[0114] 8 is a functional block diagram of the control unit 20 of the third embodiment. As in the first embodiment, the control unit 20 has a product counting unit 11, a first product identification unit 12, a second product identification unit 13, an accounting processing unit 15, and a product information storage unit 16. These elements are the same as those in the first embodiment, so a description thereof will be omitted.

[0115] The control unit 20 includes an error detection unit 24. Similar to the error detection unit 14 of the first embodiment, the error detection unit 24 detects an error when the number of items in the product group 30 acquired by the first product identification unit 12 and the second product identification unit 13 differs from the total number of items counted by the product counting unit 11. The error detection unit 24 may also be configured to perform at least some of the following functions: detecting overlapping of the product group 30 moving through the passage area 21; detecting duplicate readings of the same item by the first product identification unit 12 and the second product identification unit 13; or detecting fraudulent behavior by a purchaser of the product group 30. This makes it possible to detect various errors that may occur in automated checkout processes, thereby improving the accuracy and reliability of the automated checkout processes.

[0116] Weight information of the group of products placed on each section is input from the first placement section 22 and the second placement section 23 to the error detection section 24, and the error detection section 24 detects errors based on this weight information. For example, when products are moved from the first placement section 22 to the second placement section 23, if the amount of weight decrease measured by the first placement section 22 differs from the amount of weight increase measured by the second placement section 23, the error detection section 24 can detect fraudulent behavior by the purchaser of the group of products 30 as a fraudulent error.

[0117] FIG. 9 is a flowchart of the transaction process performed by the transaction system 1B of the third embodiment.

[0118] In step S201, reading is started by the camera 3 and the RFID reader 4. The trigger for starting reading can be, for example, when a customer places the shopping basket 40 on the first placement unit 22 and the first placement unit 22 detects a change in weight.

[0119] The processing in steps S202 to S206 is the same as steps S102 to S106 in FIG. 3, and therefore a description thereof will be omitted.

[0120] In step S207, the error detection unit 24 obtains weight information of the product identified in step S203 or step S204 from the product information storage unit 16. This information is used as reference weight information in a later stage.

[0121] In step S208, the first and second loading sections 22 and 23 measure the weight of the product group 30 in each section after the products have been moved, and the error detection section 24 measures the weight difference between the product group 30 in the first loading section 22 and the product group 30 in the second loading section 23 based on this weight information.

[0122] In step S209, it is determined whether the weight difference measured in step S208 is equal to or greater than the reference weight information for the corresponding product acquired in step S207. This weight difference corresponds to the weight of the corresponding product moved from the first placement unit 22 to the second placement unit 23. Therefore, if the weight difference is equal to or greater than the reference value (YES in S209), a fraud error is detected by the error detection unit 14 in step S210, and the process returns to step S102. A fraud error can occur, for example, when a second product 32 is stacked on top of a first product 31 and the customer moves the first product 31 while hiding it under the second product 32, as shown in FIG. 3(D). In the state shown in FIG. 3(D), unlike the case shown in FIG. 3(C), double detection cannot be detected. Therefore, a weight difference greater than the reference value is detected, and the overlapping of the products is detected. Possible actions to be taken after a fraud error is detected include, for example, informing the customer of the error by voice or text and having them re-scan the corresponding product.

[0123] The processing in steps S211 to S214 is the same as steps S111 to S114 in FIG. 3, and therefore a description thereof will be omitted.

[0124] The accounting system 1B of the third embodiment has a passing area 21 that overlaps the imaging area B of the camera 3 and the receiving range C of the RFID reader 4, and the camera 3 and the RFID reader 4 read information from the group of products 30 that are passed through the passing area 21 by the purchaser of the group of products 30.

[0125] This configuration makes it possible to apply the accounting system 1B of this embodiment to existing systems that do not have a conveying unit such as a conveyor 2, such as so-called self-checkout systems in which customers manually scan products to check out, thereby improving versatility.

[0126] The checkout system 1B of the third embodiment also includes a first placement unit 22 on which the group of products 30 is placed before passing through the passage area 21 and which measures the weight of the placed group of products 30, and a second placement unit 23 on which the group of products 30 is placed after passing through the passage area 21 and which measures the weight of the placed group of products 30. When products are moved from the first placement unit 22 to the second placement unit 23, the error detection unit 24 detects fraudulent behavior by the purchaser of the group of products 30 if the weight loss measured by the first placement unit 22 differs from the weight increase measured by the second placement unit 23.

[0127] This configuration makes it possible to detect fraud, such as customers hiding products while making payments, even in existing systems that do not have a conveyor 2 or other transport unit, such as so-called self-checkout systems where customers manually scan products to make payments, thereby improving the accuracy of payment processing.

[0128] In the third embodiment, a configuration has been exemplified in which the error detection unit 24 uses the weight difference between the group of products placed on the first placement unit 22 and the group of products moved to the second placement unit 23 to detect fraudulent behavior, such as a customer stacking multiple products to hide some of the products when processing a transaction, as a fraudulent error. However, any configuration may be used that uses information other than weight difference as long as it is information that can be used to determine fraudulent behavior.

[0129] For example, the upper flat surface of the base of the passage area 21 is made of a transparent material, and an imaging device separate from the camera 3 is installed inside the base. This imaging device is installed so that part of the upper flat surface directly above it serves as the imaging area. In this configuration, the imaging device can capture images of products moving through the passage area 21 from below the passage area 21, so the error detection unit 24 can use the images captured by the imaging device to detect fraudulent behavior such as performing a checkout process while hiding some of the products underneath, as described above.

[0130] Furthermore, in the above embodiment, in addition to the camera 3 that captures image information of the product group 30, a separate imaging device that captures images of the entire checkout system 1, 1A, 1B may be provided, and the control units 9, 20 may use the images captured by the imaging device to monitor the behavior of customers purchasing the product group 30. This allows for more reliable detection of fraudulent behavior by customers.

[0131] The present embodiment has been described above with reference to specific examples. However, the present disclosure is not limited to these specific examples. Design modifications to these specific examples made by a person skilled in the art as appropriate are also included within the scope of the present disclosure as long as they comprise the features of the present disclosure. The elements of the above-described specific examples, as well as their arrangement, conditions, shape, etc., are not limited to those exemplified and can be modified as appropriate. The elements of the above-described specific examples can be combined in various ways as appropriate, as long as no technical contradictions arise. [Explanation of symbols]

[0132] 1, 1A, 1B Accounting System 2 Conveyor (transport section) 2D Branch 3 Camera (imaging unit) 4 RFID Reader 5. Overlap detection unit 6 Weight scale (weighing section) 7 Photoelectric sensor (height measurement section) 8, 8A, 8B, 8C Accounting terminals 9, 20 Control section 11 Product Counting Department 12 First Product Identification Department 13 Second Product Identification Department 14, 24 Error detection unit 15 Accounting Department 16 Product information storage section 21 Passage area 22 First placement section 23 Second placement section 30 product groups 31 1st product 32 2nd product 33 RFID tags B. Imaging area C. Reception range

Claims

1. an imaging unit that captures images of a group of products that move relatively within an imaging area; an RFID reader that reads information from RFID tags attached to the group of products that move relatively within a receiving range; Equipped with the product group includes a first product and a second product classified based on a predetermined criterion; a first commodity identification unit that acquires the number and price information of the first commodity identified based on image information of the commodity group captured by the imaging unit; a second commodity identification unit that acquires quantity and price information of the second commodity to which the RFID tag is attached based on the information read by the RFID reader; an accounting processing unit that calculates a total price of the first product and the second product based on the number and price information acquired by the first product identification unit and the second product identification unit; a product counting unit that counts the total number of the product group; an error detection unit that detects an error when the points acquired by the first commodity identification unit and the second commodity identification unit differ from the total points counted by the commodity counting unit; An accounting system that includes:

2. The error detection unit At least a part of the detection of overlap of the relatively moving product group, the detection of double reading of the same product by the first product identification unit and the second product identification unit, or the detection of fraudulent behavior by a purchaser of the product group is carried out. The accounting system of claim 1 .

3. a conveying unit that conveys the group of products; the imaging unit and the RFID reader read the information from the group of products being transported by the transport unit; 3. The accounting system according to claim 1 or 2.

4. a product information storage unit that stores weight information of each of the product groups; a weighing unit that measures the weight of the group of products being transported on the transport unit; Equipped with the error detection unit acquires the weight information from the product information storage unit for the group of products read by the first product identification unit or the second product identification unit, and detects an error of multiple products being overlapped when the acquired weight information differs from the weight measured by the weighing unit; The accounting system according to claim 3 .

5. a product information storage unit that stores height information of each of the product groups; a height measuring unit that measures the height of the group of products being transported on the transport unit; Equipped with the error detection unit acquires the height information from the product information storage unit for the group of products read by the first product identification unit or the second product identification unit, and detects an error in which multiple products are overlapping when the acquired height information differs from the height measured by the height measurement unit; 5. The accounting system according to claim 3 or 4.

6. a plurality of accounting terminals that display the total price calculated by the accounting processing unit and through which purchasers of the group of products can make payments; the transport unit has a branch unit that transports the group of products to one of the plurality of checkout terminals downstream of the imaging area of ​​the imaging unit and the receiving range of the RFID reader, the accounting processing unit causes one of the accounting terminals selected by the branching unit to display the total price; The accounting system according to any one of claims 3 to 5.

7. providing a passing area that overlaps the imaging area of ​​the imaging unit and the receiving range of the RFID reader; the imaging unit and the RFID reader read information from the group of products that are passed through the passage area by a purchaser of the group of products; 3. The accounting system according to claim 1 or 2.

8. a first placement unit that places the group of products before they pass through the passage area and measures the weight of the placed group of products; a second placement unit that places the group of products after they have passed through the passage area and measures the weight of the placed group of products; Equipped with the error detection unit detects fraudulent behavior by a purchaser of the group of products when a weight decrease measured by the first placement unit differs from a weight increase measured by the second placement unit when the products are moved from the first placement unit to the second placement unit; The accounting system according to claim 7.

Images