Information sharing device and information sharing system

Abstract

The objective is to provide an information sharing device and information sharing system that can improve security in stores. [Solution] The information sharing device includes a control device that acquires information on fraudulent activities, including shoplifting and theft, at affected stores, and determines shared stores to share the information with based on the information and store information, including the location of the affected stores, and a communication device that transmits crime prevention information, including the information on the information on the affected stores.

Description

[Technical Field]

[0001] This invention relates to an information sharing device and information sharing system for sharing information on theft and other incidents that occur in stores. [Background technology]

[0002] Traditionally, shoplifting, where customers take goods without paying for them, has been a problem in stores such as convenience stores and supermarkets, and technologies have been proposed to detect and deter shoplifting. For example, Patent Document 1 proposes a device that analyzes video footage of the store where goods are placed, identifies the relationship between the goods in the video and the person purchasing them, and outputs an alert if the identified relationship is abnormal. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2024-89579 [Overview of the project] [Problems that the invention aims to solve]

[0004] Because shoplifting is a habitual crime, the same perpetrator may repeatedly commit shoplifting at multiple different stores. As described in Patent Document 1, conventionally, shoplifting was detected at each store, and security measures were implemented at each store or by the retailer managing each store. Therefore, security measures could not be implemented at stores or retailers where shoplifters had not visited.

[0005] The present invention aims to solve the above-mentioned problems and to provide an information sharing device and information sharing system that can improve security in stores. [Means for solving the problem]

[0006] The information sharing device according to the present invention comprises a control device that acquires information on fraudulent activities, including shoplifting and theft, at affected stores, and determines shared stores to share the information with based on the information and store information, including the location of the affected stores; and a communication device that transmits crime prevention information, including the information on the information on the affected stores, to the shared stores.

[0007] The information sharing system according to the present invention comprises the above-mentioned information sharing device, a first in-store management device that manages multiple POS terminals in the affected store, and a second in-store management device that manages multiple POS terminals in the shared store and receives and displays crime prevention information from the information sharing device. [Effects of the Invention]

[0008] According to the information sharing device and information sharing system of the present invention, security at stores can be improved by determining which stores share damage information and transmitting crime prevention information, including damage information, to these stores. [Brief explanation of the drawing]

[0009] [Figure 1] This is a schematic diagram of the information sharing system according to Embodiment 1. [Figure 2] This is a control block diagram of the information sharing system according to Embodiment 1. [Figure 3] This diagram illustrates the operation flow of the information sharing system according to Embodiment 1. [Figure 4] This figure illustrates an example of a method for determining a shared area in Embodiment 1. [Figure 5] This figure illustrates an example of a method for determining a shared area in Embodiment 1. [Figure 6] This figure illustrates an example of a method for determining a shared area in Embodiment 1. [Figure 7] This diagram illustrates the damage prediction in Embodiment 1. [Figure 8] This figure shows an example of security information displayed on the second in-store management device according to Embodiment 1. [Figure 9] This figure shows an example of security information displayed on the second in-store management device according to Embodiment 1. [Figure 10] This diagram illustrates the operation flow of the information sharing system according to Embodiment 2. [Figure 11] This diagram illustrates the flow of damage reporting in Embodiment 2. [Figure 12] This is a control block diagram of the first management server according to Embodiment 3. [Figure 13] This is a flowchart showing the flow of the fraud detection process in Embodiment 3. [Figure 14] This figure illustrates an example of a fraudulent pattern in Embodiment 3. [Figure 15] This figure illustrates an example of a fraudulent pattern in Embodiment 3. [Figure 16] This figure illustrates an example of a fraudulent pattern in Embodiment 3. [Figure 17] This figure illustrates an example of a fraudulent pattern in Embodiment 3. [Figure 18] This figure shows an example of security information displayed on the second in-store management device in a modified configuration. [Modes for carrying out the invention]

[0010] Hereinafter, an information sharing device 1 and an information sharing system 100 of the present invention will be described with reference to the drawings. In each drawing, the same or corresponding parts are denoted by the same reference numerals, and their descriptions are omitted or simplified as appropriate. Furthermore, the shape, size, and arrangement of the components shown in each drawing can be appropriately modified within the scope of the present invention.

[0011] Embodiment 1. FIG. 1 is a schematic configuration diagram of an information sharing system 100 according to Embodiment 1. The information sharing system 100 of the present embodiment is a system for sharing damage information related to theft such as shoplifting among a plurality of retailers A to C. In the following description, theft acts such as shoplifting may be referred to as illegal acts, and shoplifters and thieves may be referred to as illegal persons.

[0012] As shown in FIG. 1, retailer A has a plurality of stores S installed at different locations B and a head office H A that manages the plurality of stores S. A Retailer B has a plurality of stores S installed at different locations B and a head office H B that manages the plurality of stores S. B Retailer C has a plurality of stores S installed at different locations C and a head office H C that manages the plurality of stores S. C The business types, formats, and scales of each of the retailers A to C may be the same or different. Also, the business types, formats, and scales of each store S of retailer A A may be the same or different. Similarly, the business types, formats, and scales of each store S of retailer B B and the business types, formats, and scales of each store S of retailer C C may be the same or different, respectively.

[0013] The information sharing system 100 includes an information sharing device 1 for sharing information among retailers A to C. The information sharing device 1 collects damage information related to illegal acts including shoplifting and theft that occur in any of the stores S of retailers A to C A to S C and transmits it to any of the stores S of the other retailers A to C A to S C that may be affected by the theft for sharing. In the following, when damage occurs in store S of retailer A A , the case of sharing information with store S of retailer B B will be described as an example.

[0014] Figure 2 is a control block diagram of the information sharing system 100 according to Embodiment 1. Note that in Figure 2, among the configurations shown in Figure 1, the headquarters H of retailer A is shown. A and the headquarters of retailer B H B And one of the affected stores, store S A and one shared store S B The configuration shown is only an example of the above. As shown in Figure 2, the information sharing system 100 consists of an information sharing device 1 and a store S A The first in-store management device 21, multiple POS terminals 22, and multiple camera devices 23 are installed in the building, and Headquarters H A The first management server 25 is located at store S B A second in-store management device 31, multiple POS terminals 32, and multiple camera devices 33 are installed in the building, and headquarters H B The information sharing device 1 is equipped with a second management server 35 located at the same location. The information sharing device 1 is connected to the first in-store management device 21, the first management server 25, the second in-store management device 31, and the second management server 35 via a network such as the Internet, either by wired or wireless communication.

[0015] The information sharing device 1 comprises a control device 11, a storage device 12, and a communication device 13. The control device 11 consists of a dedicated processing circuit such as an ASIC or FPGA, a processor such as a CPU or DSP that executes a program stored in memory, or both. The control device 11 acquires information on theft from stores, determines which stores will share the acquired information, and transmits the information to the shared stores.

[0016] The storage device 12 is composed of, for example, non-volatile semiconductor memory such as ROM or flash memory, volatile semiconductor memory such as RAM, HDD, or SSD. The storage device 12 stores a security information database 121 and a store information database 122. In Figure 2, "database" is abbreviated as "DB". The storage device 12 also stores programs used for information sharing processing in the control device 11, various data such as parameters used for executing the program, and map information.

[0017] The crime prevention information database 121 contains information on multiple stores S included in the information sharing system 100. A ~S C The system stores crime prevention information, including details of past theft incidents. The crime information includes the date the theft was reported, the reporter, the store where the theft occurred, the time of the theft, the name, number, and value of the stolen items, the number, gender, and characteristics of the perpetrators, as well as images of the perpetrators entering and leaving the store, images of their fraudulent activities, and facial photographs. The crime prevention information includes the crime information and the sharing intensity and damage predictions described below.

[0018] The store information database 122 contains information for each store S included in the information sharing system 100. A ~S C Stores location information and business type information for each store. A ~S C Location information is for each store S A ~S C This is the address. Each store S A ~S C Industry information is available for each store S A ~S C This information pertains to the industry, business type, and scale of each retailer A to C. The control device 11 controls the headquarters H of each retailer A to C. A ~H C It communicates periodically with each store S in the store information database 122. A ~S C Update location and industry information.

[0019] The communication device 13 is a network interface that performs bidirectional communication with the first in-store management device 21, the first management server 25, the second in-store management device 31, and the second management server 35. For example, the communication device 13 receives damage information from the first in-store management device 21 and outputs it to the control device 11. The communication device 13 also transmits the security information created by the control device 11 to the second in-store management device 31.

[0020] The first in-store management device 21 is located in store S A This refers to an information terminal such as a tablet or smartphone operated by an employee of Store S.A The employees of store S use the first in-store management device 2. A The system manages multiple POS terminals 22 within the store, reports damage from theft such as shoplifting, displays crime prevention information shared by the information sharing device 1, and implements crime prevention measures. As shown in Figure 2, the first in-store management device 21 comprises a control device 211, an operation display device 212, a storage device 213, and a communication device 214.

[0021] The control device 211 controls each part of the first in-store management device 21. The control device 211 consists of a dedicated processing circuit such as an ASIC or FPGA, a processor such as a CPU or DSP that executes a program stored in memory, or both. The control device 211 creates damage information based on the information entered by the employee and the information stored in the storage device 213, and transmits it to the information sharing device 1.

[0022] The operation display device 212 is, for example, a touch panel. Employees operate the store S A If shoplifting or other illegal activity is detected within the facility, the operation display device 212 is operated to input the damage information and to instruct the information sharing device 1 to transmit it.

[0023] The storage device 213 is composed of, for example, a non-volatile semiconductor memory such as ROM or flash memory, a volatile semiconductor memory such as RAM, an HDD, or an SSD, and store S A The system stores information used for managing multiple POS terminals 22 installed in the system, product information registered in the multiple POS terminals 22, operation history information of the POS terminals 22, and image information captured by multiple camera devices 23.

[0024] POS terminal 22 is located in store S A This is a POS register that registers, processes, and settles payments for products purchased by the user (customer). POS terminal 22 is located in store S AThe POS terminal 22 may be operated by an employee, or it may be operated by a user as a self-service register. Furthermore, the POS terminal 22 may be fixed in the checkout area or attached to a shopping cart. Additionally, the POS terminal 22 may be a communication terminal such as a tablet or smartphone owned by the user, running an application program with POS register functionality. The POS terminal 22 reads product information from the product's barcode and registers the product. The product information read by the POS terminal 22 and the operation history information of the POS terminal 22 are transmitted to the first in-store management device 21 and the first management server 25.

[0025] The camera 23 is connected to the POS terminal 22, or to the store S A It is installed at the entrance, ceiling, wall, or shelf of the store S A This is a camera that takes pictures of users inside the store. Images taken by the camera 23 are transmitted to the first in-store management device 21 and the first management server 25. The images taken by the camera 23 may be either still images or videos.

[0026] The first management server 25 manages multiple stores S operated by retailer A. A Headquarters H, which oversees the operation. A Established in multiple stores S A It has the function of a POS server that centrally manages user information, product information, and store information. As shown in Figure 2, the first management server 25 has a control device 251, a storage device 252, and a communication device 253.

[0027] The control device 251 controls various parts of the first management server 25. The control device 251 consists of a dedicated processing circuit such as an ASIC or FPGA, a processor such as a CPU or DSP that executes a program stored in memory, or both. The control device 251 stores product information and operation history information received from the POS terminal 22, as well as image information received from the camera 23, in the storage device 252.

[0028] The storage device 252 is composed of, for example, non-volatile semiconductor memory such as ROM or flash memory, volatile semiconductor memory such as RAM, HDD, or SSD. The storage device 252 contains user information about users who use the POS terminal 22, and store S A Product information regarding the products handled by each store S A Store information related to the user. User information includes user identification information such as the user's name, age, and gender, user payment information, user's past operation history, and user's past purchase history. Product information includes product identification information such as product name, JAN code, and price. Store information is stored for each store S A This includes the type of business, business model, scale, and location information such as address. The storage device 252 also stores product information registered in multiple POS terminals 22, operation history information of the POS terminals 22, and image information captured by multiple camera devices 23.

[0029] The communication device 253 is a network interface that performs bidirectional communication with the information sharing device 1, the first in-store management device 21, the POS terminal 22, and the camera 23. The communication device 253 receives product information and operation history information from the POS terminal 22 and image information from the camera 23. The communication device 253 also transmits store information to the information sharing device 1.

[0030] The second in-store management device 31 is located in store S B This refers to an information terminal such as a tablet or smartphone operated by an employee of Store S. B The employees of store S use the second in-store management device 31. B The system manages multiple POS terminals 32 within the store, reports damage from theft such as shoplifting, displays crime prevention information shared by the information sharing device 1, and implements crime prevention measures. As shown in Figure 2, the second in-store management device 31 comprises a control device 311, an operation display device 312, a storage device 313, and a communication device 314.

[0031] The control device 311 controls various parts of the second in-store management device 31. The control device 311 consists of a dedicated processing circuit such as an ASIC or FPGA, a processor such as a CPU or DSP that executes a program stored in memory, or both. The control device 311 displays the security information received from the information sharing device 1 on the operation display device 312. The control device 311 also implements security measures based on the security information.

[0032] The operation display device 312 is, for example, a touch panel. The operation display device 312 displays security information received from the information sharing device 1. Employees also operate the operation display device 312 to give instructions regarding security measures.

[0033] The storage device 313 is composed of, for example, a non-volatile semiconductor memory such as ROM or flash memory, a volatile semiconductor memory such as RAM, an HDD, or an SSD, and the store S B The system stores information used for managing multiple POS terminals 32 installed in the system, product information registered in the multiple POS terminals 32, operation history information of the POS terminals 32, and image information captured by multiple camera devices 33.

[0034] POS terminal 32 is located in store S B This is a POS register that registers, processes, and settles payments for products purchased by the user (customer). POS terminal 32 is located in store S B The POS terminal 32 may be operated by an employee, or it may be operated by a user as a self-service register. Furthermore, the POS terminal 32 may be fixed in the checkout area or attached to a shopping cart. Additionally, the POS terminal 32 may be a communication terminal such as a tablet or smartphone owned by the user, running an application program with POS register functionality. The POS terminal 32 reads product information from the product's barcode and registers the product. The product information read by the POS terminal 32 and the operation history information of the POS terminal 32 are transmitted to the second in-store management device 31 and the second management server 35.

[0035] The camera 33 is connected to a POS terminal 32 or a store S B It is installed at the entrance, ceiling, wall, or shelf of the store S B This is a camera that takes pictures of users inside the store. The images taken by the camera 33 are transmitted to the second in-store management device 31 and the second management server 35. The images taken by the camera 33 may be either still images or videos.

[0036] The second management server 35 manages multiple stores S operated by retailer B. B Headquarters H, which oversees the operation. B Established in multiple stores S B It has the function of a POS server that centrally manages user information, product information, and store information. As shown in Figure 2, the second management server 35 has a control device 351, a storage device 352, and a communication device 353.

[0037] The control device 351 controls various parts of the second management server 35. The control device 351 consists of a dedicated processing circuit such as an ASIC or FPGA, a processor such as a CPU or DSP that executes a program stored in memory, or both. The control device 351 stores product information and operation history information received from the POS terminal 32, as well as image information received from the camera 33, in the storage device 352.

[0038] The storage device 352 is composed of, for example, non-volatile semiconductor memory such as ROM or flash memory, volatile semiconductor memory such as RAM, HDD, or SSD. The storage device 352 contains user information about users who use the POS terminal 32, and store S B Product information regarding the products handled by each store S B Store information related to the user. User information includes user identification information such as the user's name, age, and gender, user payment information, user's past operation history, and user's past purchase history. Product information includes product identification information such as product name, JAN code, and price. Store information is stored for each store S BThis includes the type of business, business model, scale, and location information such as address. The storage device 352 also stores product information registered in multiple POS terminals 32, operation history information of the POS terminals 32, and image information captured by multiple camera devices 33.

[0039] The communication device 353 is a network interface that performs bidirectional communication with the information sharing device 1, the second in-store management device 31, the POS terminal 32, and the camera 33. The communication device 353 receives product information and operation history information from the POS terminal 32 and image information from the camera 33. The communication device 353 also transmits store information to the information sharing device 1.

[0040] Figure 3 is a diagram illustrating the operation flow of the information sharing system 100 according to Embodiment 1. Here, we will explain the operation when an employee, after noticing suspicious behavior by a user, discovers shoplifting or other fraudulent activity after the fact (after the fraudulent person has left the store) based on the person's image and the operation history of the POS terminal 22. First, store S A When an employee discovers fraudulent activity by a fraudulent individual, the first in-store management device 21 creates information about the fraudulent activity (S1). Here, the employee first operates the operation display device 212 of the first in-store management device 21 and inputs information such as the date and time of reporting, the reporter, the affected store, the date the incident occurred, the time of the incident, the name of the affected product, the number of items and the amount, the number of fraudulent individuals, their gender and characteristics, etc. The control device 211 creates information about the fraud by linking the information entered by the employee with images of the fraudulent individuals entering and leaving the store, images of the fraudulent activity, and facial photographs taken by the camera 23. The created information about the fraud is transmitted to the information sharing device 1 via the communication device 214 of the first in-store management device 21 (S2).

[0041] When the control device 11 of the information sharing device 1 receives damage information via the communication device 13, it determines a shared area for sharing the damage information based on the received damage information and the store information contained in the store information database 122 stored in the storage device 12 (S3). For example, damaged store S AIf the store is located in Kyushu, there is little need to share information with the store located in Hokkaido. Therefore, the control device 11 will not share information with the affected store S A The area where the theft occurred or the range of the theft perpetrator's movements is estimated and designated as a shared area.

[0042] The determination of the shared area will be explained with reference to Figures 4 to 6. Figures 4 to 6 illustrate an example of the method for determining the shared area in Embodiment 1. First, the control device 11 of the information sharing device 1 determines the shared area according to the size of the affected store. For example, many users visit small stores such as convenience stores on foot or by bicycle, and the range of movement of users of small stores is considered to be relatively narrow. On the other hand, many users visit large stores such as shopping malls by car, and the range of movement of users of large stores is considered to be wider than that of users of small stores. Therefore, for example, if the affected store is a large store, the shared area will be a radius of 30 km centered on the affected store; if the affected store is a medium-sized store, the shared area will be a radius of 15 km centered on the affected store; and if the affected store is a small store, the shared area will be a radius of 5 km centered on the affected store. The radius of each area is just an example, and the larger the affected store, the larger the shared area should be set. In the example of Figure 4, affected store S1 is a medium-sized store such as a supermarket, and affected store S2 is a small store such as a convenience store. Therefore, the sharing area R1 for damage information of the affected store S1 is larger than the sharing area R2 for the affected store S2.

[0043] Furthermore, the control device 11 of the information sharing device 1 may designate the area between the shared areas of multiple affected stores as a shared area. For example, if the same person commits multiple acts of fraud at different stores, there is a possibility that other stores located between the store where the first act of fraud occurred and the store where the second act of fraud occurred may also be affected. Therefore, the control device 11 may designate the area around the road connecting the store where the first act of fraud occurred and the store where the second act of fraud occurred as a shared area. For example, as shown in Figure 5, if a crime occurs at affected store S2, and prior to that, a crime was committed at store S1 by the same person, then the following areas are determined as shared areas: area R1 with a predetermined radius centered on affected store S1, area R2 with a predetermined radius centered on affected store S2, and area R3 at a predetermined distance (e.g., 5 km) on both sides of the road connecting the store where the first act of fraud occurred and the store where the second act of fraud occurred.

[0044] Furthermore, the control device 11 of the information sharing device 1 may designate an area where the fraudulent person is estimated to move based on multiple pieces of damage information as a shared area. For example, if the same fraudulent person causes multiple consecutive incidents of damage at different stores, the movement route and direction of the fraudulent person can be estimated from the store where the first incident occurred and the store where the second incident occurred, and it is considered that damage may occur at other stores located in the estimated direction of movement. Therefore, the control device 11 may designate a movement prediction area estimated from the store where the first incident occurred and the store where the second incident occurred as a shared area. For example, as shown in Figure 6, if the first incident occurs at store S2 and the second incident occurs at store S1 on the same day, an area R2 with a predetermined radius centered on the first incident store S2, an area R1 with a predetermined radius centered on the second incident store S1, and a movement prediction area R4 estimated from the direction from the first incident store S2 to the second incident store S1 are determined as shared areas. The predicted movement area R4 is defined, for example, by identifying the direction of movement from the first damaged store S2 to the second damaged store S1, and is an area within a predetermined distance (e.g., 5 km) from the damaged store S1 in the direction of movement along the road connecting the damaged stores S2 and S1, and is an area within a predetermined distance (e.g., 5 km) on both sides of the road.

[0045] Returning to Figure 3, the control device 11 of the information sharing device 1 determines the shared stores from among the stores located within the shared area based on the business type information of the affected store (S4). For example, if the affected product is a home appliance, there is little need to share the damage information with convenience stores of different business types. Therefore, the control device 11 determines the stores within the shared area that are of the same business type or business format as the affected store as the shared stores, based on the damage information and the store information stored in the store information database 122. Alternatively, the control device 11 may determine the shared business type from the type of affected product included in the damage information and determine the stores within the shared area that belong to the shared business type as the shared stores. For example, if the affected product is a home appliance, the control device 11 may determine the stores within the shared area that handle home appliances as the shared stores.

[0046] Next, the control device 11 determines the sharing intensity of the damage information based on the damage information (S5). The sharing intensity can be, for example, "caution" or "vigilance," which is stronger than "caution." The sharing intensity is determined based on at least one of the number of perpetrators, the amount of damage, and the number of times the damage has occurred. The number of times the damage has occurred is obtained from past crime prevention information stored in the crime prevention information database 121. For example, if there is one perpetrator, it is considered shoplifting, so the sharing intensity is set to "caution," and if there are multiple perpetrators, it is considered a theft ring, so the sharing intensity is set to "vigilance," which is stronger than "caution." Alternatively, even if there is only one perpetrator, if the damage has occurred multiple times, or if the amount of damage is high (for example, 5,000 yen or more), the sharing intensity may be set to "vigilance." Note that the sharing intensity is not limited to two levels, "caution" and "vigilance," but may be three or more levels, and may be expressed numerically.

[0047] Furthermore, the control device 11 stores information about the affected stores S stored in the crime prevention information database 121. A Based on past crime prevention information, damage predictions are created (S6). Figure 7 is a diagram illustrating the damage prediction in Embodiment 1. The control device 11 uses the crime prevention information database 121 to store damage information for the past month for stores S A Based on crime prevention information, damage predictions will be made to identify the days and times when shoplifting is most likely to occur. Note that damage predictions are not limited to those based on crime prevention information from the past month; they may also be based on crime prevention information from any period, such as the past three months.

[0048] First, the control device 11 detects the victimized store S included in the crime information for the past month stored in the crime information database 121. AFrom the date and time of damage occurrence, the day of the week with the most damage occurrences is identified and designated as the "specified day of the week." Then, the time period during which damage occurs on the specified day of the week is identified. In the example in Figure 7, the specified day of the week with the most damage occurrences is Saturday, and the time period during which damage occurs is from 13:10 to 20:30. The control device 11 then determines the "time period of concern" from the time period during which damage occurred on the specified day of the week. The time period of concern is the time period during which damage occurs most frequently. For example, if multiple times during which damage occurs follow a normal distribution, the range of ±1σ (68.2%) is considered the time period of concern. In the example in Figure 7, for example, 15:00 to 16:30 is considered the time period of concern.

[0049] Furthermore, if the control device 11 can identify the fraudulent person's visit history from the fraudulent person's membership number and prepaid card number, it may predict the damage based on the fraudulent person's visit history. In this case, the control device 11 obtains the fraudulent person's visit dates and times for the past month from the first management server 25 and designates the days of the week with the most visits as specific days. Then, it identifies the visit times on the specific days of the week and identifies the times of day that require special attention from the visit times.

[0050] Returning to Figure 3, the control device 11 controls the affected store S A Store S, a store that shares damage information, sharing intensity, and damage predictions as crime prevention information, receives damage information from the store. B The second in-store management device 31 is transmitted (S7). The control device 11 then controls the shared store S B The security information transmitted is stored in the security information database 121 of the storage device 12 (S8).

[0051] Shared store S BThe second in-store management device 31 receives security information from the information sharing device 1 and displays it on the operation display device 312 (S9). Figures 8 and 9 show examples of security information displayed on the second in-store management device 31 according to Embodiment 1. As shown in Figures 8 and 9, the operation display device 312 displays the person who reported the damage and the date of the report, the damaged store, the date of the damage, the time of the damage, the damaged goods, the amount of damage, the number of perpetrators, images of the perpetrators entering and leaving the store, images of the perpetrators during the fraudulent activity, and facial photographs, as well as the sharing intensity and damage prediction. Figure 8 is an example of security information display when there is one perpetrator and the sharing intensity is "caution," and Figure 9 is an example of security information display when there are multiple theft rings and the sharing intensity is "warning." As shown in Figures 8 and 9, the method of displaying the sharing intensity differs depending on the sharing intensity.

[0052] It should be noted that the method of displaying security information is not limited to the examples in Figures 8 and 9. For example, the color of the text or background on the security information screen may be different for "caution" and "warning," or the shared access strength indicator may be made to flash. By displaying security information differently according to the shared access strength in this way, it is possible to alert employees of the shared store.

[0053] Returning to Figure 3, the shared store S that received the security information. B Then, security measures are implemented (S10). Security measures include, for example, store S B This could include increasing the number of employees, raising awareness among employees, or strengthening surveillance cameras. Strengthening surveillance cameras would involve increasing the number of cameras activated and their operating time. This would allow store S to... A Store S B In this regard, countermeasures can be taken in advance.

[0054] As described above, in the information sharing system 100 of this embodiment, the store S where the damage occurred A and different retailer stores S B In this case, the affected store S A Damage information can be shared in store S. BEven if no suspicious person has visited the store, preventive measures can be implemented in advance, improving the security of store S B and enhancing the security of the entire area by sharing information across business boundaries. Furthermore, by determining the information sharing targets based on damage information and store information, unnecessary measures due to redundant information sharing can be avoided, suppressing an increase in the burden on employees.

[0055] In the above, information sharing when a suspicious person leaves the store without being secured inside the damaged store has been described. However, even when a suspicious person is secured inside the damaged store, similar information sharing may be performed.

[0056] Embodiment 2. The information sharing system 100 of Embodiment 2 will be described. The information sharing system 100 of Embodiment 2 differs from Embodiment 1 in the method of reporting damage at the damaged store. The configuration of the information sharing system 100 is the same as that of Embodiment 1. However, it is assumed that the first in-store management device 21 of this embodiment includes a camera for reading two-dimensional codes such as QR codes (registered trademarks).

[0057] FIG. 10 is a diagram for explaining the operation flow of the information sharing system 100 according to Embodiment 2. FIG. 11 is a diagram for explaining the damage reporting flow in Embodiment 2. Referring to FIGS. 10 and 11, the damage reporting flow of the information sharing system 100 according to Embodiment 2 will be described. In this embodiment, the first in-store management device 21 and the first management server 25 cooperate to report damage.

[0058] First, when an employee of store S A detects suspicious behavior by a user, the employee uses the first in-store management device 21 to read a two-dimensional code for reporting to headquarters H A . Suspicious behavior refers to behavior that may indicate shoplifting by the user. The two-dimensional code is for store S AThe codes may be affixed to walls or shelves, or to manuals or employee ID cards carried by employees. The codes read by the first in-store management device 21 are not limited to two-dimensional codes, but may also be one-dimensional codes such as barcodes.

[0059] As shown in Figure 11(a), when the two-dimensional code is read by the first in-store management device 21, a shoplifting report button for creating a report email is displayed on the operation display device 212 of the first in-store management device 21, as shown in Figure 11(b). When an employee presses the shoplifting report button, the report email format shown in Figure 11(c) is displayed on the operation display device 212 of the first in-store management device 21. As shown in Figure 11(c), the report email format includes the recipient, subject, and report content fields. The recipient is the first management server 25, and the subject is "Shoplifting Report," which are entered by default. The report content includes, for example, the store name, sender name, suspicious person information, location of occurrence, date and time of occurrence, and remarks. The suspicious person information is identification information such as the membership number or prepaid card number entered when the person who committed the suspicious act logged into the POS terminal 22.

[0060] Returning to Figure 10, the employee operates the operation display device 212 of the first in-store management device 21 and enters the report content into the report email format, thereby creating the report email (S22). Then, when the employee gives the instruction to send the report email, the report email is sent to headquarters H A It is sent to the first management server 25 (S23).

[0061] The control device 251 of the first management server 25 communicates with store S via the communication device 253. A Upon receiving a report email, the system verifies whether or not actual fraudulent activity occurred based on the report (S24). Here, it is determined whether or not shoplifting occurred based on the report content, the image information of the suspicious person stored in the storage device 252, and the operation history information of the suspicious person. This verification is carried out by Headquarters H AIt may be visually performed by employees or the like, or may be automatically performed by the first management server 25 or other information processing devices.

[0062] And when fraud is confirmed (S24: YES), the control device 251 creates damage information to be transmitted to the information sharing device 1 based on the report content (S25). Specifically, the control device 251 identifies the reporter and report date of the damage, and the damaged store, damage date, and damage time zone included in the damage information from the report content. Further, the control device 251 identifies the damaged goods and the damage amount from the report content and the product information stored in the storage device 252. Furthermore, the control device 251 extracts images of when the unauthorized person enters and exits the store, images of the fraud, and face photos from the image information stored in the storage device 252 based on the report content, and identifies the number of unauthorized persons.

[0063] The control device 251 of the first management server 25 transmits the created damage information to the information sharing device 1 (S26). The subsequent information sharing process in the information sharing device 1 is the same as in the first embodiment. If fraud is not confirmed (S24: NO), the first management server 25 does not create damage information or transmit it to the information sharing device 1. Note that the first management server 25 may transmit the confirmation result of fraud to the first in-store management device 21, and if fraud is confirmed, the first in-store management device 21 may implement security measures.

[0064] As described above, in the information sharing system 100 of the present embodiment, in addition to the effects of the first embodiment, A the fraud in the store S can be promptly transmitted to the headquarters H, A and the workload of the employees in the store S can be reduced. Also, by confirming whether shoplifting or other fraud actually occurs at the headquarters H, A it is possible to avoid on-site risks such as misidentifying fraud by employees in the store S, and reduce the stress of on-site employees. A A

[0065] Embodiment 3. ​​The information sharing system 100 of Embodiment 3 will now be described. The information sharing system 100 of Embodiment 3 differs from Embodiments 1 and 2 in its method for detecting fraudulent activity at affected stores.

[0066] Shoplifting and other fraudulent activities in stores can be detected by employees actually witnessing the fraudulent activity, or by employees reviewing recorded images of suspicious individuals after witnessing suspicious behavior. Conventional technologies have also been proposed that analyze images of suspicious individuals to detect suspicious or fraudulent activities based on the movements of the suspicious person or merchandise. However, because the methods of shoplifting and other fraudulent activities are diverse, conventional technologies may not always be able to detect fraudulent activities. The information sharing system 100 implemented in this project performs fraud detection processing that can automatically identify a variety of fraudulent activities.

[0067] The fraud detection process is performed by a fraud detection device. The fraud detection device may be the control device 211 of the first in-store management device 21, the control device 251 of the first management server 25, or any other information processing device. The following description will focus on the case where the first management server 25 is the fraud detection device that performs the fraud detection process.

[0068] Figure 12 is a control block diagram of the first management server 25 according to Embodiment 3. The functions and configurations of the first in-store management device 21, POS terminal 22, and camera 23 in this embodiment are the same as in Embodiment 2. The functions and configurations of the information sharing device 1, second in-store management device 31, POS terminal 32, camera 33, and second management server 35 in this embodiment are the same as in Embodiment 1 and are therefore not shown. The configuration of the second management server 35 may be the same as that of the first management server 25 in this embodiment.

[0069] As shown in Figure 12, the first management server 25 includes a control device 251, a storage device 252, and a communication device 253. The configuration and functions of the communication device 253 are the same as in Embodiment 1. In addition to the functions of Embodiment 1, the control device 251 in this embodiment performs fraud detection processing based on information stored in the storage device 42.

[0070] The storage device 252 in this embodiment stores an image information database 51, an operation history information database 52, and a judgment model 53. In Figure 12, "database" is abbreviated as "DB".

[0071] The image information database 51 contains information on multiple stores S managed by the first management server 25. A The system stores image information transmitted from multiple camera devices 23 installed in the store. The image information includes images of the user taken by the camera devices 23, the date and time the images were taken, identification information of the store where the camera devices 23 are installed, and identification information of the POS terminal 22 being photographed.

[0072] The operation history information database 52 contains information on multiple stores S managed by the first management server 25. A The system stores operation history information received from multiple POS terminals 22 installed in the store. The operation history information database 52 stores multiple operation histories for a single transaction at the POS terminal 22 as a single operation history entry. The operation history includes identification information of the POS terminal 22 on which the operation was performed, identification information of the store where the POS terminal 22 is installed, user login information, registration information such as the department (category), product code, product name, number of items, and price of the registered product, cancellation information such as the department (category), product code, product name, number of items, and price of the product whose registration was canceled, and the date and time each operation was performed. User login information is identification information such as a membership number or prepaid card number.

[0073] The judgment model 53 is a pre-trained model that takes the operation history information and image information of a suspicious person as input data and outputs a judgment result of whether or not the action is fraudulent. The judgment model 53 is an AI model that has been trained through supervised learning using an image recognition algorithm such as a convolutional neural network (CNN) and a large number of images of past fraudulent acts that have been accumulated.

[0074] Figure 13 is a flowchart showing the flow of the fraud detection process in Embodiment 3. Below, we will explain the case in which the first management server 25 detects fraud when a user operates the POS terminal 22, which is a self-checkout device, to make a payment. First, similar to Embodiment 2, store S A When an employee discovers suspicious activity by a customer, the employee uses the first in-store management device 21 to send a report email to the first management server 25. When the control device 251 of the first management server 25 receives the report email from the first in-store management device 21, it performs fraud detection processing on the reported suspicious person.

[0075] First, the control device 251 extracts the operation history information of the suspicious person from the storage device 252 (S31). Here, operation history information corresponding to the suspicious person information (such as a membership number or prepaid card number) included in the report is extracted from the operation history information database 52.

[0076] Next, the control device 251 extracts image information of the POS terminal 22 corresponding to the operation history information from the storage device 252 (S32). Here, the control device 251 first identifies the date and time when the operation was performed by a suspicious person and the identification information of the POS terminal 22 based on the operation history information. Then, the control device 251 extracts image information from the image information database 51 that corresponds to the identified date and time and the identification information of the POS terminal 22.

[0077] The control device 251 then inputs the extracted operation history information and image information into the judgment model 53 and obtains a judgment result as output (S33). This allows the control device 251 to automatically determine whether or not a suspicious person has committed an illegal act such as shoplifting.

[0078] Next, the training of the judgment model 53 will be explained. The judgment model 53 extracts suspicious persons and products from image information and detects the movement of the suspicious person, the state of the product, and the movement of the product. Then, it determines whether or not something is fraudulent based on at least one of the movement of the suspicious person, the state of the product, and the movement of the product, and the operation content included in the operation history information. Figures 14 to 17 are diagrams illustrating examples of fraudulent patterns in Embodiment 3. The judgment model 53 is trained to determine something is fraudulent when it detects the fraudulent patterns shown in Figures 14 to 17 from the input operation history information and image information.

[0079] First, as shown in Figure 14, if an item in an image moves from the unscanned area T1 through the POS terminal 22 to the scanned area T2, and there is no scan history of the item in the operation history information within 5 seconds of that movement (i.e., the item is not registered), it is determined to be fraudulent. "Passing through the POS terminal 22" means passing in front of the scanner on the POS terminal 22 that reads the item's barcode. This makes it possible to detect fraudulent activities such as scanning an item at a distance from the scanner on the POS terminal 22, scanning the side of the item that does not have a barcode, or covering the item's barcode with a finger while scanning.

[0080] Furthermore, as shown in Figure 15, if multiple items in an image are moved from the unscanned area T1 to the scanned area T2 via the POS terminal 22 while stacked or held in both hands, and the operation history information within 3 seconds of the movement shows only a scan history for one item, it is determined to be fraudulent. This makes it possible to detect fraudulent activities such as scanning only one item while stacking multiple items, or scanning only one item while moving multiple items simultaneously.

[0081] Furthermore, as shown in Figure 16, if a bundled product moves from the unscanned area T1 through the POS terminal 22 to the scanned area T2 in the image, and within 5 seconds of that movement, the operation history information only shows a scan history for individual products, it is determined to be fraudulent. A bundled product is a product sold as a set of multiple products of the same type or multiple products of different types, such as the 6-pack of beverages shown in Figure 16. Bundled products have a barcode indicating the price of the bundled product displayed on the tape or box that bundles the multiple products together. In addition, each product included in the bundled product also has a barcode indicating the price of the individual product. The above fraud detection makes it possible to detect fraudulent activity such as scanning the barcodes of individual products despite them being bundled products.

[0082] Furthermore, as shown in Figure 17, if an item in an image is moved from the scanned area T2 to the front of the POS terminal 22, and then moved back to the scanned area, and the operation in the operation history information before the second movement of the item is an item cancellation operation, and there is no scan history of the canceled item within 5 seconds of the second movement of the item, it is determined to be fraudulent. This makes it possible to detect fraud such as taking an item home without returning it, even though the item was canceled at the time of payment for reasons such as insufficient money.

[0083] As described above, the information sharing system 100 of this embodiment can improve the accuracy of fraud detection by using a judgment model 53 that has learned various fraudulent methods to determine whether fraud has occurred. This can deter fraudulent activities such as shoplifting, further improve security, and reduce the number of personnel required to perform fraud detection.

[0084] The above describes the embodiments, but the present invention is not limited to the above embodiments, and can be modified or combined in various ways without departing from the spirit of the present invention. For example, in the above embodiments, the information sharing device 1 was described as an independent device, but the management server of any of retailers A to C may have the functions of the information sharing device 1. Furthermore, the affiliation of each function of each device is not limited to the examples of the above embodiments. For example, some or all of the databases in the information sharing device 1 or the first management server 25 may be stored in an external device such as online storage.

[0085] Furthermore, in the above embodiment, the control device 11 of the information sharing device 1 determined the shared area according to the size of the affected stores, but it is not limited to this. For example, the shared area may be arbitrarily set by the administrator using the information sharing device 1. Specifically, the information sharing device 1 may be equipped with a display device such as a liquid crystal display and an operating device such as a mouse, and a map including the affected stores may be displayed on the display device, allowing the administrator to select the shared area by operating the operating device. Alternatively, as shown in Figures 4 to 6, the display device may show a map including the affected stores and the determined shared area, allowing the administrator to arbitrarily change the shared area.

[0086] Alternatively, the control device 11 may designate a shared area with a predetermined radius centered on the affected store, regardless of the size of the affected store. The predetermined radius may be a fixed value set in advance, or it may be set according to the location of the affected store. For example, if the affected store is in a sparsely populated area, the predetermined radius may be larger than if the affected store is in an urban area. Furthermore, as in Embodiment 1 above, the shared area may be determined according to the size of the affected store, and then the shared area may be adjusted according to the location of the affected store.

[0087] Furthermore, the control device 11 may determine the shared area according to the alert level based on the damage information. For example, if the sharing intensity is "Alert," the shared area will be a radius of 30 km centered on the damaged store, and if the sharing intensity is "Caution," the shared area will be a radius of 10 km centered on the damaged store. The radius of each area is just an example, and the shared area should be set to be larger as the sharing intensity increases.

[0088] Furthermore, the information sharing device 1 shares information about the shared area as security information with the shared store S B It may be sent to shared store S. B The second in-store management device 31 may display a shared area screen on the operation display device 312. Figure 18 shows an example of security information displayed on the second in-store management device 31 according to a modified example. As shown in Figure 18, the second in-store management device 31 displays information about the victimized store S A And, shared store S B A map including the shared area on the map and the sharing level are displayed. This allows shared stores S B Among the employees of the affected store S A It becomes easy to understand the positional relationship between objects.

[0089] Furthermore, the information sharing device 1 shares security information with the store S B In addition to sending to shared store S B Headquarters H of retailer B to which it belongs B It may also be sent to the second management server 35 located at [location]. This will enable multiple stores S belonging to retailer B to be transmitted. B Damage information can be shared at headquarters H. B From the second management server 35, shared store S B They can instruct the authorities to take crime prevention measures.

[0090] Furthermore, in the above embodiment 3, the judgment model 53 is used to determine fraud when a user operates the POS terminal 22, which is a self-checkout device, to make a payment. However, the judgment method used by the judgment model 53 is not limited to this. For example, the judgment model 53 may take image information from the sales floor as input data and output a result indicating fraud. For example, the judgment model 53 may be trained to determine fraud when a user takes unpaid items and leaves the store without paying, puts items in their pocket or shopping bag on the sales floor, or takes unpaid items into the restroom.

[0091] Furthermore, while the information sharing device 1 in the above embodiment is configured to transmit damage information, sharing strength, and damage prediction as security information to the sharing store, it is sufficient if it transmits at least damage information. In this case, the information sharing device 1 does not need to determine the sharing strength or create the damage prediction. Damage prediction may be performed not by the information sharing device 1, but by the management server of the headquarters to which the affected store belongs. [Explanation of symbols]

[0092] 1 Information sharing device, 11 Control device, 12 Storage device, 13 Communication device, 21 First in-store management device, 22 POS terminal, 23 Camera, 25 First management server, 31 Second in-store management device, 32 POS terminal, 33 Camera, 35 Second management server, 42 Storage device, 51 Image information database, 52 Operation history information database, 53 Judgment model, 100 Information sharing system, 121 Crime prevention information database, 122 Store information database, 211 Control device, 212 Operation display device, 213 Storage device, 214 Communication device, 251 Control device, 252 Storage device, 253 Communication device, 311 Control device, 312 Operation display device, 313 Storage device, 314 Communication device, 351 Control device, 352 Storage device, 353 Communication device.

Claims

1. A control device that acquires information on fraudulent activities, including shoplifting and theft, at affected stores, and determines shared stores to share the information with based on the said information and store information, including the location information of the affected stores. An information sharing device comprising: a communication device that transmits crime prevention information, including the aforementioned damage information, to the aforementioned shared store.

2. The control device determines a shared area for sharing the damage information based on the size of the affected stores, and determines the stores within the shared area as the shared stores. The aforementioned shared area is an area with a predetermined radius that includes the affected store. The information sharing device according to claim 1, wherein the predetermined radius is larger the larger the scale of the affected store.

3. The control device is Based on the number of individuals who committed the fraudulent acts or the amount of damages included in the aforementioned damage information, the sharing strength of the aforementioned damage information is determined. Based on the aforementioned sharing strength, a shared area for sharing the damage information is determined, and stores within the shared area are designated as the shared stores. The aforementioned shared area is an area with a predetermined radius that includes the affected store. The information sharing device according to claim 1, wherein the predetermined radius is larger the higher the sharing strength.

4. The control device is If information regarding the damage caused by the same fraudulent individual is received from multiple affected stores, The information sharing device according to claim 1, which determines an area between multiple affected stores, or a movement prediction area estimated from multiple affected stores, as a shared area, and determines stores within the shared area as the shared stores.

5. The device further includes a memory device that stores multiple past incident reports, The information sharing device according to claim 1, wherein the control device creates a damage prediction for the affected store based on a plurality of damage information stored in the storage device and transmits it to the shared store as crime prevention information.

6. The aforementioned store information includes the type of business and business format of the affected store, The information sharing device according to any one of claims 2 to 4, wherein the control device determines as the shared store a store of the same type or business as the damaged store in the shared area, or a store of the same type as the business corresponding to the damaged goods included in the damage information in the shared area.

7. An information sharing device according to any one of claims 1 to 5, A first in-store management device that manages multiple POS terminals in the aforementioned affected store, An information sharing system comprising: a second in-store management device that manages multiple POS terminals in the shared store and receives and displays the security information from the information sharing device.

8. The information sharing system according to claim 7, wherein the second in-store management device displays at least one of the damage information, the sharing strength, the damage prediction, and the shared area as the security information.

9. The system further includes a first management server for managing the aforementioned affected stores, The first in-store management device reads a one-dimensional code or a two-dimensional code installed in the affected store and reports the fraudulent activity to the first management server. The information sharing system according to claim 7, wherein the first management server, upon receiving the damage report, creates the damage information and transmits it to the information sharing device.

10. The information sharing system according to claim 7, further comprising a fraud detection device that determines whether or not fraudulent activity occurred at the victimized store using a trained determination model that takes the operation history information of the fraudulent person's POS terminal at the victimized store and image information of the fraudulent person as input data and outputs whether or not fraudulent activity occurred.

11. The information sharing system according to claim 10, wherein the determination model determines whether or not fraudulent activity has occurred based on a combination of the state or movement of the product in the image information and the operation content of the operation history information.

12. The aforementioned determination model is If the product in the image information moves from the unscanned area to the scanned area by passing through the POS terminal, and there is no scan history of the product in the operation history information within a predetermined time period from the time of the move, In the aforementioned image information, if multiple products are stacked or held in both hands and moved from the unscanned area through the POS terminal to the scanned area, and the operation history information within a predetermined time period from that movement contains only a scan history for one product, In the aforementioned image information, if a bundled product moves from the unscanned area through the POS terminal to the scanned area, and within a predetermined time period from that movement, the operation history information only contains a scan history for a single product, or If, in the aforementioned image information, the product is moved from the scanned area to in front of the POS terminal, and then moved back to the scanned area, and the operation in the operation history information prior to the second movement of the product is a product cancellation operation, and there is no scan history of the canceled product within a predetermined time from the second movement of the product, The information sharing system according to claim 11, which has been trained to determine that the aforementioned fraudulent activity has occurred.

Images