Suspicious person detection device, suspicious person detection method, and program

Abstract

The present invention provides a technique for reducing the processing load in detection of a suspicious person. This suspicious person detection device is provided with: a position information acquisition means for acquiring position information indicating the position of a cart used by a person in a store; a determination means for determining that the movement of the cart is suspicious on the basis of the time-series data of the position information; a vital data acquisition means for acquiring vital data of a person using the cart when it is determined that the movement of the cart is suspicious; a determination means for determining whether the person is a suspicious person on the basis of the vital data; and notification means for performing notification when it is determined that the person is a suspicious person.

Description

Suspicious person detection device, suspicious person detection method, and recording medium

[0001] The present disclosure relates to a suspicious person detection device, a suspicious person detection method, and a recording medium.

[0002] There are technologies that use vital signs and video footage to detect suspicious individuals. For example, there are technologies that detect a person as suspicious based on their mental state analyzed from vibrations detected in video footage. There are also technologies that detect a person as suspicious if their behavior or actions are suspicious in video footage.

[0003] Patent document 1 describes a suspicious person alarm notification system that issues a report when a crime index calculated based on vital information based on radar signals and movement information such as a person's posture and direction of gaze based on skeletal information of each person shown in an image is high.

[0004] International Publication No. 2022 / 003447

[0005] However, continually identifying suspicious individuals based on the vital signs of all people within the monitored area can place a heavy processing load on the system.

[0006] An example of an objective of the present disclosure is to provide a technology that reduces the processing load when detecting a suspicious person.

[0007] A suspicious person detection device in one aspect of the present disclosure includes a location information acquisition means for acquiring location information indicating the location of a cart used by a person in a store, a determination means for determining whether the movement of the cart is suspicious based on time series data of the location information, a vital data acquisition means for acquiring vital data of the person using the cart if the movement of the cart is determined to be suspicious, a determination means for determining whether the person is suspicious based on the vital data, and a notification means for issuing a notification if the person is determined to be suspicious.

[0008] In one aspect of the present disclosure, a suspicious person detection method involves a computer acquiring location information indicating the location of a cart used by a person in a store, determining that the movement of the cart is suspicious based on time series data of the location information, and if the movement of the cart is determined to be suspicious, acquiring vital data of the person using the cart, determining whether the person is suspicious based on the vital data, and if the person is determined to be suspicious, issuing a notification.

[0009] A suspicious person detection program in one aspect of the present disclosure causes a computer to acquire location information indicating the location of a cart used by a person in a store, determine whether the movement of the cart is suspicious based on time series data of the location information, acquire vital data of the person using the cart if the movement of the cart is determined to be suspicious, determine whether the person is suspicious based on the vital data, and execute a process to issue a notification if the person is determined to be suspicious.

[0010] Each program may be stored in a non-transitory computer-readable recording medium.

[0011] One example of the effect of the present disclosure is that the processing load can be reduced when detecting a suspicious person.

[0012] FIG. 1 is a diagram showing an overview of a system including a suspicious person detection device. FIG. 2 is a block diagram showing the functional configuration of the suspicious person detection device. FIG. 3 is a diagram showing an example of an output screen for a terminal device used by a store employee. FIG. 4 is a diagram showing an example of an output screen for a terminal device used by a person determined to be a suspicious person. FIG. 5 is a diagram showing another example of an output screen for a terminal device used by a person determined to be a suspicious person. FIG. 6 is a flowchart showing the operation of the suspicious person detection device. FIG. 7 is a diagram showing a hardware configuration for realizing the suspicious person detection device of the present disclosure using a computer and its peripheral devices.

[0013] Embodiments of the present disclosure will be described in detail with reference to the drawings.

[0014] First Embodiment Fig. 1 is a diagram illustrating an overview of a system including a suspicious person detection device 100 according to the present disclosure. The suspicious person detection device 100 detects suspicious persons in a store and issues a notification to prevent fraudulent activities by the suspicious person. In the present disclosure, the suspicious person detection device 100 is primarily intended to detect people who visit a store, i.e., customers of the store. The suspicious person detection device 100 is communicably connected to a terminal device and an output device.

[0015] In FIG. 1 , the terminal device is a terminal device provided in a cart used by a person visiting a store. The terminal device is, for example, a tablet or a smartphone. The terminal device has at least a function for acquiring vital data of the person using the cart. For example, the terminal device may be a terminal device used by the person using the cart to scan products in self-scanning shopping, where the person shops around the store while registering the products they purchase. The terminal device may be provided in the cart or may be detachably installed on the cart. Note that the cart may be a shopping cart. Alternatively, a basket used for self-scanning shopping may be used instead of a cart.

[0016] The output device outputs a notification from the suspicious person detection device 100. The output device may be at least one of a display device and an audio output device. For example, the output device may be at least one of a signage provided in the store, a speaker provided in the store, a terminal device used by a store clerk, and a terminal device used by a customer of the store.

[0017] Next, the configuration of the suspicious person detection device 100 according to the embodiment will be described.

[0018] 2 is a block diagram showing the configuration of the suspicious person detection device 100 according to the first embodiment. Referring to FIG. 2, the suspicious person detection device 100 includes a location information acquisition unit 101, a determination unit 102, a vital data acquisition unit 103, a judgment unit 104, and a notification unit 105.

[0019] In FIG. 2 , the location information acquisition unit 101 is an example of a location information acquisition means that acquires location information indicating the location of a cart used by a person in a store. The location information is information indicating the location of each cart in the store. The location information acquisition unit 101 only needs to acquire location information of at least a cart used by a customer of the store among the carts present in the store. A cart used by a customer of the store may be a cart for which self-scan shopping is being performed. Furthermore, a moving cart may be considered to be a cart used by a customer of the store.

[0020] For example, the location information acquisition unit 101 acquires location information of a terminal device used in self-scan shopping. In this case, the terminal device is a terminal device provided in a cart or a terminal device detachably installed in the cart. The location information acquisition unit 101 may acquire location information of the terminal device acquired by, for example, a beacon installed in a store performing short-range communication with the terminal device.

[0021] Furthermore, for example, the location information acquisition unit 101 may acquire location information based on video captured by a camera such as a surveillance camera within a store. For example, the location information acquisition unit 101 may identify the location of the cart from the position of the cart detected from the video within the video capture range and the position and orientation within the store of the camera that captured the video.

[0022] The method of acquiring location information by the location information acquisition unit 101 is not limited to these examples.

[0023] The location information acquisition unit 101 may acquire location information of the cart over time. At this time, the location information acquisition unit 101 may store the acquired location information of the cart as time-series data for each cart in a storage unit (not shown). Furthermore, the location information acquisition unit 101 may periodically acquire data for a certain period of time or longer, in which the location information of the cart is associated with time, as time-series data.

[0024] The determination unit 102 is an example of a determination means for determining whether cart movement is suspicious based on time-series data of cart position information. The determination unit 102 may determine whether the time-series data of cart position information meets a condition for determining that cart movement is suspicious. For example, the determination unit 102 may determine that cart movement is suspicious if the cart movement falls outside the range of statistical values ​​of customer behavior in a store. In this case, cart movement determined to be suspicious is the cart's movement line, cart movement characteristics, etc., identified from the time-series data of cart position information. The cart movement line represents the trajectory, frequency, and direction of a person's movement over time. The cart movement line may be any line that at least indicates the amount of time a person stays in an appropriately set area. In other words, the cart movement line and cart movement characteristics may be the cart's speed, cart movement characteristics, etc. The determination unit 102 may also determine that cart movement is suspicious using a trained model. The trained model will be described later.

[0025] For example, the determination unit 102 determines that the movement of a cart is suspicious based on the cart's movement path identified from time-series data of the cart's location information. For example, the determination unit 102 may determine that the movement of a cart is suspicious if the cart's movement path includes a specific area. Here, the specific area is an area in a store where fraudulent activity is likely to occur. The area where fraudulent activity is likely to occur may be an area where fraudulent activity has actually occurred or an area that generally meets conditions for fraudulent activity. The area where fraudulent activity is likely to occur may be an area where products that are likely to be used for fraudulent activity are placed. Products that are likely to be used for fraudulent activity may be, for example, high-priced products or popular products that are in short supply. Furthermore, the area that generally meets conditions for fraudulent activity may be a blind spot in a store. The blind spot in a store may be at least one of an area that is blind spot of a surveillance camera and an area that is likely to be blind spot of a store employee. Furthermore, the area that generally meets conditions for fraudulent activity may be an area in a store where customers do not normally enter. Areas in a store that customers normally do not enter are, for example, the back of the store or an area where cashiers operate the cash registers.

[0026] For example, the determination unit 102 determines that the movement of the cart is suspicious when the time-series data of the cart's location information indicates that the cart has entered a specific area. The determination unit 102 may also determine that the movement of the cart is suspicious when the time-series data of the cart's location information indicates that the cart has stayed in the specific area for a predetermined period of time or longer. The fact that the cart has entered the specific area and that the cart has stayed in the specific area for a predetermined period of time or longer may be determined from the movement path based on the time-series data of the cart's location information. The predetermined period of time for determining that the cart has stayed in the specific area may be set to be longer than the length of time that a person who is not committing fraud might stay in a store when he or she is unsure about what to purchase. The predetermined period of time for determining that the cart has stayed in the specific area may be, for example, 3 minutes, 5 minutes, or the like.

[0027] Furthermore, for example, the determination unit 102 may determine that the movement of the cart is suspicious if the time-series data of the cart's location information indicates that the cart has entered and exited a specific area in the store a predetermined number of times or more. The fact that the cart has entered and exited a specific area in the store a predetermined number of times or more may be identified from the movement path based on the time-series data of the cart's location information. In this case, the predetermined number may be set to three, five, or the like, which is greater than the number of times a non-fraudulent person would go back and forth between other locations when unsure about purchasing an item in the store.

[0028] Furthermore, the determination unit 102 may determine that the movement of the cart is suspicious if the time-series data of the cart's location information indicates that the cart went directly from the vicinity of the store entrance to a specific area in the store. The fact that the cart went directly from the vicinity of the store entrance to a specific area in the store may be identified from the movement path based on the time-series data of the cart's location information. For example, in a store such as a supermarket, it is expected that people will move around the store. In contrast, it is expected that a person who uses the store with the intention of committing fraudulent acts on products in a specific area will go directly to the specific area without moving around the store. Therefore, the determination unit 102 may determine that the movement of the cart is suspicious if it indicates that the cart went directly from the vicinity of the store entrance to a specific area in the store.

[0029] Predetermined cart movements in relation to a particular area that will determine the cart movements as suspicious may be set for each particular area.

[0030] Furthermore, the determination unit 102 may determine whether the movement characteristics of the person using the cart are suspicious based on the time-series data of the cart's position information. For example, the determination unit 102 may determine that the movement of the cart is suspicious if the time-series data of the cart's position information indicates that the cart's movement speed is outside a predetermined range. For example, the determination unit 102 may determine that the movement of the cart is suspicious if the time-series data of the cart's position information indicates that the cart is meandering.

[0031] Furthermore, as described above, the determination unit 102 may determine that the movement of the cart is suspicious using a trained model. The trained model is a trained model that has learned the relationship between time-series data of the position information of a cart used by a person and information indicating whether the person has committed fraud. The determination unit 102 determines whether the movement of the cart is suspicious by inputting the time-series data of the cart position information into the trained model.

[0032] The vital data acquisition unit 103 is an example of a vital data acquisition means that acquires vital data of the person using the cart when the determination unit 102 determines that the movement of the cart is suspicious. The vital data of the person may be at least one of physiological information of the person, information representing the person's internal state, and information representing the person's mental state. The vital data may be, for example, physiological information such as heart rate, respiratory rate, body temperature, and pupil dilation. Furthermore, for example, the vital data may be a numerical representation of the person's internal state or mental state, such as stress level, tension, or aggression, estimated using various physiological information.

[0033] The vital data acquiring unit 103 may acquire vital data measured by a sensor. The sensor may be, for example, a sensor provided in the terminal device or a sensor communicatively connected to the terminal device. For example, the vital data acquiring unit 103 may acquire vital data measured by a sensor provided in a wearable device connected to the terminal device. Alternatively, the vital data acquiring unit 103 may acquire vital data detected by a sensor communicatively connected to the terminal device and provided in a handle of a cart used to carry merchandise in a store.

[0034] The vital data acquisition unit 103 may also acquire vital data estimated based on video. In this case, the video may be video of a person captured by a camera of a terminal device installed on the cart. The vital data acquisition unit 103 may acquire video from a camera provided in the terminal device, such as a tablet or smartphone, and extract vital data from the video. For example, the vital data acquisition unit 103 may also acquire vital data extracted from the video by the terminal device. The video-based vital data is extracted by analyzing changes in facial color, facial movement, gaze, blinking, pupil diameter, iris movement, facial expression, and the like included in the facial video of the person captured by the camera.

[0035] As described above, when the location information acquisition unit 101 acquires the location information of the cart based on the video, the vital data acquisition unit 103 identifies the terminal device corresponding to the location information of the cart. For example, the vital data acquisition unit 103 may identify the terminal device from the location information of the cart by comparing a facial image of the person using the terminal device taken by the terminal device with a facial image of the person included in an image taken by a camera for identifying location information, such as a surveillance camera.

[0036] The vital data acquisition unit 103 may acquire vital data of the person from the terminal device when the determination unit 102 determines that the movement of the cart is suspicious. Furthermore, the vital data acquisition unit 103 may cause the terminal device to execute an operation for detecting vital data of the person when the determination unit 102 determines that the movement of the cart is suspicious. For example, when the determination unit 102 determines that the movement of the cart is suspicious, the vital data acquisition unit 103 may cause the terminal device to start capturing video of the person's face for detecting vital data.

[0037] The vital data acquisition unit 103 may acquire vital data at the time when the determination unit 102 determines that the person is suspicious. Alternatively, the vital data acquisition unit 103 may acquire time-series data of the vital data. For example, the vital data acquisition unit 103 may acquire vital data for a predetermined period up to the time when the determination unit 102 determines that the person is suspicious. Alternatively, for example, the vital data acquisition unit 103 may acquire vital data over time from the time when the determination unit 102 determines that the person is suspicious.

[0038] The determination unit 104 is an example of a determination means that determines whether a person is a suspicious person based on the person's vital data acquired by the vital data acquisition unit 103. The determination unit 104 determines that the person is a suspicious person when the person's vital data indicates that the person is in a mental state that suggests the person is about to commit a fraudulent act.

[0039] Here, fraudulent acts include criminal acts, violence, intimidation, etc. For example, fraudulent acts in a store include shoplifting, robbery, and damage to merchandise. Generally, when a person commits fraud or is about to commit fraud, the person experiences emotional changes such as excitement, increased aggression, increased tension, and increased stress. These emotional changes can be reflected in vital data. For example, changes in vital data when a person commits fraud or is about to commit fraud include increased heart rate, increased skin temperature, increased respiratory rate, and dilated pupils.

[0040] Therefore, the determination unit 104 may determine that a person is suspicious if the value indicated by the vital data exceeds a predetermined threshold or if the amount of change in the value indicated by the vital data exceeds a predetermined threshold. Furthermore, for example, the determination unit 104 may determine that a person is suspicious if the change in the vital data is similar to the change in vital data observed when committing fraud. Here, similarity may mean that the difference is equal to or less than a certain value or that the similarity is greater than a threshold. The determination unit 104 is not limited to these examples and may determine that a person is suspicious under conditions appropriately determined depending on the vital data used.

[0041] The notification unit 105 is an example of a notification means that issues a notification when a person is determined to be a suspicious person. Specifically, the notification unit 105 causes an output device to issue a notification. As described above, the notification unit 105 issues a notification to prevent fraudulent acts by suspicious people. The notification to prevent fraudulent acts by suspicious people may be, for example, a notification to prevent fraudulent acts by attracting the attention of the person determined to be suspicious. Furthermore, the notification to prevent fraudulent acts by suspicious people may be a notification to inform store employees of information about the person determined to be suspicious, so that the store employees can take measures to prevent fraudulent acts. The notification unit 105 may issue at least one or more of the above notifications.

[0042] The content of the notification that prevents fraudulent activities by attracting the attention of a person determined to be suspicious is not particularly limited. For example, the notification that prevents fraudulent activities by attracting the attention of a person determined to be suspicious may be content that regulates fraudulent activities, such as "Shoplifting is a crime," "Security guards on patrol," or "Watch out for suspicious persons." Furthermore, the notification that prevents fraudulent activities by attracting the attention of a person determined to be suspicious may be content unrelated to fraudulent activities, such as a store advertisement or a notification to initiate an inquiry to a store clerk, such as "Can I help you?" By the notification unit 105 issuing a notification with content unrelated to fraudulent activities, it is possible to prevent a person determined to be suspicious from feeling suspicious even if the person is not attempting to commit a fraudulent activity, thereby preventing a decrease in customer satisfaction at the store and preventing the person from becoming increasingly aggressive from being provoked.

[0043] The notification of information about a person determined to be suspicious so that store employees can take measures to prevent fraudulent activity may include information that allows the store employees to recognize the suspicious person or their location. For example, the notification in this case may include information indicating at least one of the suspicious person's physical characteristics and the suspicious person's location within the store. The notification may also include information indicating at least one of the person's vital data, which is the reason the person was determined to be suspicious, and the movement of the cart used by the person.

[0044] The device that outputs the notification by the notification unit 105 may be at least one of a terminal device used by a person, an output device installed in a store, and a terminal device used by an employee of the store.

[0045] For example, when a person is determined to be a suspicious person, the notification unit 105 causes the output device to issue a notification by voice. In this case, the voice is a voice that attracts the attention of the person determined to be a suspicious person. By outputting a voice that attracts the attention of the person from the output device, if the person is attempting to commit a fraudulent act, the person's attention can be distracted and the fraudulent act can be prevented. In this case, the notification unit 105 causes a speaker used for announcements in the store or a speaker of a terminal device used by the person to output a notification voice. Furthermore, for example, the notification voice when a person is determined to be a suspicious person may be a voice notification that notifies a store employee that a suspicious person is present in the store. In this case, the notification unit 105 causes a speaker used for announcements in the store or an audio output device of a terminal device used by the employee to output a notification voice.

[0046] Furthermore, for example, when a person is determined to be a suspicious person, the notification unit 105 notifies the person by displaying a message. In this case, the message is a message that attracts the person's attention to the person determined to be a suspicious person. By displaying a message that attracts the person's attention, the output device can distract the person when the person is attempting to commit a fraudulent act and prevent the fraudulent act. In this case, the notification unit 105 notifies the person by displaying a message on a display unit of a terminal device used by the person or a display device such as signage installed in the store. Furthermore, for example, when a person is determined to be a suspicious person, the notification may be a message that notifies a store employee that a suspicious person is present in the store. In this case, the notification unit 105 notifies the employee by displaying a message on a tablet terminal, smartphone, POS (Point of Sales) terminal, or other display device used by the employee.

[0047] FIG. 3 is an example of a notification output screen used by a store employee. In FIG. 3, the output screen includes information indicating that a suspicious individual has been detected, the suspicious individual's location within the store, and an image of the suspicious individual's face. Also, in FIG. 3, the alert output screen includes an "Alert" button. For example, when the "Alert" button is selected on the alert output screen for employees, the notification unit 105 may output an alert informing other employees or security guards that a suspicious individual has been detected. Furthermore, for example, when the above button is selected, the notification unit 105 may output an alert to the person detected as a suspicious individual. In this way, allowing employees to decide whether or not to issue an alert at their own discretion may prevent the employee from accidentally alerting someone who is unlikely to commit fraud or from provoking someone who is becoming increasingly aggressive.

[0048] FIG. 4 illustrates an example of a notification output screen displayed on a terminal device used by a person determined to be suspicious. In FIG. 4 , the output screen includes a self-scanning shopping screen with a message asking, "Can we help you?" Displaying this message along with a notification sound may be effective in attracting the person's attention and preventing fraudulent activity. As described above, the output screen may include a message prohibiting fraudulent activity, such as "No shoplifting," in addition to a message that does not mention fraudulent activity, such as "Can we help you?" Note that the self-scanning shopping screen includes a list of registered items for purchase and the total price of the currently purchased items, but the self-scanning shopping screen is not limited to this example. In the example of FIG. 4 , the notification is displayed on a portion of the screen of the person's terminal device, but the notification may also be displayed on the entire screen of the person's terminal device.

[0049] FIG. 5 shows another example of a notification output screen displayed on a terminal device used by a person determined to be suspicious. In FIG. 5 , the output screen includes messages such as "No shoplifting," "Surveillance cameras in operation," and "Please speak to a store employee if you need assistance" to attract the person's attention and prevent fraudulent activity. Also, in FIG. 5 , the output screen includes a button for returning to the self-scan shopping screen. For example, the terminal device may return the display from the output screen of FIG. 5 to the self-scan shopping screen when the button for returning to the self-scan shopping screen is pressed or when a new product is registered.

[0050] The above describes examples of notifications when a person is determined to be suspicious, but these notifications may be combined as appropriate. The notification method may also be changed as appropriate depending on the movement of the person's cart and the person's vital data. For example, if the vital data indicates that the person is becoming increasingly aggressive, a notification may be sent to a store clerk to avoid directly provoking the person.

[0051] The operation of the suspicious person detection device 100 configured as above will be described with reference to the flowchart of FIG.

[0052] As shown in FIG. 6, first, the location information acquisition unit 101 acquires location information of a cart used by a person (step S101).

[0053] Next, the determination unit 102 determines whether the movement of the cart is suspicious based on the time-series data of the cart's position information (step S102). If it is determined in step S102 that the movement of the cart is not suspicious based on the time-series data of the cart's position information (step S102: No), the suspicious person detection device 100 ends the series of operations. Note that even if it is determined in step S102 that the movement of the cart is not suspicious based on the time-series data of the cart's position information, if a person is using the cart, the suspicious person detection device 100 may return to the operation of step S101 again.

[0054] In step S102, if the movement of the cart based on the time series data of the cart's location information is determined to be suspicious (step S102: Yes), the vital data acquisition unit 103 acquires the vital data of the person using the cart (step S103).

[0055] Next, the determination unit 104 determines whether the person is a suspicious person based on the vital sign data acquired in step S103 (step S104). If it is determined in step S104 that the person is not a suspicious person (step S104: No), the suspicious person detection device 100 ends the series of operations. Note that, as in step S102, even if it is determined in step S104 that the person is not a suspicious person, if the person is using a cart, the suspicious person detection device 100 may return to the operation of step S101 again.

[0056] In step S104, if the person is determined to be a suspicious person (step S104: Yes), the notification unit 105 issues a notification to prevent fraudulent acts by the suspicious person.

[0057] This completes the series of operations of the suspicious person detection device 100.

[0058] In the suspicious person detection device 100 of the present embodiment described above, the determination unit 102 determines whether the movement of the cart is suspicious based on time-series data of the cart's location information. Then, the vital data acquisition unit 103 acquires vital data of the person using the cart if the determination unit 102 determines that the movement of the cart is suspicious. Then, the determination unit 104 determines whether the person is suspicious based on the person's vital data, and the notification unit 105 issues a notification if the person is determined to be suspicious.

[0059] As a result, the suspicious person detection device 100 of this embodiment can reduce the processing load when detecting a suspicious person. For example, when detecting a suspicious person based on vital data, the suspicious person detection device 100 of this embodiment uses location information of a cart used by the person and the movement of the cart identified from the location information to narrow down the target of monitoring vital data for detecting a suspicious person. As a result, the suspicious person detection device 100 of this embodiment does not need to monitor the vital data of all people who are targets of suspicious person detection. Therefore, the suspicious person detection device 100 of this embodiment can reduce the processing load for detecting a suspicious person based on vital data.

[0060] [Hardware Configuration] Some or all of the components of each device or system in each embodiment of the present disclosure described above are realized by any combination of an information processing device 1000 and a program, for example, as shown in Fig. 7. The information processing device 1000 includes, as an example, the following configuration.

[0061] - CPU (Central Processing Unit) 1001 - ROM (Read Only Memory) 1002 - RAM (Random Access Memory) 1003 - Program 1004 loaded into RAM 1003 - Storage device 1005 for storing program 1004 - Drive device 1007 for reading and writing from and to recording medium 1006 - Communication I / F 1008 for connecting to communication network 1009 - Input / output I / F 1010 for inputting and outputting data - Bus 1011 for connecting each component. Note that I / F is an abbreviation for Interface.

[0062] Each component of each device or system in each embodiment is realized by the CPU 1001 acquiring and executing a program that realizes the function of that component. The program that realizes the function of each component of each device is stored in the storage device 1005 or RAM 1003 in advance, for example, and is read by the CPU 1001 as needed. The program 1004 may be supplied to the CPU 1001 via a communication network, or may be stored in the recording medium 1006 in advance, and the drive device 1007 may read the program and supply it to the CPU 1001.

[0063] There are various variations in the method of realizing each device. For example, each device or system may be realized by any combination of a separate information processing device 1000 and a program for each component. Furthermore, multiple components included in each device may be realized by any combination of a single information processing device 1000 and a program.

[0064] Furthermore, some or all of the components of each device or system may be realized by general-purpose or dedicated circuits (circuitry) including a processor or the like, or a combination of these. Examples of circuits include a CPU, a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), and an AI (Artificial Intelligence) processing LSI (Large Scale Integration). These may be configured by a single chip or by multiple chips connected via a bus. Some or all of the components of each device may be realized by a combination of the above-mentioned circuits and a program.

[0065] When some or all of the components of each device or system are realized by multiple information processing devices, circuits, etc., the multiple information processing devices, circuits, etc. may be centrally or decentralized. For example, the information processing devices, circuits, etc. may be realized as a client-server system, a cloud computing system, or the like, in which they are connected via a communication network.

[0066] Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

[0067] Furthermore, although the operations are described in a sequential order in the form of a flowchart, the order of description does not limit the order in which the operations are performed. Therefore, when implementing each embodiment, the order of the operations may be changed as long as it does not affect the content.

[0068] Some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0069] [Supplementary Note 1] A suspicious person detection device comprising: a location information acquisition means for acquiring location information indicating the location of a cart used by a person in a store; a determination means for determining whether the movement of the cart is suspicious based on time series data of the location information; a vital data acquisition means for acquiring vital data of the person using the cart if the movement of the cart is determined to be suspicious; a determination means for determining whether the person is a suspicious person based on the vital data; and a notification means for issuing a notification if the person is determined to be a suspicious person.

[0070] [Supplementary Note 2] The suspicious person detection device according to Supplementary Note 1, wherein the determining means determines that movement of the cart is suspicious based on a line of movement of the cart identified based on time-series data of the position information.

[0071] [Supplementary Note 3] The suspicious person detection device according to Supplementary Note 2, wherein the determination means determines that the movement of the cart is suspicious when the time-series data of the location information indicates that the cart has entered a specific area in the store.

[0072] [Appendix 4] The suspicious person detection device described in Appendix 2 or 3, wherein the judgment means judges that the movement of the cart is suspicious if the time series data of the location information indicates that the cart has stayed in a specific area of ​​the store for more than a predetermined time.

[0073] [Appendix 5] The suspicious person detection device described in any of Appendices 2 to 4, wherein the judgment means judges that the movement of the cart is suspicious if the time series data of the location information indicates that the cart has entered and exited a specific area in the store a predetermined number of times or more.

[0074] [Appendix 6] The suspicious person detection device described in any of Appendices 2 to 5, wherein the judgment means judges that the movement of the cart is suspicious if the time series data of the location information indicates that the cart went directly from around the entrance of the store to a specific area in the store.

[0075] [Supplementary Note 7] The suspicious person detection device according to any one of Supplementary Notes 3 to 6, wherein the specific area is an area where fraudulent acts are often committed.

[0076] [Supplementary Note 8] The suspicious person detection device according to any one of Supplementary Notes 3 to 7, wherein the specific area is a blind spot in the store.

[0077] [Supplementary Note 9] The suspicious person detection device according to any one of Supplementary Notes 1 to 8, wherein the determination means determines that the movement of the cart is suspicious based on characteristics of the movement of the cart identified based on time-series data of the position information.

[0078] [Supplementary Note 10] The suspicious person detection device according to Supplementary Note 9, wherein the determination means determines that the movement of the cart is suspicious when the time-series data of the position information indicates that the movement speed of the cart is outside a predetermined range.

[0079] [Supplementary Note 11] The suspicious person detection device according to Supplementary Note 9 or 10, wherein the determination means determines that the movement of the cart is suspicious when the time-series data of the position information indicates that the cart is meandering.

[0080] [Supplementary Note 12] The suspicious person detection device described in any one of Supplementary Notes 1 to 11, wherein the judgment means judges that the movement of the cart is suspicious by inputting the time series data of the location information of the cart used by a person into a trained model that has learned the relationship between the time series data of the location information of the cart and information indicating whether the person has committed fraud.

[0081] [Supplementary Note 13] The suspicious person detection device according to any one of Supplementary Notes 1 to 12, wherein the vital data acquisition means acquires the vital data of the person from an image of the person captured by a camera of a terminal device installed on the cart.

[0082] [Supplementary Note 14] The suspicious person detection device according to any one of Supplementary Notes 1 to 13, wherein the notification means, when it is determined that the person is a suspicious person, notifies a terminal device used by the person.

[0083] [Supplementary Note 15] The suspicious person detection device according to any one of Supplementary Notes 1 to 14, wherein the notification means issues a notification by voice when the person is determined to be a suspicious person.

[0084] [Supplementary Note 16] The suspicious person detection device described in any one of Supplementary Notes 1 to 15, wherein the notification means notifies the location information and the vital data to a terminal device used by an employee of the store when the person is determined to be a suspicious person.

[0085] [Supplementary Note 17] The suspicious person detection device according to any one of Supplementary Notes 1 to 16, wherein the determination means determines that the person is a suspicious person when the vital data indicates that the person is excited.

[0086] [Supplementary Note 18] The suspicious person detection device according to any one of Supplementary Notes 1 to 17, wherein the determination means determines that the person is a suspicious person when the vital data indicates that the person is becoming increasingly aggressive.

[0087] [Supplementary Note 19] A suspicious person detection method, in which a computer acquires location information indicating the location of a cart used by a person in a store, determines that the movement of the cart is suspicious based on time-series data of the location information, acquires vital data of the person using the cart if the movement of the cart is determined to be suspicious, determines whether the person is a suspicious person based on the vital data, and issues a notification if the person is determined to be a suspicious person.

[0088] [Supplementary Note 20] A recording medium that records a program that causes a computer to execute the following processes: acquiring location information indicating the location of a cart used by a person in a store; determining whether the movement of the cart is suspicious based on time-series data of the location information; acquiring vital data of the person using the cart if the movement of the cart is determined to be suspicious; determining whether the person is a suspicious person based on the vital data; and issuing a notification if the person is determined to be a suspicious person.

[0089] REFERENCE SIGNS LIST 100 Suspicious person detection device 101 Location information acquisition unit 102 Determination unit 103 Vital data acquisition unit 104 Determination unit 105 Notification unit 1000 Information processing device 1001 CPU 1002 ROM 1003 RAM 1004 Program 1005 Storage device 1006 Recording medium 1007 Drive device 1008 Communication I / F 1009 Communication network 1010 Input / output I / F 1011 Bus

Claims

1. a location information acquisition means for acquiring location information indicating the location of a cart used by a person in a store; a determination means for determining whether the movement of the cart is suspicious based on the time-series data of the position information; a vital data acquisition means for acquiring vital data of a person using the cart when the movement of the cart is determined to be suspicious; a determination means for determining whether the person is a suspicious person based on the vital data; a notification means for issuing a notification when the person is determined to be a suspicious person; A suspicious person detection device comprising:

2. The determining means determines that the movement of the cart is suspicious based on the movement line of the cart identified based on the time series data of the position information. The suspicious person detection device according to claim 1 .

3. The determination means determines that the movement of the cart is suspicious based on characteristics of the movement of the cart identified based on the time-series data of the position information. The suspicious person detection device according to claim 1 .

4. The determination means determines that the movement of the cart is suspicious by inputting the time series data of the location information into a trained model that has learned the relationship between the time series data of the location information of the cart used by a person and information indicating whether the person has committed fraud. The suspicious person detection device according to claim 1 .

5. The vital data acquisition means acquires the vital data of the person from an image of the person captured by a camera of a terminal device installed on the cart. The suspicious person detection device according to claim 1 .

6. When the person is determined to be a suspicious person, the notification means notifies the terminal device used by the person. The suspicious person detection device according to claim 1 .

7. The determining means determines that the person is suspicious if the vital data indicates that the person is excited. The suspicious person detection device according to claim 1 .

8. The determining means determines that the person is suspicious if the vital data indicates that the person is becoming more aggressive. The suspicious person detection device according to any one of claims 1 to 7.

9. The computer Acquire location information indicating the location of a cart used by a person in a store; determining that the movement of the cart is suspicious based on the time series data of the location information; If the movement of the cart is determined to be suspicious, vital data of the person using the cart is acquired; determining whether the person is a suspicious person based on the vital data; If the person is determined to be suspicious, a notification is sent. Method for detecting suspicious persons.

10. Acquire location information indicating the location of a cart used by a person in a store; determining that the movement of the cart is suspicious based on the time series data of the location information; If the movement of the cart is determined to be suspicious, vital data of the person using the cart is acquired; determining whether the person is a suspicious person based on the vital data; If the person is determined to be suspicious, a notification is sent. A program that causes a computer to perform a process.