Fraud detection device, fraud detection method, and program

Abstract

The present invention provides a technique for reducing erroneous detection in detecting a fraudulent act. This fraud detection device is provided with: an acquisition means for acquiring first vital data, of a person detected to have started a specific motion related to execution of a fraudulent act, at a certain point in time during which the specific motion is performed, and second vital data, of the person, at the time when the specific motion is completed; a determination means for determining whether a change between the first vital data and the second vital data matches a change in vital data at the time when a fraudulent act has been performed; and a notification means for notifying, in accordance with the result of the determination, that there is a possibility that the fraudulent act has been performed.

Description

FRAUD DETECTION DEVICE, FRAUD DETECTION METHOD, AND RECORDING MEDIUM

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

[0002] There are technologies that detect and report suspicious individuals who may be committing crimes or other illegal acts. For example, there are technologies that detect and report suspicious individuals by detecting specific behaviors or combinations of specific behaviors by individuals. There are also technologies that detect and report suspicious individuals by using vibration patterns emitted by individuals or thermal imaging data of individuals.

[0003] Patent Document 1 describes a suspicious person notification system that calculates a stress index from vital data such as brain waves, cerebral blood flow, pulse waves, blood pressure, respiratory rate, body temperature, and sweat rate, presumes that a person with a stress index higher than a reference value is a suspicious person, and issues a warning if the suspicious person enters a specific area.

[0004] Japanese Patent Application Laid-Open No. 2018-037075

[0005] However, the suspicious person reporting system described in Patent Document 1 may erroneously assume that a person whose stress index is high for reasons other than committing fraud is a suspicious person.

[0006] An example of an objective of the present disclosure is to provide a technology for reducing false positives in detecting fraudulent activities.

[0007] A fraud detection device in one aspect of the present disclosure includes an acquisition means for acquiring first vital data at a certain point during the specific action of a person in which the start of a specific action related to the commission of fraudulent activity has been detected, and second vital data of the person at the time the specific action ends; a determination means for determining whether a change from the first vital data to the second vital data corresponds to a change in vital data that would occur if fraudulent activity were committed; and a notification means for notifying the person that fraudulent activity may have been committed based on the result of the determination.

[0008] In one aspect of the fraud detection method of the present disclosure, a computer acquires first vital data at a certain point during the specific action of a person in which the start of a specific action related to the commission of fraudulent activity has been detected, and second vital data of the person at the time the specific action ends, determines whether changes in the first vital data to the second vital data correspond to changes in vital data that would occur if fraudulent activity were committed, and notifies the person that fraudulent activity may have been committed based on the results of the determination.

[0009] A fraud detection program in one aspect of the present disclosure acquires first vital data at a certain point during the specific action of a person in which the start of a specific action related to the commission of fraudulent activity has been detected, and second vital data of the person at the time the specific action ends, determines whether changes in the first vital data to the second vital data correspond to changes in vital data that would occur if fraudulent activity were committed, and, depending on the results of the determination, causes a computer to execute a process of notifying the person that fraudulent activity may have been committed.

[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 it is possible to reduce false positives in detecting fraudulent activities.

[0012] FIG. 1 is a diagram showing an example of an outline of a system including a fraud detection device. FIG. 2 is a block diagram showing the functional configuration of the fraud detection device. FIG. 3 is a diagram showing an example of a notification screen for a terminal device used by a store employee. FIG. 4 is a flowchart showing an example of the operation of the fraud detection device. FIG. 5 is a diagram showing another example of an outline of a system including a fraud detection device. FIG. 6 is a diagram showing another example of an outline of a system including a fraud detection device. FIG. 7 is a flowchart showing another example of the operation of the fraud detection device. FIG. 8 is a diagram showing a hardware configuration for realizing the fraud detection device of the present disclosure using a computer and its peripheral devices.

[0013] [Embodiment] FIG. 1 is a diagram illustrating an example of a system including a fraud detection device 100 according to the present disclosure. The fraud detection device 100 issues a notification when fraudulent activity is likely to have occurred based on a person's vital data. In the present disclosure, the fraud detection device 100 uses vital data from at least two points in time associated with a specific behavior associated with the commission of fraudulent activity to determine whether the changes in vital data correspond to changes in vital data that would occur if fraudulent activity had occurred. Then, the fraud detection device 100 issues a notification that fraudulent activity is likely to have occurred based on the results of the determination.

[0014] Here, the vital data of a person may be at least one of physiological information of the person and information representing the person's internal state. The physiological information of a person includes, but is not limited to, heart rate, respiratory rate, body temperature, pupil dilation, etc. The information representing the person's internal state is a quantification of the person's internal state, such as stress level, tension level, and aggression. The person's internal state may also be referred to as the person's mental state. The information representing the person's internal state may be estimated, for example, using various physiological information or based on the person's facial expression captured in an image of the person. Furthermore, the vital data used for fraud detection by the fraud detection device 100 may be a combination of multiple vital data.

[0015] Furthermore, fraudulent acts are acts that do not comply with norms such as laws. Examples of fraudulent acts include criminal acts, violence, and intimidation. For example, fraudulent acts in stores include shoplifting, robbery, fraudulent accounting using counterfeit bills or fake coupons, and damage to merchandise. Note that fraudulent acts in this disclosure are not limited to fraudulent acts committed in stores, but also refer to fraudulent acts committed in various other facilities, such as homes, apartments, buildings, companies, factories, warehouses, plazas, parks, and public restrooms, or in a single room within a facility.

[0016] Generally, when a person commits or is about to commit a fraudulent act, the person experiences emotional changes such as excitement, increased aggression, increased tension, and increased stress. Such emotional changes may be reflected in vital data. For example, changes in vital data when a person commits or is about to commit a fraudulent act include an increased heart rate, increased skin temperature, increased respiratory rate, and dilated pupils. Specific behaviors associated with the commission of fraudulent acts will be described below.

[0017] The fraud detection device 100 is communicably connected to a measuring device for measuring vital data and an output device for outputting notifications.

[0018] In FIG. 1 , the measurement device for measuring vital data is not particularly limited. For example, it may be a measurement system that measures vital data from video of a person captured by a camera. In this case, the camera may be a surveillance camera installed in a facility or a camera of a terminal device attached to a cart used in a store to carry products and register purchased products. The measurement system may acquire video of a person from the camera and measure vital data from the video. The video-based vital data is measured by analyzing changes in facial color, facial movement, gaze, blinking, pupil diameter, iris movement, facial expression, and the like contained in the facial video of the person captured by the camera. For example, the measurement system may measure vital data from video of a person used to detect specific movements, as described below. Furthermore, for example, the measurement system may identify and track a person and measure vital data from the captured video.

[0019] Furthermore, for example, the measuring device for measuring vital data is a sensor. The sensor is, for example, a sensor provided in a terminal device or a sensor communicatively connected to the terminal device. Here, the terminal device may be, for example, a smartphone, a tablet, a smartwatch, or smart glasses. The terminal device may be a terminal device owned by a person or a terminal device used by a person in a facility or the like. Furthermore, for example, the terminal device may be a terminal device such as a tablet provided in a cart used in a store to carry products and register purchased products. In this case, the sensor may be, for example, a sensor provided in the handle of the cart used to carry products in a store, which measures vital data when a person touches the handle.

[0020] 1 , the output device outputs a notification from the fraud 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 a facility such as a store, a speaker provided in the facility, and a terminal device used by employees or customers of the facility.

[0021] Next, the configuration of the fraud detection device 100 according to the embodiment will be described.

[0022] 2 is a block diagram showing an example of the configuration of the fraud detection device 100. Referring to FIG. 2, the fraud detection device 100 includes an acquisition unit 101, a determination unit 102, and a notification unit 103.

[0023] In FIG. 2 , the acquisition unit 101 is an example of an acquisition means for acquiring first vital data at a certain time point while a specific movement associated with the execution of fraudulent activity is being performed by a person whose start has been detected, and second vital data at a certain time point after the specific movement is completed by the person. The acquisition unit 101 may acquire the person's vital data from the measurement device for measuring vital data described above. The first vital data is vital data measured from the start of the specific movement to the end of the specific movement. For example, the first vital data may be vital data measured at the start of the specific movement, but this example is not limited thereto. The second vital data is vital data measured after the specific movement is completed. For example, the second vital data may be vital data measured at the end of the specific movement, but this example is not limited thereto. In the following description, the term "vital data" is used when referring to the nature or characteristics of the data and includes both first vital data and second vital data. The terms "first vital data" and "second vital data" are used when distinguishing between the times at which the vital data are measured.

[0024] First, the timing at which the acquiring unit 101 acquires vital data will be described. Note that, here, acquiring vital data refers to acquiring measured vital data from a measurement device. In other words, the timing at which the acquiring unit 101 acquires vital data is not limited to the timing at which the vital data is measured.

[0025] The acquisition unit 101 may acquire the person's vital data at predetermined times from the start to the end of the specific movement. For example, the acquisition unit 101 may acquire the first vital data and the second vital data when the end of the specific movement is detected. For example, when the start of the specific movement is detected, the acquisition unit 101 may acquire the person's first vital data when the start of the specific movement is detected, and when the end of the specific movement is detected, the acquisition unit 101 may acquire the person's second vital data when the end of the specific movement is detected. Furthermore, the acquisition unit 101 may continuously acquire the person's vital data from the start to the end of the specific movement. In this case, the acquisition unit 101 may acquire the first vital data and the second vital data measured at each time point from the continuously acquired vital data of the person using information indicating the times when the start and end of the specific movement were detected. Note that the acquisition unit 101 may acquire vital data before the start of the specific movement or continuously from before the start of the specific movement.

[0026] Similarly, the measuring device for measuring vital data may measure the vital data continuously or in response to an instruction from another device, such as fraud detection device 100 or a device for detecting the start and end of a specific action.

[0027] The specific action will now be described. A specific action is an action associated with the commission of fraudulent activity. A plurality of specific actions, as exemplified below, may be predefined as specific actions for which the fraud detection device 100 detects fraud. First, consider an example in which the fraud detection device 100 is used in a store. For example, a specific action associated with the commission of fraudulent activity is an action related to a person's handling of at least one of the store's merchandise and the store's money. Examples of actions related to the handling of at least one of the store's merchandise and the store's money include paying for merchandise, using a fitting room, placing merchandise in a container other than a cart or basket, picking up a specific product associated with the commission of fraudulent activity, or a person being in the store. Here, a specific product may be, for example, a product that is likely to be resold, priced above a certain price, or a product that is popular to a certain extent, and is therefore likely to be fraudulently purchased by shoplifting or other such means. Furthermore, a specific product may be, for example, a product that is often subject to fraudulent activity, such as the addition of foreign objects or damage. The specific product may be predefined by a store employee or other person. Furthermore, the specific action in various facilities including a store may be, for example, entering a specific room where fraudulent activity may be carried out.

[0028] The method for detecting a specific action is not particularly limited. Specific actions may be detected using a detection method and device appropriate for each action. The device for detecting a specific action may be included in the fraud detection device 100, or may be another device communicatively connected to the fraud detection device 100. The device for detecting a specific action in the fraud detection device 100 may acquire data from a camera or various sensors and detect the specific action based on the acquired data. In detecting a specific action, it may be possible to detect whether or not the specific action is being performed, or it may be possible to detect only the start or end of the specific action.

[0029] For example, the performance of a specific action may be detected by detecting the action of a person photographed by a camera.

[0030] Furthermore, for example, specific actions such as paying for a product or using a fitting room are generally performed at a predetermined location. In this case, the specific action may be detected by a person remaining at a predetermined location where the specific action is performed or by the person entering the predetermined location where the specific action is performed. In this case, the person remaining at a predetermined location where the specific action is performed or the person entering the predetermined location where the specific action is performed may be detected by analyzing an image captured by a camera, a human presence sensor, or the like. For example, when a person is detected remaining at a location in front of a checkout counter where the specific action of paying for a product is performed, the specific action of paying for a product may be detected as being performed. Note that a person may stay at the checkout counter for reasons other than paying for a product, and these may also be detected as a specific action in the same way as paying for a product. Other reasons that may cause a person to stay at the checkout counter include returning or exchanging a purchased product, exchanging a product using store benefits such as points or coupons, requesting a refund for an incorrect payment, or making an inquiry about another store or a product or service at the store. For example, regarding the specific behavior of using a fitting room, the specific behavior may be detected as having started when it is detected that the person has entered one of the fitting rooms or a fitting area including multiple fitting rooms and an aisle.Similarly, regarding the specific behavior of using a fitting room, the specific behavior may be detected as having ended when it is detected that the person has left one of the fitting rooms or a fitting area including multiple fitting rooms and an aisle.

[0031] Furthermore, if the specific operation is a transaction for a product, this specific operation may be detected from the processing status of the POS (Point of Sales) terminal device that performs the transaction. In the following description, the POS terminal device includes peripheral devices communicatively connected to the POS terminal device, such as an electronic payment terminal or a receipt printer. For example, the start of the specific operation may be detected when a store membership code or product identification code is entered into the POS terminal device without product registration. Furthermore, the end of the specific operation may be detected when the POS terminal device has finished issuing a receipt or invoice, or when the POS terminal device has dispensed change or a card.

[0032] Furthermore, if the specific action is a person picking up a specific product, this specific action may be detected based on the weight value or change measured by a weight sensor on the shelf on which the specific product is placed, or the reception strength of radio waves received by a reading antenna from the RFID attached to the specific product.

[0033] Furthermore, if the specific action is a person being in a store or a person entering a specific room where fraudulent activity may be occurring, the specific action may be detected by authentication when the person enters the store or the specific room. Authentication when a person enters a store or a specific room may be authentication for opening a door, such as biometric authentication or holding a card used for authentication over a card reader. Furthermore, authentication when entering a store may be authentication for confirming that the person has entered the store, such as reading a code provided in the store with a terminal device carried by the person. Furthermore, for example, this specific action may be detected by location information of a terminal device carried by the person.

[0034] The specific actions and the method for detecting the specific actions have been described above. As described above, the detection of the specific actions may be performed by another device communicably connected to fraud detection device 100, or may be performed by a detection unit (not shown) provided in fraud detection device 100.

[0035] The acquisition unit 101 acquires first vital data and second vital data of a person who has started a specific movement in response to detection of the specific movement by each of the above-mentioned detection methods. For example, when vital data is measured by a measurement system from video of a person captured by a camera, the acquisition unit 101 acquires the first vital data and second vital data measured based on video captured by a camera capturing the position where the specific movement is being performed. Also, for example, the acquisition unit 101 acquires the first vital data and second vital data based on the video used to detect the specific movement.

[0036] Furthermore, for example, if the person's terminal device can be identified in detecting the specific behavior, the acquisition unit 101 may acquire, from the person's terminal device, vital data measured by the person's terminal device or a sensor communicatively connected to the person's terminal device. For example, if a POS terminal device reads the person's member identification information during product checkout, the acquisition unit 101 may acquire vital data from a terminal device that is logged in to the store's website or application using the read member identification information. The member identification information is information such as a number that the store uses to identify the customer as a member. Furthermore, for example, if the person's facial image is registered in advance, the acquisition unit 101 may acquire vital data from a terminal device associated with the person detected in the video used to detect the specific behavior.

[0037] The determination unit 102 is an example of a determination means for determining whether a change from the first vital data to the second vital data corresponds to a change in vital data that occurs when fraudulent activity is committed. When a person commits fraudulent activity, it is considered that a change in the person's mental state may occur when the fraudulent activity is completed. For this reason, the determination unit 102 determines whether a change in vital data occurs when fraudulent activity is committed, based on a change from the first vital data at a certain point during the specific action to the second vital data at the point when the specific action is completed. The change in vital data that occurs when fraudulent activity is committed may be stored in advance inside the fraud detection device 100 or in an external storage device connected for communication.

[0038] The change in vital data resulting from the misconduct represents a change in the person's internal state after the misconduct. For example, the change in vital data resulting from the misconduct may be a change from vital data indicating a tense state to vital data indicating a relaxed state after the misconduct is completed. Furthermore, the change in vital data resulting from the misconduct may be a change from normal vital data to vital data indicating an excited state after the misconduct is completed, but is not limited to these examples. The change in vital data resulting from the misconduct may not only be a change from one value to another, but also the amount of change in the vital data. For example, the change in vital data resulting from the misconduct may be a change from a heart rate of 120 or more to 100 or less, or a decrease of 20 or more heart rates, but is not limited to these examples.

[0039] For example, a change in vital data indicating relief due to the completion of fraudulent activity is a change in vital data when tension decreases. A change in vital data when tension decreases is a change in vital data when tension decreases from a high tension state. A decrease in tension can be, for example, a decrease in heart rate, a decrease in respiratory rate, a decrease in body temperature, etc., but examples of changes in vital data when tension decreases are not limited to these.

[0040] Furthermore, for example, a change in vital data indicating that a person is in a happier mood after completing a fraudulent act is a change in the person's facial expression, but is not limited to this example.

[0041] Furthermore, the determination unit 102 may use a trained model to determine whether a change from the first vital data to the second vital data corresponds to a change in vital data that would occur if fraudulent activity were committed. In this case, the trained model is a model that has learned information indicating changes in a person's vital data and whether the person has committed fraudulent activity. The determination unit 102 inputs the first vital data and the second vital data, or the change from the first vital data to the trained model, to determine whether the change corresponds to a change in vital data that would occur if fraudulent activity were committed.

[0042] The notification unit 103 is an example of a notification means that notifies that fraudulent activity may have been performed based on the determination result. The notification unit 103 notifies that fraudulent activity may have been performed when the determination unit 102 determines that the change from the first vital data to the second vital data matches the change in vital data that would occur if fraudulent activity had been performed. Furthermore, the notification unit 103 may notify that fraudulent activity may not have been performed when the determination unit 102 determines that the change from the first vital data to the second vital data does not match the change in vital data that would occur if fraudulent activity had been performed.

[0043] The notification unit 103 outputs a notification to at least one of the output devices described above.

[0044] FIG. 3 illustrates an example of a notification screen displayed on a terminal device used by a store employee. In FIG. 3 , the notification screen includes information indicating that possible fraudulent activity has been detected, the location within the store where the possible fraudulent activity occurred, and an image of the person who may have committed the fraudulent activity. The image of the person who may have committed the fraudulent activity may also be real-time video footage tracking the person who may have committed the fraudulent activity. 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 an employee, the notification unit 103 may output an alert to another output device informing other employees or security guards that possible fraudulent activity has been detected. Furthermore, for example, when the above button is selected, the notification unit 103 may output an alert to the person who has been detected as possibly committing fraudulent activity. In this way, allowing the employee to decide whether to issue an alert may prevent the employee from accidentally alerting a person who has not committed fraud or from provokeing a person who is becoming increasingly aggressive.

[0045] The notification screen is not limited to the example shown in Fig. 3. In the case of a notification screen for a terminal device used by a store employee, the notification screen may be one that enables the employee to determine that a fraudulent act may have occurred.

[0046] The operation of the fraud detection device 100 configured as above will be described with reference to the flowchart of FIG.

[0047] As shown in Figure 4, first, the acquisition unit 101 acquires first vital data at a certain point during the specific action of a person in which the start of a specific action related to the commission of fraudulent activity has been detected, and second vital data of the person at the point when the specific action has ended (step S101).

[0048] Next, the determination unit 102 determines whether the change from the first vital data to the second vital data corresponds to a change in vital data when fraudulent activity is committed (step S102).

[0049] Next, the notification unit 103 notifies the user that there is a possibility that fraudulent activity has occurred, depending on the result of the determination (step S103).

[0050] This completes the series of operations of the fraud detection device 100.

[0051] The fraud detection device according to the present embodiment includes an acquisition unit, a determination unit, and a notification unit. The acquisition unit acquires first vital data at a certain point during a specific action of a person for which the start of the specific action associated with the execution of fraudulent activity has been detected, and second vital data of the person at the time the specific action ends. The determination unit determines whether a change from the first vital data to the second vital data corresponds to a change in vital data that would occur if fraudulent activity were performed. The notification unit then notifies the person that fraudulent activity may have been performed based on the determination result. As a result, the fraud detection device according to the present embodiment can reduce false positives when detecting fraudulent activity.

[0052] When a person commits fraud, it is believed that the completion of the fraud may cause a change in the person's mental state. In contrast, it is believed that mental states caused by factors other than the fraud may not change before and after a specific action related to the fraud, or may change differently from when the fraud is committed. The fraud detection device of this embodiment determines whether the change from the first vital data to the second vital data corresponds to a change in vital data when fraud is committed. This allows the fraud detection device of this embodiment to reduce false positives that a person who is in a state of high tension or high spirits due to factors other than fraud may have committed fraud.

[0053] [Variation 1] The determination unit 102 may further use vital data at other times besides the first vital data and the second vital data to determine whether a change in the vital data related to the specific movement corresponds to a change in the vital data resulting from misconduct. For example, the determination unit 102 may determine whether a continuous change in the vital data while the specific movement is being performed corresponds to a change in the vital data resulting from misconduct. Furthermore, for example, the determination unit 102 may determine whether changes in the vital data at multiple times while the specific movement is being performed correspond to a change in the vital data resulting from misconduct. In this case, the acquisition unit 101 acquires vital data at other times to be used in the determination by the determination unit 102.

[0054] For example, the determination unit 102 may determine whether a change in vital data while the specific movement is being performed corresponds to a change in vital data that would occur if fraudulent activity were committed. In this case, the acquisition unit 101 acquires the vital data while the specific movement is being performed. The vital data while the specific movement is being performed may be vital data from the start of the specific movement to the end of the specific movement, as shown in FIG. 5 , which illustrates another example of an outline of a system including the fraud detection device 100.

[0055] Furthermore, for example, the determination unit 102 may use vital data at a time other than the time during which the specific behavior is being performed to determine whether a change in vital data corresponds to a change in vital data that would occur if fraudulent activity were committed. The vital data at a time other than the time during which the specific behavior is being performed may be, for example, at least one of third vital data at a time before the specific behavior is being performed and fourth vital data at a time when the person exits a facility such as a store within the fraud detection range of the fraud detection device 100, as shown in FIG. 6 . The third vital data may be vital data at a time after the person enters a facility such as a store within the fraud detection range of the fraud detection device 100, or vital data at a time when the person enters a facility such as a store within the fraud detection range of the fraud detection device 100.

[0056] In this case, for example, the determination unit 102 determines whether the third vital data indicates that the person is nervous. Based on the determination of the third vital data, the determination unit 102 determines changes in the first vital data and the second vital data for a person who is nervous before performing a specific movement. Note that in this case, the acquisition unit 101 may acquire the first vital data and the second vital data if it is determined that the third vital data indicates that the person is nervous. Based on the determination of the third vital data, the determination unit 102 may not determine changes in the first vital data and the second vital data as described in the embodiment for a person who is not nervous before performing a specific movement. That is, if the third vital data indicates that the person is not nervous, the determination unit 102 may determine that there is no possibility of fraud. This is because a person who is not nervous before performing a specific movement is considered to be less likely to commit fraud. That is, the determination unit 102 may use the third vital data, the first vital data, and the second vital data to determine whether to perform fraud detection.

[0057] Furthermore, for example, when a change from the first vital data to the second vital data does not correspond to a change in vital data resulting from the commission of fraudulent activity, the determination unit 102 may use the fourth vital data to determine whether the change in vital data corresponds to a change in vital data resulting from the commission of fraudulent activity. In this case, the determination unit 102 may determine whether a change from the first vital data to the fourth vital data corresponds to a change in vital data resulting from the commission of fraudulent activity. This is because, even after the fraudulent activity in the specific action has been completed, the person may be pointed out for the fraudulent activity or may be called out by a facility employee or security guard until the person leaves the facility, and a change in mental state may appear upon leaving the facility due to the completion of the fraudulent activity.

[0058] An example of the operation of the fraud detection device 100 when the third vital data and the fourth vital data are further used in this way will be described with reference to the flowchart of FIG.

[0059] As shown in FIG. 7, first, the acquisition unit 101 starts acquiring vital data of a person who is detected as having entered a store (step S201).

[0060] Next, when the start of the specific movement is detected, the determination unit 102 determines whether the third vital data at a certain point in time before the start of the specific movement corresponds to vital data when the person is nervous (step S202). If the third vital data does not correspond to vital data when the person is nervous (step S202: No), the fraud detection device 100 may end the series of processes.

[0061] If the third vital data corresponds to vital data when the person is nervous (step S202: Yes), the judgment unit 102 determines whether the change from the first vital data to the second vital data corresponds to a change in vital data when fraudulent activity is committed (step S203).

[0062] If the change from the first vital data to the second vital data does not correspond to the change in vital data that would occur if fraudulent activity were committed (step S203: No), the determination unit 102 further determines whether the change from the first vital data to the fourth vital data at the time of leaving the store corresponds to the change in vital data that would occur if fraudulent activity were committed (step S204).If it is determined that the change from the first vital data to the fourth vital data at the time of leaving the store does not correspond to the change in vital data that would occur if fraudulent activity were committed (step S204: No), the fraud detection device 100 may end the series of processes.

[0063] If the change from the first vital data to the second vital data corresponds to the change in vital data that would occur if fraudulent activity were committed (step S203: Yes), or if the change from the first vital data to the fourth vital data at the time of leaving the store corresponds to the change in vital data that would occur if fraudulent activity were committed (step S204: Yes), the notification unit 103 notifies that fraudulent activity may have occurred (step S205).

[0064] This completes the series of operations of the fraud detection device 100.

[0065] 6 illustrates the acquisition of first, second, third, and fourth vital data, respectively, but the timing of acquiring the vital data is not limited to these. For example, the acquisition unit 101 may acquire vital data from the time of entering the store until the time of leaving the store. Furthermore, in this case, the determination unit 102 may determine whether a change in vital data corresponds to a change in vital data resulting from fraudulent activity, based on a change in vital data from the time of entering the store until the time of leaving the store, including the time of performing a specific action.

[0066] [Variation 2] The fraud detection device 100 may perform the processing described in the embodiment or Variation 1 on a specific individual. The specific individual is a person associated with the commission of fraudulent activity. Examples of a person associated with the commission of fraudulent activity include a person who is likely to commit fraud or a person who has committed fraud in the past. In Variation 2, the acquisition unit 101 acquires first vital data and second vital data of the specific individual when, for example, a person determined to be the same person as a pre-stored image of the specific individual is detected from video. The determination unit 102 determines whether changes in the first vital data and the second vital data of the specific individual correspond to changes in vital data that would occur if fraud had been committed. Then, the notification unit 103 may notify the user that fraudulent activity may have been committed by the specific individual based on the determination result by the determination unit 102. Note that the method for detecting the specific individual is not particularly limited. Furthermore, the detection of the specific individual may be performed by another device externally provided to the fraud detection device 100 and capable of communicating with the fraud detection device 100, as with the device that detects the specific behavior, or by a detection unit (not shown) provided in the fraud detection device 100. Alternatively, the fraud detection device 100 may include a person determination unit (not shown). In this case, the person determination unit may determine whether the person who performed the specific action is a specific person, and if the person who performed the specific action is a specific person, the acquisition unit 101 may acquire the first vital data and the second vital data.

[0067] This allows the processing of the fraud detection device 100 in the second modification to be reduced.

[0068] [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. 8. The information processing device 1000 includes, as an example, the following configuration.

[0069] - 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.

[0070] 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.

[0071] 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.

[0072] 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.

[0073] 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.

[0074] 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.

[0075] 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.

[0076] Some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0077] [Supplementary Note 1] A fraud detection device comprising: an acquisition means for acquiring first vital data of a person whose start of a specific action related to the execution of fraudulent activity has been detected at a certain point while the specific action is being performed, and second vital data of the person at the point when the specific action has ended; a determination means for determining whether a change in the second vital data from the first vital data corresponds to a change in vital data that would occur if fraudulent activity were to be performed; and a notification means for notifying that fraudulent activity may have been performed based on the result of the determination.

[0078] [Supplementary Note 2] The fraud detection device according to Supplementary Note 1, wherein the specific behavior is a behavior by the person that involves handling at least one of merchandise in the store and money in the store at the store visited by the person.

[0079] [Supplementary Note 3] The fraud detection device according to Supplementary Note 2, wherein the action relating to at least one of merchandise in the store and money in the store is a transaction for the merchandise.

[0080] [Supplementary Note 4] The fraud detection device according to Supplementary Note 2 or 3, wherein the action relating to at least one of merchandise in the store and money in the store is use of a fitting room.

[0081] [Supplementary Note 5] The fraud detection device according to any one of Supplementary Notes 2 to 4, wherein the action relating to at least one of the store's merchandise and the store's money is putting the merchandise into a container other than a cart or basket.

[0082] [Supplementary Note 6] The fraud detection device according to any one of Supplementary Notes 2 to 5, wherein the action relating to at least one of the merchandise in the store and the money in the store is picking up a specific merchandise that is associated with the commission of the fraudulent act.

[0083] [Supplementary Note 7] The fraud detection device according to Supplementary Note 6, wherein the specific product is a product that may be resold.

[0084] [Supplementary Note 8] The fraud detection device according to Supplementary Note 6 or 7, wherein the specific product is a product priced higher than a predetermined price.

[0085] [Supplementary Note 9] The fraud detection device according to Supplementary Note 2, wherein the action relating to at least one of the merchandise in the store and the money in the store is the person being in the store.

[0086] [Supplementary Note 10] The fraud detection device according to Supplementary Note 9, wherein the first vital data is vital data when the person enters the store, and the second vital data is vital data when the person leaves the store.

[0087] [Supplementary Note 11] The fraud detection device described in any of Supplementary Notes 1 to 10, wherein the acquisition means continuously acquires vital data from the first vital data to the second vital data, and the determination means determines whether a change in vital data from the first vital data to the second vital data corresponds to a change in vital data when fraudulent activity is committed.

[0088] [Supplementary Note 12] The fraud detection device described in any of Supplementary Notes 1 to 11, wherein the acquisition means continuously acquires vital data from the time the person enters the store, and the determination means determines whether a change in the vital data from a certain point before the specific action is performed to a point at which the specific action ends corresponds to a change in vital data that would occur if fraudulent activity were performed.

[0089] [Supplementary Note 13] The fraud detection device according to any one of Supplementary Notes 1 to 12, wherein the change in vital data when fraudulent behavior is committed is a change in vital data when tension level decreases.

[0090] [Supplementary Note 14] The fraud detection device according to Supplementary Note 13, wherein the change in vital data when the tension level decreases is a change in vital data when the tension level decreases from a high tension state.

[0091] [Supplementary Note 15] The fraud detection device according to any one of Supplementary Notes 12 to 14, wherein the change in vital data when fraudulent behavior is committed is a change in vital data when a person is in an elevated mood.

[0092] [Appendix 16] The fraud detection device described in any of Appendices 1 to 15, wherein the acquisition means acquires the first vital data and the second vital data if the third vital data at a certain point before the specific action indicates a high level of tension.

[0093] [Supplementary Note 17] The fraud detection device described in any one of Supplementary Notes 1 to 16, wherein the determination means determines that there is no possibility of fraud if the third vital data at a certain point before the specific action is performed indicates a low level of tension.

[0094] [Supplementary Note 18] The fraud detection device according to any one of Supplementary Notes 1 to 17, wherein the acquisition means acquires vital data of a specific person when a specific action by the specific person that is associated with the commission of fraudulent activity is detected.

[0095] [Supplementary Note 19] The fraud detection device according to Supplementary Note 18, wherein the specific person is a person who has committed fraud in the past.

[0096] [Supplementary Note 20] A fraud detection method in which a computer acquires first vital data at a certain point while a specific action related to the performance of fraudulent activity is being performed by a person who has been detected as having started the specific action, and second vital data at the point when the specific action ends; determines whether a change in the second vital data from the first vital data corresponds to a change in vital data that would occur if fraudulent activity were to be performed; and notifies the person that fraudulent activity may have been performed based on the result of the determination.

[0097] [Supplementary Note 21] A recording medium that records a program that causes a computer to execute the following processes: acquiring first vital data at a certain point while a specific action related to the execution of fraudulent activity is being performed by a person who has been detected as having started the specific action, and second vital data at the point when the specific action ends; determining whether a change in the second vital data from the first vital data corresponds to a change in vital data that would occur if fraudulent activity were to be performed; and notifying the person that fraudulent activity may have been performed based on the result of the determination.

[0098] REFERENCE SIGNS LIST 100 Fraud detection device 101 Acquisition unit 102 Determination unit 103 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. Acquisition means for acquiring first vital data of a person at a certain point in time while the specific action related to the execution of fraudulent activity is being performed, and second vital data of the person at the time the specific action is completed. A determination means for determining whether the change from the first vital data to the second vital data corresponds to the change in vital data that occurs when fraudulent activity is performed, A notification means for notifying that fraudulent activity may have occurred, based on the results of the aforementioned determination, A fraud detection device equipped with the following features.

2. The aforementioned specific action is an action performed by the person in a store that the person has visited, relating to the handling of at least one of the store's goods and the store's money. The fraud detection device according to claim 1.

3. Actions relating to at least one of the store's goods and the store's money include at least one of the following: paying for the goods, using a fitting room, placing the goods in a container other than a cart or basket, and picking up specific goods related to the performance of the fraudulent act. The fraud detection device according to claim 2.

4. The aforementioned specific products are products that may be resold, or products that are priced higher than a predetermined price. The fraud detection device according to claim 3.

5. An action relating to at least one of the goods of the store and the money of the store is the presence of the person in the store. The first vital data mentioned above is the vital data of the person at the time of entering the store. The second set of vital data refers to the vital data of the person at the time they left the store. The fraud detection device according to claim 2.

6. The acquisition means continuously acquires vital data from the first vital data to the second vital data, The determination means determines whether the changes in vital data from the first vital data to the second vital data correspond to the changes in vital data that occur when fraudulent activity is performed. The fraud detection device according to claim 1.

7. The acquisition means continuously acquires vital data from the time the person enters the store. The determination means determines whether the changes in vital data from a certain point in time before the specific action is performed to the end of the specific action correspond to the changes in vital data that would occur if fraudulent activity were performed. The fraud detection device according to claim 1.

8. The changes in vital data that occur when the aforementioned misconduct is performed are those that occur when the level of tension decreases, or those that occur when the mood is elevated. The fraud detection device according to any one of claims 1 to 7.

9. Computers For a person in whom the commencement of a specific action associated with the execution of fraudulent activity is detected, first vital data at a certain point in time while the specific action is being performed, and second vital data at the time the specific action is completed are obtained. Determine whether the change from the first vital data to the second vital data corresponds to the change in vital data that occurs when fraudulent activity is performed. Depending on the results of the above determination, we will notify you that there is a possibility that fraudulent activity has occurred. Fraud detection methods.

10. For a person in whom the commencement of a specific action associated with the execution of fraudulent activity is detected, first vital data at a certain point in time while the specific action is being performed, and second vital data at the time the specific action is completed are obtained. Determine whether the change from the first vital data to the second vital data corresponds to the change in vital data that occurs when fraudulent activity is performed. Depending on the results of the above determination, we will notify you that there is a possibility that fraudulent activity has occurred. A program that instructs a computer to perform a process.

Images