Fraud prevention device, fraud prevention method, and fraud prevention system

Abstract

In the present invention, a vehicle is provided with: an image acquisition unit (11) that acquires an image captured by an image capture device provided so as to be able to capture an image of an area around the driver's seat, including at least the driver's foot area; a detection unit (12) that detects an occupant captured in a first region of the captured image and an occupant captured in a second region of the captured image; a fraud provisional determination unit (13) that provisionally determines that there is no fraud if one occupant is detected in the first region and no occupant is detected in the second region; and a fraud determination result confirmation unit (14) that confirms that there is no fraud if the fraud provisional determination unit (13) provisionally determines that there is no fraud and furthermore it is presumed that an occupant that takes a test and the occupant detected in the first region are the same person. The detection unit (12) detects an occupant captured in the second region by using a method different from the method used to detect an occupant captured in the first region.

Description

Anti-fraud device, anti-fraud method, and anti-fraud system

[0001] The present disclosure relates to a fraud prevention device, a fraud prevention method, and a fraud prevention system that prevent fraud in driver breathalyzer tests.

[0002] In recent years, breathalyzer tests before driving have become mandatory. Accordingly, technologies have been developed to prevent illicit conduct in pre-driving breathalyzer tests, in which a passenger other than the driver pretends to be the driver and administers the breathalyzer test. For example, Patent Document 1 discloses a determination device that determines the reliability of an alcohol concentration when breath is detected, the alcohol concentration in the breath is measured, and whether the measured alcohol concentration is equal to or greater than a predetermined value. The determination device determines that the reliability of the measured alcohol concentration is low if it determines that a person image extracted from an image captured from the driver's seat is not a single person. The determination device disclosed in Patent Document 1 recognizes facial contours from an image captured from the driver's seat to extract a person image and counts the extracted person images.

[0003] JP 2009-90866 A

[0004] One example of a fraudulent case in which a passenger other than the driver pretends to be the driver and conducts a breathalyzer test is when the driver changes their posture while seated in the driver's seat, positions their face where the driver's face should be, and then acts as if they are seated in the driver's seat to conduct the breathalyzer test. To give a specific example, the driver may lose their posture and move their face downward, and a rear seat passenger may lean forward and bring their face near the headrest where the driver's face is expected to be, thereby conducting the breathalyzer test. In this case, because the driver's posture is poor, the contours of the driver's face may not be extracted from the captured image. Conventional technologies, which fail to take into account the above-mentioned fraudulent cases, may erroneously determine that a single person is extracted from the captured image, even though there are actually multiple people near the driver's seat and multiple person images should be extracted from the captured image. In other words, the conventional technology has the problem that it may not be possible to completely prevent illicit conduction of breathalyzer tests by passengers other than the driver pretending to be the driver.

[0005] The present disclosure has been made to solve the above-mentioned problems, and aims to provide an anti-fraud device that prevents fraud in breathalyzer tests administered to drivers in vehicles.

[0006] The fraud prevention device according to the present disclosure includes an image acquisition unit that acquires an image captured by an imaging device that is provided in a vehicle and is capable of capturing an image of the periphery of the driver's seat including at least the foot area, a detection unit that detects an occupant captured in a first area around the headrest of the driver's seat in the captured image and an occupant captured in a second area below the headrest of the driver's seat in the captured image including the foot area of ​​the driver's seat, based on the captured image acquired by the image acquisition unit, and a fraud provisional determination process that provisionally determines whether or not fraud has occurred based on the presence or absence of an occupant detected in the first area or the second area, and when one occupant is detected in the first area and one occupant is detected in the second area, and a fraud determination result confirmation unit that performs a fraud determination result confirmation process to determine whether or not there is fraud, and that, if the fraud determination unit provisionally determines that there is no fraud and detects the face of the test occupant performing the alcohol test based on the captured image acquired by the image acquisition unit and compares the face of the test occupant with the face of the occupant detected in the first area, determines that there is no fraud if it is presumed that the test occupant and the occupant detected in the first area are the same person, and the detection unit detects the occupant imaged in the second area by a method different from the method for detecting the occupant imaged in the first area.

[0007] According to the present disclosure, it is possible to prevent fraud in breathalyzer tests administered to drivers in vehicles.

[0008] 9A and 9B are diagrams illustrating an example of the configuration of the anti-fraud device according to the first embodiment.

[0034] FIG. 9A is a diagram illustrating a first area and a second area detected by the detection unit according to the first embodiment.

[0035] FIG. 9B is a flowchart illustrating an example of the operation of the anti-fraud device according to the first embodiment.

[0036] FIG. 9B is a flowchart illustrating details of step ST20 of FIG. 3.

[0037] FIG. 9B is a flowchart illustrating details of the processing of step ST30 of FIG. 3.

[0038] FIG. 9B is a flowchart illustrating an example of the operation of the anti-fraud device according to the second embodiment.

[0039] FIG. 9A is a diagram illustrating an example of the configuration of the anti-fraud device according to the first embodiment.

[0039] FIG. 9B is a diagram illustrating an example of the configuration of the anti-fraud device according to the second embodiment.

[0039] Fig. 12 is a flowchart for explaining details of the process of step ST35 in Fig. 11. Fig. 13 is a diagram showing an example of the configuration of an anti-tamper device provided with a recheck execution unit in the first embodiment.

[0009] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. Embodiment 1. An anti-fraud device according to embodiment 1 prevents a driver from cheating by impersonating a passenger other than the driver in a breathalyzer test administered to the driver of a vehicle, thereby allowing the driver to evade the breathalyzer test. For example, when a driver of a vehicle gets into the vehicle, the breathalyzer test is administered before driving the vehicle. In the following embodiment 1, when the term "fraud" is used simply, the "fraud" refers to a passenger other than the driver impersonating the driver in an attempt to evade the breathalyzer test.

[0010] FIG. 1 is a diagram showing an example configuration of an anti-fraud device 1 according to embodiment 1. In embodiment 1, it is assumed that the anti-fraud device 1 is mounted on a vehicle (not shown). The anti-fraud device 1 is connected to an imaging device 2 and an output device 3. In embodiment 1, the anti-fraud device 1 and the imaging device 2 constitute an anti-fraud system 100. Note that this is merely an example, and for example, the anti-fraud system 100 may also be constituted by the anti-fraud device 1, the imaging device 2, and the output device 3.

[0011] The imaging device 2 is, for example, a near-infrared camera or a visible light camera, and captures images of occupants present in the vehicle cabin. The imaging device 2 may be, for example, a camera commonly used with a so-called "Driver Monitoring System (DMS)." Note that, for example, when the vehicle is powered on, the imaging device 2 constantly captures images of occupants present in the vehicle cabin. The imaging device 2 is installed in the vehicle cabin so as to be able to capture images of the area around the driver's seat, including at least the foot area. The area around the driver's seat, including the foot area, in the vehicle cabin refers to, for example, a space near the front of the driver's seat headrest, a space near the front of the driver's seat backrest, and a space near the front of the driver's seat below the seat cushion. The imaging device 2 is installed, for example, in the center of the vehicle in the vehicle width direction. Note that, in the first embodiment, the "center" in the vehicle width direction is not limited to being strictly "center" but also includes "approximately center." For example, the imaging device 2 is installed on an overhead console, a mirror such as a rearview mirror, the center console of the vehicle, or near the center of the dashboard where a car navigation system or the like is installed. The imaging device 2 may be a wide-angle camera that can capture images of the area around the driver's seat, including the foot area, as well as the area where seats other than the driver's seat are located within the vehicle's cabin. In the following first embodiment, as an example, it is assumed that the imaging device 2 is installed near the vehicle's rearview mirror and captures images of the area around the driver's seat, including the foot area, the area around the passenger seat, including the foot area, and the space between the driver's seat and the passenger seat, from near the vehicle's rearview mirror. The imaging device 2 outputs the captured images to the anti-fraud device 1.

[0012] Based on the captured image captured by the imaging device 2, the anti-fraud device 1 determines whether or not a passenger other than the driver is impersonating the driver and attempting to evade the breathalyzer test administered to the driver before the vehicle is driven. The anti-fraud device 1 also acquires information (hereinafter referred to as "test result information") indicating the breath alcohol concentration of the passenger who actually performs the breathalyzer test by blowing his / her breath (hereinafter referred to as the "test passenger") from a detection device (not shown), and determines whether or not the test passenger is drunk based on the acquired test result information. The detection device is, for example, an alcohol checker that measures the breath alcohol concentration of the passenger, and includes a blowing port through which the passenger blows his / her breath and a gas measurement sensor that measures the breath alcohol concentration of the passenger blown into the blowing port. The detection device is, for example, attached to the dashboard and positioned adjacent to the steering wheel.

[0013] If the anti-fraud device 1 determines that there is no tampering and that the test occupant is not drinking alcohol, it releases the vehicle interlock. If the anti-fraud device 1 determines that there is tampering or that the test occupant is drinking alcohol, it maintains the vehicle interlock. In embodiment 1, the vehicle has an interlock function. The interlock function is a function that controls whether the vehicle engine is startable or unstartable. In embodiment 1, for example, a control unit (not shown) of the anti-fraud device 1 places the vehicle in an interlock state, in other words, a state in which the engine cannot be started, when the vehicle's power is turned on. Note that the control to place the vehicle in the interlock state when the vehicle's power is turned on may be performed by a device external to the anti-fraud device 1. In embodiment 1, releasing the vehicle interlock means changing the engine from an unstartable state to an engine startable state. Maintaining the vehicle interlock means maintaining the vehicle's engine unstartable state. In the first embodiment, the continuing of the vehicle interlock also includes changing the vehicle engine from an unstartable state to a startable state, and then making the vehicle engine unstartable again before the vehicle engine actually starts. Details of an example configuration of the anti-tamper device 1 will be described later.

[0014] The output device 3 is mounted on, for example, a vehicle. The output device 3 is, for example, a display device or an audio output device provided in a navigation device (not shown) mounted on the vehicle. The output device 3 outputs various information based on instructions from the anti-fraud device 1. For example, the output device 3 displays or outputs audio to prompt the driver to blow into the detection device based on instructions from the anti-fraud device 1. The test occupant blows into the detection device in accordance with the display or audio output from the output device 3 and performs the alcohol test.

[0015] The following describes in detail an example configuration of the anti-fraud device 1. The anti-fraud device 1 includes an image acquisition unit 11, a detection unit 12, a provisional fraud determination unit 13, a fraud determination result confirmation unit 14, a test result acquisition unit 15, a drinking determination unit 16, and an interlock unit 17. The detection unit 12 includes a first detection unit 121 and a second detection unit 122.

[0016] The image acquisition unit 11 acquires the captured image from the imaging device 2. The image acquisition unit 11 outputs the acquired captured image to the detection unit 12 and the fraud determination result confirmation unit 14.

[0017] The detection unit 12 detects an occupant who is imaged in an area around the headrest of the driver's seat in the captured image (hereinafter referred to as a "first area") and an occupant who is imaged in an area below the headrest of the driver's seat and including the area at the driver's feet in the captured image (hereinafter referred to as a "second area"), based on the captured image acquired by the image acquisition unit 11. In the first embodiment, the process by which the detection unit 12 detects the occupant who is imaged in the first area and the occupant who is imaged in the second area is referred to as a "detection process."

[0018] Specifically, the first detection unit 121 of the detection unit 12 detects a first region in the captured image based on the captured image acquired by the image acquisition unit 11, and detects the occupant whose image is captured in the detected first region. The second detection unit 122 of the detection unit 12 detects a second region in the captured image based on the captured image acquired by the image acquisition unit 11, and detects the occupant whose image is captured in the detected second region.

[0019] FIG. 2 is a diagram illustrating the first and second regions detected by the detection unit 12 in the first embodiment. In FIG. 2, the captured image is indicated by "I." In FIG. 2, the captured image shows the driver and the passenger seat occupant seated correctly in the driver's seat and the passenger seat, respectively. As an example, the vehicle is a right-hand drive vehicle, but for convenience, structures inside the vehicle cabin other than the seats, such as the steering wheel, are omitted from FIG. 2. In FIG. 2, the first region is indicated by "201" and the second region is indicated by "202." In FIG. 2, the driver's seat, the driver seat, the passenger seat, and the passenger seat occupant are indicated by "S1," "D," "S2," and "P," respectively, all of which are captured in the captured image.

[0020] In the first embodiment, the first area on the captured image is an area set for authenticating an occupant seated in the driver's seat who is assumed to be about to undergo an alcohol test, and is set to suppress erroneous authentication of a person located outside the vehicle, a person seated in the rear seat of the vehicle, and a person seated in the passenger seat of the vehicle. As described above, the first area is set around the headrest of the driver's seat. The first area is, for example, a rectangular area. The position of the first area on the ceiling side of the vehicle interior is set at a predetermined distance (e.g., approximately 15 cm) from the top of the headrest of the driver's seat toward the ceiling side of the vehicle interior. The position of the first area on the exterior side of the vehicle is set at a predetermined distance (e.g., approximately 20 cm) from the side of the backrest of the driver's seat toward the exterior side of the vehicle. The position of the first area on the passenger seat side is set at a predetermined distance (e.g., approximately 10 cm) from the side of the backrest of the driver's seat toward the passenger seat. The position of the first area on the floor side of the vehicle interior is set at the position of the bottom of the headrest of the driver's seat. The distance from the headrest to the ceiling side of the vehicle interior and the predetermined distance from the seatback are appropriately set so that the faces of people outside the vehicle and people in the backseats do not enter the area. The first detection unit 121 stores the coordinate values ​​of the first area in a buffer or memory area (not shown).

[0021] After detecting the first region, the first detection unit 121 detects an occupant whose image is captured in the first region. Specifically, the first detection unit 121 detects a face in the first region, thereby detecting the occupant whose image is captured in the first region. The first detection unit 121 performs a known image analysis process on the captured image to detect an area in which a human face is captured (hereinafter referred to as a "face region"). At this time, the first detection unit 121 also detects feature information of the captured image of the face region. The first detection unit 121 sets a rectangle that surrounds the face region on the captured image and acquires coordinate values ​​of each vertex of the set rectangle. The coordinate values ​​acquired by the first detection unit 121 are coordinate values ​​in the captured image. The first detection unit 121 outputs the feature information of the captured image of the acquired face region and the coordinate values ​​of the face region to the provisional fraud determination unit 13 as a first detection result. Here, the feature information of the captured image includes, for example, contrast values ​​of the eyes, nose, mouth, and cheeks after normalizing the size of the face.

[0022] In the first embodiment, the second region on the captured image is a region set to detect an occupant (specifically, the driver) hiding in the driver's seat in an attempt to commit an illegal act. The second region is typically set to detect the face or head of an occupant located around the driver's seat below the headrest where the driver's face is expected to be located when the driver sits upright. In other words, the second region on the captured image is typically set to detect the face or head of an occupant located in a position where the driver's face is expected not to be present when the driver is about to blow breath into the seat for a breathalyzer test. As described above, the second region is set as a region below the headrest of the driver's seat and including the area around the driver's feet. The second region is, for example, a rectangular region. The ceiling-side position of the second region in the vehicle interior is set below the headrest of the driver's seat. The exterior position of the second region is set a predetermined distance (e.g., approximately 20 cm) from the side of the backrest of the driver's seat toward the exterior of the vehicle. The position of the passenger seat side of the second area is set at a position a predetermined distance (e.g., about 10 cm) away from the side of the backrest of the driver's seat. The position of the second area on the passenger seat side of the vehicle interior floor is set on the floor below the seat of the driver's seat. The second detection unit 122 stores the coordinate values ​​of the second area in a buffer or memory area.

[0023] When the second detection unit 122 detects the second area, it detects the occupant whose image is captured in the second area. Specifically, the second detection unit 122 detects the face or head in the second area, thereby detecting the occupant whose image is captured in the second area. Note that in the first embodiment, the "head" includes only the top of the head, excluding facial features such as the eyes, nose, and mouth. Here, an example of a method for detecting the occupant whose image is captured in the second area by the second detection unit 122 will be described in detail. In the first embodiment, the second detection unit 122 detects the occupant whose image is captured in the second area using a method different from the method used by the first detection unit 121 to detect the occupant whose image is captured in the first area.

[0024] Specifically, the second detection unit 122 uses an occupant recognition dictionary to detect an occupant whose image is captured in the second region based on the captured image and the occupant recognition dictionary. The occupant recognition dictionary is generated through machine learning of a whole-body image of a person sitting in the driver's seat facing forward or backward (hereinafter referred to as a "full-body image"), a whole-body image of a person sitting in the driver's seat facing sideways, and a whole-body image of a person sitting in a relaxed position in the driver's seat. Note that a whole-body image of a person sitting in a relaxed position in the driver's seat includes an image in which the person's face is not recognizably captured in the captured image and only the head is captured. The occupant recognition dictionary is trained to, for example, receive a captured image and output information indicating whether a person's face or head is captured in the second region of the captured image (hereinafter referred to as "human image presence / absence information"). The occupant recognition dictionary may also be trained to receive a captured image and output information indicating the probability that a person's face or head is captured in the second region of the captured image (hereinafter referred to as "human image probability information"). The occupant recognition dictionary is generated in advance, and the second detection unit 122 stores the occupant recognition dictionary in a buffer or a storage area.

[0025] For example, the second detection unit 122 detects an occupant imaged in the second area by inputting the captured image into the occupant recognition dictionary and obtaining human image presence / absence information. In particular, the second detection unit 122 inputs the captured image into the occupant recognition dictionary, and when human image presence / absence information indicating that a human face or head has been captured in the second area of ​​the captured image is obtained, detects that an occupant has been captured in the second area. The second detection unit 122 inputs the captured image into the occupant recognition dictionary, and when human image presence / absence information indicating that a human face or head has not been captured in the second area of ​​the captured image is obtained, detects that no occupant has been captured in the second area. Furthermore, for example, the second detection unit 122 may detect an occupant imaged in the second area by inputting the captured image into the occupant recognition dictionary and obtaining human image probability information. In detail, the second detection unit 122 inputs the captured image into the occupant recognition dictionary, and if the probability indicated in the obtained human imaging probability information is equal to or greater than a predetermined threshold value (hereinafter referred to as the "probability determination threshold value"), it detects that there is an occupant imaged in the second area, and if the probability indicated in the obtained human imaging probability information is less than the probability determination threshold value, it detects that there is no occupant imaged in the second area.

[0026] As described above, the first detection unit 121 detects an occupant whose image is captured in the first area by detecting a face in the first area. In contrast, the second detection unit 122 detects an occupant whose image is captured in the second area by using an occupant recognition dictionary. As described above, the occupant recognition dictionary has been trained to be able to detect not only a person's face but also their head. By detecting an occupant whose image is captured in the second area using the occupant recognition dictionary, the second detection unit 122 can detect the presence of the occupant even if the occupant's face is not captured in the second area in a way that allows the face to be recognized.

[0027] The second detection unit 122 outputs the detection result, whether or not it has detected the presence of an occupant in the second area, to the fraud provisional determination unit 13 as the second detection result.

[0028] In the first embodiment, the first and second regions are rectangular regions, but the first and second regions are not limited to rectangular regions. The first and second regions can be appropriately set as long as they can suppress erroneous authentication of the faces of people outside the vehicle and people in the rear seats of the vehicle, and can easily detect an occupant who is hiding in the driver's seat in an attempt to commit fraud.

[0029] The fraud provisional determination unit 13 provisionally determines whether or not there is fraud, depending on whether or not there is an occupant detected in the first area or the second area, based on the first detection result and the second detection result output from the detection unit 12. In the first embodiment, the process performed by the fraud provisional determination unit 13 to provisionally determine whether or not there is fraud, depending on whether or not there is an occupant detected in the first area or the second area, is referred to as the "fraud provisional determination process."

[0030] The fraud provisional determination unit 13 provisionally determines whether fraud has occurred based on whether pre-set conditions for provisionally determining that no fraud has occurred (hereinafter referred to as "provisional determination conditions") are met. The provisional determination conditions are set based on the presence or absence of an occupant detected in the first area or the second area. For example, an administrator or the like sets the provisional determination conditions in advance and stores them in a location that can be referenced by the fraud prevention device 1.

[0031] The provisional determination condition may be, for example, the following condition (1): Condition (1) One occupant is detected in the first area, and no occupant is detected in the second area.

[0032] The fraud provisional determination unit 13 provisionally determines that there is no fraud if the provisional determination conditions are satisfied. On the other hand, the fraud provisional determination unit 13 provisionally determines that there is fraud if the provisional determination conditions are not satisfied. Assuming that the provisional determination conditions include the above condition (1), the fraud provisional determination unit 13 provisionally determines that there is no fraud if one occupant is detected in the first area and no occupant is detected in the second area. Note that, in this case, detecting one occupant in the first area means, in detail, that only one occupant is detected in the first area. On the other hand, the fraud provisional determination unit 13 provisionally determines that there is fraud if multiple occupants are detected in the first area, if no occupant is detected in the first area, or if an occupant is detected in the second area.

[0033] The determination of whether or not there is fraud made by the provisional fraud determination unit 13 is a provisional determination of whether or not there is fraud. The final determination of whether or not there is fraud is made by the fraud determination result determination unit 14. Details of the fraud determination result determination unit 14 will be described later.

[0034] The fraud provisional judgment unit 13 outputs the provisional judgment result of the fraud provisional judgment process, that is, whether or not there is fraud (hereinafter referred to as the "fraud provisional judgment result"), to the fraud judgment result finalization unit 14. Furthermore, if the fraud provisional judgment unit 13 provisionally judges that there is no fraud as a result of the fraud provisional judgment process, it outputs information to the output device 3 instructing it to display or output a sound (hereinafter referred to as "output control information") that prompts the detection device to blow into the detection device. For example, the fraud provisional judgment unit 13 outputs output control information to the output device 3 to display or output a sound saying, "Please blow into the detection device."

[0035] The fraud determination result finalizing unit 14 performs a process to finalize the determination result of whether or not fraud has occurred. In the first embodiment, the process to finalize the determination result of whether or not fraud has occurred, performed by the fraud determination result finalizing unit 14, is referred to as the "fraud determination result finalizing process."

[0036] The fraud determination result determination process performed by the fraud determination result determination unit 14 will now be described. In the fraud determination result determination process, the fraud determination result determination unit 14 first determines whether the fraud temporary determination unit 13 has made a provisional determination that there is no fraud. The fraud determination result determination unit 14 may determine that the fraud temporary determination unit 13 has made a provisional determination that there is no fraud based on the fraud temporary determination result output from the fraud temporary determination unit 13.

[0037] When the fraud determination result confirmation unit 14 determines that the fraud provisional determination unit 13 has provisionally determined that no fraud is present, the fraud determination result confirmation unit 14 causes the image acquisition unit 11 to acquire an image, and acquires the image from the image acquisition unit 11. As described above, the imaging device 2 constantly captures images of the interior of the vehicle. The fraud determination result confirmation unit 14 acquires the latest image from the image acquisition unit 11 after the fraud provisional determination unit 13 provisionally determines that no fraud is present and outputs output control information to the output device 3, prompting the detection device to blow into the detection device. It is assumed that the image captures the test occupant blowing into the detection device. Note that the arrow from the fraud determination result confirmation unit 14 to the image acquisition unit 11 is omitted in FIG. 1 .

[0038] The fraud determination result confirmation unit 14 performs a known image analysis process on the captured image acquired by the image acquisition unit 11 to detect the area in which the face of the test occupant is captured (hereinafter referred to as the "test occupant face area"). The fraud determination result confirmation unit 14 may, for example, detect the test occupant face area by performing a known pattern matching on the captured image. Alternatively, the fraud determination result confirmation unit 14 may, for example, input the captured image into a trained model (hereinafter referred to as the "machine learning model") that receives the captured image as input and outputs the test occupant's face area, thereby detecting the test occupant face area.

[0039] After detecting the test occupant face area, the fraud determination result confirmation unit 14 compares the test occupant face area with the face area of ​​the occupant detected in the first area, and determines whether the test occupant and the occupant detected in the first area can be considered to be the same person, in other words, whether the test occupant and the occupant detected in the first area are presumed to be the same person. Note that the fraud determination result confirmation unit 14 may obtain the first detection result from the first detection unit 121 via the fraud provisional determination unit 13.

[0040] The fraud determination result determination unit 14 may determine whether the test occupant and the occupant detected in the first area can be considered to be the same person, for example, using a known image matching technique. For example, the fraud determination result determination unit 14 calculates the degree of matching by calculating the sum of differences between corresponding pixels, the number of pixels where the number of differences between pixels is equal to or less than a threshold, the correlation coefficient between images, etc., and determines that the test occupant and the occupant detected in the first area can be considered to be the same person if the calculated degree of matching is equal to or greater than a predetermined threshold. In this case, it is presumed that the test occupant and the occupant detected in the first area are the same person.

[0041] Furthermore, for example, the fraud determination result confirmation unit 14 may compare feature information of the captured image of the test occupant's face region with feature information of the captured image of the occupant's face region detected in the first region, and if the degree of match is equal to or greater than a preset threshold, determine that the test occupant and the occupant detected in the first region can be considered to be the same person. In this case, when detecting the test occupant's face region, the fraud determination result confirmation unit 14 also detects feature information of the captured image of the test occupant's face region.

[0042] If the result of comparing the face of the test occupant with the face of the occupant detected in the first area, more specifically, the result of comparing the test occupant face area with the face area of ​​the occupant detected in the first area, indicates that the test occupant and the occupant detected in the first area are the same person, the fraud determination result finalizing unit 14 finalizes the result of the fraud determination as “no fraud.” On the other hand, if the result of the fraud determination result finalizing unit 14 determines that the test occupant and the occupant detected in the first area cannot be considered to be the same person, in other words, if the test occupant and the occupant detected in the first area are not assumed to be the same person, the fraud determination result finalizing unit 14 finalizes the result of the fraud determination as “yes.”

[0043] When the provisional fraud determination unit 13 provisionally determines that fraud has occurred, the fraud determination result determination unit 14 determines that fraud has occurred as a result of the determination of whether or not fraud has occurred.

[0044] The fraud determination result finalizing unit 14 outputs a finalizing result indicating whether or not fraud has occurred (hereinafter referred to as the “fraud finalizing result”) to the interlock unit 17 .

[0045] The test result acquisition unit 15 acquires test result information from the detection device. That is, the test result acquisition unit 15 acquires the alcohol test result obtained from the alcohol test conducted by the test occupant, specifically, the test result information indicating the alcohol concentration obtained from the breath of the test occupant by blowing breath. The test result acquisition unit 15 outputs the acquired test result information to the drinking determination unit 16.

[0046] The alcohol consumption determination unit 16 determines whether the test occupant is drinking alcohol based on the test result information acquired by the test result acquisition unit 15. The alcohol consumption determination unit 16 may also have the functions of the test result acquisition unit 15. In this case, the anti-fraud device 1 does not require the test result acquisition unit 15. For example, if the alcohol concentration indicated in the test result information is equal to or greater than a predetermined threshold (hereinafter referred to as the "drinking determination threshold"), the alcohol consumption determination unit 16 determines that the test occupant is drinking alcohol. On the other hand, if the alcohol concentration indicated in the test result information is less than the drinking determination threshold, the alcohol consumption determination unit 16 determines that the test occupant is not drinking alcohol. The alcohol consumption determination unit 16 outputs the determination result of whether the test occupant is drinking alcohol (hereinafter referred to as the "drinking determination result") to the interlock unit 17.

[0047] The interlock unit 17 determines whether to release the interlock in the vehicle based on the determination result of the fraud determination result determination unit 14 as to whether or not there has been fraud (i.e., the fraud determination result) and the determination result of the drinking alcohol determination unit 16 as to whether or not the test occupant has been drinking (i.e., the drinking alcohol determination result). The interlock unit 17 releases the interlock when the fraud determination result determination unit 14 determines that there has been no fraud and the drinking alcohol determination unit 16 determines that the test occupant has not been drinking. The interlock unit 17 continues the interlock when the fraud determination result determination unit 14 does not determine that there has been no fraud or when the drinking alcohol determination unit 16 determines that the test occupant has been drinking.

[0048] An example of the operation of the anti-fraud device 1 according to the first embodiment will be described. FIG. 3 is a flowchart illustrating an example of the operation of the anti-fraud device 1 according to the first embodiment. The anti-fraud device 1 starts the operation shown in the flowchart of FIG. 3 when, for example, the vehicle is powered on. When the vehicle is powered on, the control unit of the anti-fraud device 1 causes the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking alcohol determination unit 16, and the interlock unit 17 to start processing. The anti-fraud device 1 repeats the operation shown in the flowchart of FIG. 3 until, for example, the vehicle interlock is released or instruction information instructing the suspension of the breathalyzer test is input. For example, a vehicle occupant can input instruction information from an input device (not shown) installed in the vehicle. The input device is, for example, a touch panel display provided in a navigation device. When the control unit of the anti-fraud device 1 acquires the instruction information, it causes the image acquisition unit 11, the detection unit 12, the provisional fraud judgment unit 13, the fraud judgment result confirmation unit 14, the test result acquisition unit 15, the drinking judgment unit 16, and the interlock unit 17 to terminate processing.

[0049] The image acquisition unit 11 acquires the captured image from the imaging device 2 (step ST10). The image acquisition unit 11 outputs the acquired captured image to the detection unit 12 and the fraud determination result confirmation unit 14.

[0050] The detection unit 12 performs a detection process based on the captured image acquired by the image acquisition unit 11 in step ST10 (step ST20). The detection unit 12 outputs the first detection result and the second detection result to the provisional fraud determination unit 13.

[0051] The fraud provisional determination unit 13 performs a fraud provisional determination process based on the first detection result and the second detection result output from the detection unit 12 in step ST20 (step ST30). The fraud provisional determination unit 13 outputs the fraud provisional determination result to the fraud determination result confirmation unit 14. Furthermore, if the fraud provisional determination unit 13 provisionally determines that there is no fraud as a result of performing the fraud provisional determination process, it outputs output control information to the output device 3 to cause it to display or output audio prompting the user to blow into the detection device. When the output control information is output from the fraud prevention device 1, the output device 3 displays or outputs audio prompting the user to blow into the detection device. In response to this, the test occupant blows into the detection device and performs an alcohol test.

[0052] The fraud determination result finalizing unit 14 performs a fraud determination result finalizing process (step ST40-1). The fraud determination result finalizing unit 14 outputs a fraud determination result indicating that no fraud has occurred to the interlock unit 17.

[0053] The test result acquisition unit 15 acquires the test result information from the detection device, and the drinking determination unit 16 determines whether the test occupant is drinking alcohol based on the test result information acquired by the test result acquisition unit 15 (step ST40-2). The drinking determination unit 16 outputs the drinking determination result to the interlock unit 17.

[0054] The interlock unit 17 determines whether to release the interlock in the vehicle based on the determination result of the fraud determination result determination unit 14 in step ST40-1 (i.e., the fraud determination result) and the determination result of the drinking alcohol determination unit 16 in step ST40-2 as to whether the test occupant is drinking alcohol (i.e., the drinking alcohol determination result). The interlock unit 17 releases the interlock when the fraud determination result determination unit 14 determines that there is no fraud and the drinking alcohol determination unit 16 determines that the test occupant is not drinking alcohol. The interlock unit 17 continues the interlock when the fraud determination result determination unit 14 does not determine that there is no fraud or when the drinking alcohol determination unit 16 determines that the test occupant is drinking alcohol.

[0055] 3, the processing of step ST40-1 and the processing of step ST40-2 are performed in parallel, but this is merely an example. For example, the processing of step ST40-2 may be performed after the processing of step ST40-1, or the processing of step ST40-1 may be performed after the processing of step ST40-2. It is sufficient that the processing of step ST40-1 and the processing of step ST40-2 are performed before the processing of step ST50 is performed.

[0056] 4 is a flowchart for explaining the details of step ST20 in FIG. 3 . The first detection unit 121 of the detection unit 12 detects a first region in the captured image based on the captured image acquired by the image acquisition unit 11 in step ST10, and detects an occupant captured in the detected first region (step ST101). The first detection unit 121 outputs feature information of the captured image of the acquired face region and coordinate values ​​of the face region to the fraud provisional determination unit 13 as a first detection result. The second detection unit 122 of the detection unit 12 detects a second region in the captured image based on the captured image acquired by the image acquisition unit 11 in step ST10 and the occupant recognition dictionary, and detects an occupant captured in the detected second region (step ST102). The second detection unit 122 outputs a detection result indicating whether or not an occupant is detected in the second region to the fraud provisional determination unit 13 as a second detection result.

[0057] Fig. 5 is a flowchart for explaining the details of the process of step ST30 in Fig. 3. It is assumed here that the provisional determination condition is set to "one occupant is detected in the first area and no occupant is detected in the second area."

[0058] The fraud provisional determination unit 13 determines whether or not there is one occupant detected in the first area based on the first detection result output from the first detection unit 121 (step ST111). If it determines in step ST111 that there is one occupant detected in the first area, that is, that there is only one occupant detected in the first area ("YES" in step ST111), the fraud provisional determination unit 13 determines whether or not there is no occupant detected in the second area based on the second detection result output from the second detection unit 122 (step ST112).

[0059] If it is determined in step ST112 that no occupant is detected in the second area ("YES" in step ST112), the fraud provisional determination unit 13 determines that the provisional determination condition is met and provisionally determines that no fraud is present (step ST113). The fraud provisional determination unit 13 outputs a fraud provisional determination result indicating that no fraud is present to the fraud determination result finalization unit 14. The fraud provisional determination unit 13 also outputs output control information to the output device 3 to cause a display or audio output prompting the occupant to blow into the detection device. For example, the fraud provisional determination unit 13 outputs output control information to the output device 3 to cause a display or audio output stating, "Please blow into the detection device." The test occupant blows into the detection device in accordance with the display or audio output from the output device 3 and takes the alcohol test.

[0060] If it is determined in step ST111 that there are two or more occupants detected in the first area or that there are no occupants detected in the first area (if "NO" in step ST111), and if it is determined in step ST112 that there is an occupant detected in the second area (if "NO" in step ST112), the fraud provisional determination unit 13 determines that the provisional determination conditions are not satisfied and provisionally determines that fraud has occurred (step ST114). The fraud provisional determination unit 13 outputs a fraud provisional determination result indicating that fraud has occurred to the fraud determination result confirmation unit 14.

[0061] FIG. 6 is a flowchart for explaining the details of the process of step ST40-1 in FIG.

[0062] The fraud determination result finalizing unit 14 determines whether or not the fraud provisional determination unit 13 has provisionally determined that there is fraud as a result of the fraud provisional determination process performed in step ST30 of FIG. 3 (step ST121).

[0063] If, in step ST121, the fraud provisional judgment unit 13 judges that there is no fraud (if the answer is "NO" in step ST121), the fraud judgment result confirmation unit 14 causes the image acquisition unit 11 to acquire an image, and performs a known image analysis process on the image acquired by the image acquisition unit 11 to detect the facial area of ​​the occupant undergoing the inspection (step ST122).

[0064] The fraud determination result confirmation unit 14 compares the face area of ​​the test subject occupant detected in step ST122 with the face area of ​​the occupant detected in the first area in step ST20 of Figure 3, and determines whether the test subject occupant and the occupant detected in the first area are presumed to be the same person (step ST123).

[0065] If it is determined in step ST123 that the inspected occupant and the occupant detected in the first area are presumed to be the same person (if "YES" in step ST123), the fraud determination result finalizing unit 14 determines that there is no fraud (step ST124). The fraud determination result finalizing unit 14 outputs a fraud determination result indicating that there is no fraud to the interlock unit 17.

[0066] On the other hand, if the fraud provisional determination unit 13 determines in step ST121 that fraud has been provisionally determined to have occurred (if "YES" in step ST121), and if it determines in step ST123 that the test occupant and the occupant detected in the first area cannot be assumed to be the same person (if "NO" in step ST123), the fraud determination result finalization unit 14 determines that fraud has occurred (step ST125). The fraud determination result finalization unit 14 outputs a fraud determination result indicating that fraud has occurred to the interlock unit 17.

[0067] In the above example of operation of the anti-fraud device 1 explained using the flowchart in Figure 3, the anti-fraud device 1 detects an occupant whose image is captured in the first area and an occupant whose image is captured in the second area before starting the provisional fraud determination process. However, this is merely one example, and the anti-fraud device 1 may, for example, start the provisional fraud determination process when an occupant whose image is captured in the first area is detected, and if the provisional fraud determination process determines that there is only one occupant detected in the first area, detect the occupant whose image is captured in the second area.

[0068] 7 is a flowchart showing an example of operation in which the anti-fraud device 1 starts a provisional fraud determination process when an occupant imaged in the first area is detected, and if the provisional fraud determination process determines that there is only one occupant detected in the first area, detects an occupant imaged in the second area in embodiment 1. In the flowchart shown in Fig. 7, the same steps as those shown in the flowcharts shown in Fig. 3, 4, or 5 are assigned the same step numbers, and duplicated explanations of the details of the processes are omitted.

[0069] As shown in the flowchart of Figure 7, in step ST101, the first detection unit 121 of the detection unit 12 detects a first area in the captured image based on the captured image acquired by the image acquisition unit 11 in step ST10, and when it detects an occupant imaged in the detected first area, the fraud provisional judgment unit 13 determines whether there is one occupant detected in the first area based on the first detection result output from the first detection unit 121 (step ST111).

[0070] If the fraud provisional determination unit 13 determines in step ST111 that only one occupant is detected in the first area ("YES" in step ST111), the second detection unit 122 of the detection unit 12 detects a second area in the captured image based on the captured image acquired by the image acquisition unit 11 in step ST10 and the occupant recognition dictionary, and detects the occupant imaged in the detected second area (step ST102). Then, the fraud provisional determination unit 13 determines whether or not an occupant is detected in the second area based on the second detection result output from the second detection unit 122 (step ST112). If the fraud provisional determination unit 13 determines that two or more occupants are detected in the first area or that no occupant is detected in the first area ("NO" in step ST111), the second detection unit 122 does not detect the second area or detect an occupant in the second area.

[0071] As explained using the flowchart shown in Figure 7, the anti-fraud device 1 detects the second area and the occupant in the second area only when the fraud provisional judgment unit 13 determines that there is only one occupant detected in the first area, thereby preventing unnecessary processing from being executed and reducing the processing load.

[0072] For example, a fraudulent attempt to evade a breathalyzer test administered to a vehicle driver may involve the driver changing their position while seated in the driver's seat, positioning their face where the driver's face should be, and then acting as if they were seated in the driver's seat to pose as the driver and administer the breathalyzer test. FIG. 8 shows an example of a captured image of the interior of a vehicle in which a passenger other than the driver is attempting to evade the breathalyzer test by acting as if they were seated in the driver's seat and administering the breathalyzer test. In FIG. 8 , the captured image is indicated by "I." The driver's seat, the driver's seat, the passenger seat, and the rear seat passenger are indicated by "S1," "D," "S2," and "Pb," respectively. In this example, the vehicle is a right-hand drive vehicle, but for convenience, the steering wheel and other interior structures other than the seats are omitted from FIG. 8 . FIG. 8 shows an example of a captured image of a driver (indicated by "D" in FIG. 8 ) leaning forward in the driver's seat, hiding his or her face, while a rear-seat passenger leans forward and places his or her face near the headrest of the driver's seat, where the driver's face is expected to be located. It is conceivable that a rear-seat passenger may pose as the driver and conduct a breathalyzer test in a vehicle cabin such as that shown in FIG. 8 . In this case, as shown in FIG. 8 , the driver may sag and hide his or her face, and the captured image may recognize the driver's head but not his or her face. In other words, it is conceivable that the outline of the driver's face cannot be extracted from the captured image. If the contour of the face were recognized and a person's image were extracted and counted from a captured image of a driver who has shifted his or her seat, the driver's posture would not be counted in the captured image shown in FIG. 8 , which may result in the inability to completely prevent the rear-seat passenger from impersonating the driver.

[0073] In contrast, as described above, the anti-fraud device 1 detects an occupant (more specifically, the occupant's face) who is imaged in a first area around the headrest of the driver's seat in an image captured by the imaging device 2, which is provided in a vehicle and is capable of capturing an image of at least the area around the driver's seat, including the foot area, and an occupant (more specifically, the occupant's face or head) who is imaged in a second area in the image that is below the headrest of the driver's seat and includes the area around the driver's seat. The anti-fraud device 1 detects an occupant who is imaged in the second area by a method that is different from the method for detecting an occupant who is imaged in the first area and that is capable of detecting the presence of the occupant even if the occupant's face is not imaged in a recognizable manner. The anti-fraud device 1 then performs a provisional fraud determination process. If one occupant is detected in the first area and no occupant is detected in the second area, the anti-fraud device 1 provisionally determines that there is no fraud. If the provisional determination determines that there is no fraud, and the device detects the face of the test occupant performing the alcohol test based on the captured image and compares the face of the test occupant with the face of the occupant detected in the first area, the device determines that there is no fraud. This allows the anti-fraud device 1 to prevent fraud in breathalyzer tests administered to drivers of vehicles. Specifically, the anti-fraud device 1 can prevent fraud in breathalyzer tests administered to drivers of vehicles, where an occupant other than the driver impersonates the driver and the driver attempts to evade the breathalyzer test. For example, even in cases where a driver changes their position while sitting in the driver's seat, and a passenger other than the driver positions their face where the driver's face should be, and the passenger other than the driver acts as if they are sitting in the driver's seat, and impersonates the driver to conduct a breathalyzer test, making it impossible to extract the outline of the driver who has changed their position in the image captured of the driver's seat, the fraud prevention device 1 can detect the presence of the driver and prevent fraud.

[0074] 9A and 9B are diagrams illustrating an example of the hardware configuration of the anti-fraud device 1 according to embodiment 1. In embodiment 1, the functions of the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking alcohol determination unit 16, the interlock unit 17, and a control unit (not shown) are realized by a processing circuit 1001. That is, the anti-fraud device 1 includes a processing circuit 1001 for performing control to prevent fraud in a breathalyzer test administered to a vehicle driver. The processing circuit 1001 may be dedicated hardware as shown in FIG. 9A, or a processor 1004 that executes a program stored in memory as shown in FIG. 9B.

[0075] If the processing circuit 1001 is dedicated hardware, the processing circuit 1001 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof.

[0076] When the processing circuit is the processor 1004, the functions of the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, and a control unit (not shown) are realized by software, firmware, or a combination of software and firmware. The software or firmware is written as a program and stored in the memory 1005. The processor 1004 reads and executes the program stored in the memory 1005 to execute the functions of the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, and a control unit (not shown). In other words, the anti-tampering device 1 includes a memory 1005 for storing a program that, when executed by the processor 1004, results in the execution of steps ST10 to ST50 of FIG. 3 or steps ST10 to ST50 of FIG. 7. In addition, the program stored in memory 1005 can also be said to cause the computer to execute the processing procedures or methods of the image acquisition unit 11, the detection unit 12, the provisional fraud judgment unit 13, the fraud judgment result confirmation unit 14, the test result acquisition unit 15, the drinking judgment unit 16, the interlock unit 17, and the control unit not shown. Here, the memory 1005 may be, for example, a non-volatile or volatile semiconductor memory such as a RAM, a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), or an EEPROM (Electrically Erasable Programmable Read-Only Memory), or a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, or a DVD (Digital Versatile Disc).

[0077] The functions of the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, and the control unit (not shown) may be partially implemented by dedicated hardware and partially implemented by software or firmware. For example, the image acquisition unit 11 may be implemented by a processing circuit 1001 as dedicated hardware, while the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, and the control unit (not shown) may be implemented by a processor 1004 reading and executing programs stored in a memory 1005. The anti-fraud device 1 also includes an input interface unit 1002 and an output interface unit 1003 that communicate with devices such as the imaging device 2 or the output device 3 via wired or wireless communication.

[0078] In the above-described first embodiment, the test result acquisition unit 15, the drinking determination unit 16, and the interlock unit 17 are provided in the anti-tamper device 1. However, the test result acquisition unit 15, the drinking determination unit 16, and the interlock unit 17 are not essential components of the anti-tamper device 1. The test result acquisition unit 15, the drinking determination unit 16, and the interlock unit 17 may be provided in a device connected to the anti-tamper device 1 outside the anti-tamper device 1, for example.

[0079] In the first embodiment, the fraud prevention device 1 is an on-board device mounted on a vehicle, and the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, and a control unit (not shown) are provided in the on-board device. However, the fraud prevention system 100 may be configured such that some of the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, and a control unit (not shown) are mounted in the on-board device of the vehicle, and the rest are provided in a server connected to the on-board device via a network, with the on-board device, the server, and the image capture device 2 constituting the fraud prevention system 100. In addition, the image acquisition unit 11, the detection unit 12, the provisional fraud judgment unit 13, the fraud judgment result confirmation unit 14, the test result acquisition unit 15, the drinking judgment unit 16, the interlock unit 17, and the control unit (not shown) may all be provided on the server.

[0080] As described above, according to the first embodiment, the fraud prevention device 1 includes an image acquisition unit 11 that acquires an image captured by an imaging device 2 that is provided in a vehicle so as to be able to capture an image of the periphery of the driver's seat including at least the foot area, a detection unit 12 that detects an occupant who is captured in a first area around the headrest of the driver's seat in the captured image and an occupant who is captured in a second area below the headrest of the driver's seat in the captured image including the foot area of ​​the driver's seat, based on the captured image acquired by the image acquisition unit 11, and a fraud provisional determination process that provisionally determines whether or not fraud has occurred based on the presence or absence of an occupant detected in the first area or the second area. The anti-fraud device 1 is configured to include a fraud provisional determination unit 13 that provisionally determines that there is no fraud if one occupant is detected in the first area and no occupant is detected in the second area, and a fraud determination result confirmation unit 14 that performs a fraud determination result confirmation process to confirm the determination result of whether or not there is fraud.If the fraud provisional determination unit 13 provisionally determines that there is no fraud and detects the face of a test occupant performing an action to conduct an alcohol test based on the captured image acquired by the image acquisition unit 11, and compares the face of the test occupant with the face of the occupant detected in the first area, it is estimated that the test occupant and the occupant detected in the first area are the same person.Therefore, the anti-fraud device 1 can prevent fraud in a breathalyzer test administered to a driver of a vehicle.In particular, the anti-fraud device 1 can prevent fraud in a breathalyzer test administered to a driver of a vehicle, where an occupant other than the driver impersonates the driver and the driver attempts to evade the breathalyzer test.

[0081] Second Embodiment In a second embodiment, an anti-tamper device will be described that has a personal authentication function in addition to the functions described in the first embodiment.

[0082] FIG. 10 is a diagram showing an example of the configuration of an anti-fraud device 1a according to embodiment 2. The anti-fraud device 1a is connected to an imaging device 2 and an output device 3. In embodiment 2, the anti-fraud device 1a and the imaging device 2 constitute an anti-fraud system 100a. Note that this is merely an example, and for example, the anti-fraud system 100a may also be constituted by the anti-fraud device 1a, the imaging device 2, and the output device 3. In embodiment 2, the anti-fraud device 1a is mounted on, for example, a vehicle (not shown).

[0083] 10, the same reference numerals are used to designate configuration examples similar to those of the anti-fraud device 1 described in embodiment 1 using FIG. 1, and duplicated explanations will be omitted. The anti-fraud device 1a according to embodiment 2 differs from the anti-fraud device 1 according to embodiment 1 in that it includes a personal authentication unit 18.

[0084] When the fraud temporary determination unit 13 provisionally determines that there is no fraud, the personal authentication unit 18 performs personal authentication of the occupant detected in the first area by the first detection unit 121. In the second embodiment, the process of personal authentication of the occupant detected in the first area performed by the personal authentication unit 18 is referred to as the "personal authentication process." Note that the personal authentication unit 18 may determine that the fraud temporary determination unit 13 has provisionally determined that there is no fraud based on the fraud temporary determination result output from the fraud temporary determination unit 13. In the second embodiment, the fraud temporary determination unit 13 outputs the fraud temporary determination result to the personal authentication unit 18. Furthermore, the personal authentication unit 18 may obtain the first detection result from the first detection unit 121 via the fraud temporary determination unit 13.

[0085] The personal authentication unit 18 performs personal authentication, for example, by comparing a pre-registered facial image of the driver with a captured image of the facial area of ​​the occupant detected in the first area. For example, the driver may store his or her own facial image in advance in a storage device (not shown) provided in a location accessible to the anti-fraud device 1. Note that this is merely an example, and the personal authentication unit 18 may perform personal authentication of the occupant detected in the first area using a known personal authentication technology.

[0086] If the personal authentication unit 18 can personally authenticate the occupant detected in the first area, it maintains the provisional determination result of whether or not there is any fraud made by the fraud provisional determination unit 13. That is, the personal authentication unit 18 maintains the provisional determination result that the fraud provisional determination unit 13 has provisionally determined that there is no fraud. The personal authentication unit 18 outputs the provisional determination result output from the fraud provisional determination unit 13 to the fraud determination result confirmation unit 14.

[0087] On the other hand, if the personal authentication unit 18 cannot authenticate the occupant detected in the first area, it determines that there is fraud, regardless of the provisional determination result of whether or not there is fraud made by the fraud provisional determination unit 13, and rewrites the provisional determination result output from the fraud provisional determination unit 13 to a provisional determination result indicating that there is fraud. The personal authentication unit 18 outputs the provisional determination result indicating that there is fraud to the fraud determination result confirmation unit 14.

[0088] In the second embodiment, the fraud determination result determination unit 14 performs a fraud determination result determination process based on the provisional fraud determination result output from the personal authentication unit 18. When the personal authentication unit 18 has authenticated the occupant detected in the first area, and when the fraud determination result determination unit 14 detects the face of the test occupant performing the operation of conducting an alcohol test based on the captured image acquired by the image acquisition unit 11 and compares the face of the test occupant with the face of the occupant detected in the first area, the fraud determination result determination unit 14 determines that there is no fraud if it is estimated that the test occupant and the occupant detected in the first area are the same person. In other words, when the personal authentication unit 18 has authenticated the occupant detected in the first area, this means that the personal authentication unit 18 has output a provisional determination result indicating that there is no fraud.

[0089] The fraud determination result confirmation unit 14 determines that there is fraud if the personal authentication unit 18 does not authenticate the occupant detected in the first area, in other words, if the personal authentication unit 18 outputs a provisional determination result indicating that there is fraud.

[0090] An example of the operation of the anti-fraud device 1a according to the second embodiment will be described. FIG. 11 is a flowchart for describing an example of the operation of the anti-fraud device 1a according to the second embodiment. The specific content of the processing of the anti-fraud device 1a shown in the flowchart of FIG. 11 is the same as the specific content of the processing of the anti-fraud device 1 described using the flowchart of FIG. 3 in the first embodiment, and the same step numbers are assigned to those content. The anti-fraud device 1a starts the operation shown in the flowchart of FIG. 11 when, for example, the vehicle is powered on. When the vehicle is powered on, the control unit of the anti-fraud device 1a causes the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, and the personal authentication unit 18 to start processing. The anti-fraud device 1a repeats the operation shown in the flowchart of FIG. 11 until, for example, the vehicle interlock is released or instruction information instructing the suspension of the breathalyzer test is input. When the control unit of the anti-fraud device 1a acquires the instruction information, it causes the image acquisition unit 11, the detection unit 12, the provisional fraud judgment unit 13, the fraud judgment result confirmation unit 14, the test result acquisition unit 15, the drinking judgment unit 16, the interlock unit 17, and the personal authentication unit 18 to terminate processing.

[0091] The image acquisition unit 11 acquires the captured image from the imaging device 2 (step ST10). The image acquisition unit 11 outputs the acquired captured image to the detection unit 12 and the fraud determination result confirmation unit 14.

[0092] The detection unit 12 performs a detection process based on the captured image acquired by the image acquisition unit 11 in step ST10 (step ST20).

[0093] The fraud provisional determination unit 13 performs a fraud provisional determination process based on the first detection result and the second detection result output from the detection unit 12 in step ST20 (step ST30). The fraud provisional determination unit 13 outputs the fraud provisional determination result to the personal authentication unit 18. Furthermore, if the fraud provisional determination unit 13 provisionally determines that there is no fraud as a result of performing the fraud provisional determination process, it outputs output control information to the output device 3 to cause it to display or output audio prompting the user to blow into the detection device. When the output control information is output from the fraud prevention device 1, the output device 3 displays or outputs audio prompting the user to blow into the detection device. In response to this, the test occupant blows into the detection device and performs an alcohol test.

[0094] If the fraud provisional determination unit 13 provisionally determines that there is no fraud, the personal authentication unit 18 performs personal authentication processing (step ST35). The personal authentication unit 18 outputs the provisional determination result to the fraud determination result confirmation unit 14.

[0095] The fraud determination result determining unit 14 performs a fraud determination result determining process (step ST40-1). The fraud determination result determining unit 14 outputs the fraud determination result to the interlock unit 17.

[0096] The test result acquisition unit 15 acquires the test result information from the detection device, and the drinking determination unit 16 determines whether the test occupant is drinking alcohol based on the test result information acquired by the test result acquisition unit 15 (step ST40-2). The drinking determination unit 16 outputs the drinking determination result to the interlock unit 17.

[0097] The interlock unit 17 decides whether or not to release the interlock in the vehicle based on the determination result of whether or not there is fraud by the fraud determination result determination unit 14 in step ST40-1 (i.e., the fraud determination result) and the determination result of whether or not the test occupant is in a drunk state by the drinking determination unit 16 in step ST40-2 (i.e., the drinking determination result) (step ST50).

[0098] Fig. 12 is a flowchart for explaining the details of the process of step ST35 in Fig. 11. As a result of the fraud provisional determination process performed by the fraud provisional determination unit 13 in step ST30 in Fig. 11, it is determined whether or not a provisional determination has been made that fraud has occurred (step ST131).

[0099] If the fraud provisional determination unit 13 determines that there is no fraud ("NO" in step ST131), the personal authentication unit 18 performs personal authentication of the occupant detected in the first area (step ST132). If personal authentication is successful ("YES" in step ST132), the personal authentication unit 18 maintains the provisional determination result that the fraud provisional determination unit 13 has provisionally determined that there is no fraud (step ST133). The personal authentication unit 18 outputs the provisional determination result output from the fraud provisional determination unit 13 to the fraud determination result confirmation unit 14.

[0100] If the fraud provisional judgment unit 13 judges in step ST30 of FIG. 11 that fraud has been provisionally judged to exist (if "YES" in step ST131), or if personal authentication is not possible (if "NO" in step ST132), the personal authentication unit 18 provisionally judges that fraud has existed (step ST134) and outputs the provisional judgment result indicating that fraud has been provisionally judged to exist to the fraud judgment result confirmation unit 14.

[0101] In the anti-fraud device 1a according to the second embodiment, similarly to the anti-fraud device 1 according to the first embodiment, for example, when an occupant imaged in the first area is detected, a fraud provisional determination process may be initiated, and if the fraud provisional determination process determines that there is only one occupant detected in the first area, an occupant imaged in the second area may be detected. In this case, an example of the operation of the anti-fraud device 1a is an example of the operation of the anti-fraud device 1a according to the first embodiment, in which the process of step ST35 described in the flowchart shown in FIG. 11 is added after the process of step ST113 and the process of step ST114 and before the process of step ST40-1 and the process of step ST40-1. The anti-fraud device 1a detects the second area and the occupant in the second area only when the fraud provisional determination unit 13 determines that there is only one occupant detected in the first area, thereby preventing unnecessary processing and reducing the processing load.

[0102] In this way, the anti-fraud device 1a detects an occupant (more specifically, the occupant's face) captured in a first area around the headrest of the driver's seat in an image captured by the imaging device 2, which is provided in a vehicle and capable of capturing images of at least the area around the driver's seat, including the foot area, and an occupant (more specifically, the occupant's face or head) captured in a second area below the headrest of the driver's seat in the image, including the area around the driver's seat. The anti-fraud device 1 detects an occupant captured in the second area using a method that is different from the method used to detect an occupant captured in the first area and that is capable of detecting the presence of the occupant even if the occupant's face is not recognizably captured in the image. The anti-fraud device 1a then performs a provisional fraud determination process, and if one occupant is detected in the first area and no occupant is detected in the second area, it provisionally determines that no fraud has occurred. Furthermore, if the anti-fraud device 1a provisionally determines that no fraud has occurred, it performs personal authentication of the occupant detected in the first area. When the anti-fraud device 1a performs personal authentication on the occupant detected in the first area, and detects the face of the test occupant performing the alcohol test based on the captured image and compares the face of the test occupant with the face of the occupant detected in the first area, if it is estimated that the test occupant and the occupant detected in the first area are the same person, the anti-fraud device 1a determines that there is no fraud. This allows the anti-fraud device 1a to prevent fraud in breathalyzer tests administered to drivers of vehicles. Specifically, the anti-fraud device 1a can prevent fraud in breathalyzer tests administered to drivers of vehicles, where an occupant other than the driver impersonates the driver and the driver attempts to evade the breathalyzer test. For example, even in cases where a driver changes his / her position while sitting in the driver's seat, and a passenger other than the driver positions his / her face where the driver's face should be, and the passenger other than the driver behaves as if he / she were sitting in the driver's seat, and performs a breathalyzer test while impersonating the driver, making it impossible to extract the outline of the driver who has changed his / her position in the captured image of the driver's seat, the fraud prevention device 1a can detect the presence of the driver and prevent fraud.Furthermore, when the fraud prevention device 1a performs a fraud provisional judgment process and provisionally judges that there is no fraud, it further performs personal authentication of the occupant detected in the first area, and if the personal authentication is successful and it is inferred that the occupant conducting the test and the occupant detected in the first area are the same person, it determines that there is no fraud.This more reliably prevents fraud in the breathalyzer test administered to the driver of a vehicle, where an occupant other than the driver impersonates the driver and the driver attempts to evade the breathalyzer test, compared to when personal authentication is not performed.

[0103] For example, if a vehicle is driven by only one person on a daily basis, having the anti-fraud device 1a with a personal authentication function as described above can more reliably prevent illicit behavior in which a passenger other than the driver attempts to circumvent a breathalyzer test by impersonating the driver. On the other hand, if a vehicle is driven by more than one person on a daily basis, or if a vehicle is driven by a fixed person on a daily basis but other people may drive, having the anti-fraud device 1a with a personal authentication function may result in a provisional determination of illicit behavior even when there is no illicit behavior, and then the determination of illicit behavior may be confirmed, resulting in the interlock not being released. Therefore, if a vehicle is driven by more than one person on a daily basis, or if a vehicle is driven by a fixed person on a daily basis but other people may drive, using the anti-fraud device 1 according to embodiment 1, which makes a provisional determination of illicit behavior and a confirmation of illicit behavior without performing personal authentication, can more flexibly prevent illicit behavior in a breathalyzer test administered to the vehicle driver by a passenger other than the driver attempting to circumvent the breathalyzer test by impersonating the driver.

[0104] The hardware configuration of the anti-fraud device 1a according to embodiment 2 is the same as the hardware configuration of the anti-fraud device 1 described in embodiment 1 using Figures 9A and 9B, and is therefore not shown. In embodiment 2, the functions of the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, the personal authentication unit 18, and a control unit (not shown) are realized by a processing circuit 1001. That is, the anti-fraud device 1a includes a processing circuit 1001 for performing control to prevent fraud in a breathalyzer test administered to a driver in a vehicle. The processing circuit 1001 may be dedicated hardware as shown in Figure 9A, or a processor 1004 that executes a program stored in a memory 1005 as shown in Figure 9B.

[0105] The processing circuit 1001 reads and executes a program stored in the memory 1005, thereby performing the functions of the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, the personal authentication unit 18, and a control unit (not shown). That is, the anti-fraud device 1a includes a memory 1005 for storing a program that, when executed by the processing circuit 1001, results in the execution of steps ST10 to ST50 of FIG. 11 described above. The program stored in the memory 1005 can also be said to cause a computer to execute the processing procedures or methods of the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, the personal authentication unit 18, and a control unit (not shown). The anti-tamper device 1a includes an input interface device 1002 and an output interface device 1003 that perform wired or wireless communication with devices such as an imaging device 2 or an output device 3.

[0106] In the second embodiment described above, when the fraud provisional determination unit 13 performs the fraud provisional determination process and provisionally determines that there is no fraud, it outputs output control information to the output device 3 to cause a display or audio output prompting the occupant to blow into the detection device. However, this is merely an example, and in the second embodiment, when the personal authentication unit 18 has successfully authenticated the individual of the occupant imaged in the first area, it may also output output control information to the output device 3 to cause a display or audio output prompting the occupant to blow into the detection device.

[0107] In the second embodiment, the test result acquisition unit 15, the drinking determination unit 16, and the interlock unit 17 are provided in the anti-tamper device 1a. However, the test result acquisition unit 15, the drinking determination unit 16, and the interlock unit 17 are not essential components of the anti-tamper device 1a. The test result acquisition unit 15, the drinking determination unit 16, and the interlock unit 17 may be provided in a device connected to the anti-tamper device 1a outside the anti-tamper device 1a, for example.

[0108] In the second embodiment, the fraud prevention device 1a is an on-board device mounted on a vehicle, and the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, the personal authentication unit 18, and a control unit (not shown) are provided in the on-board device. However, the fraud prevention system 100a may be configured such that some of the image acquisition unit 11, the detection unit 12, the provisional fraud determination unit 13, the fraud determination result confirmation unit 14, the test result acquisition unit 15, the drinking determination unit 16, the interlock unit 17, the personal authentication unit 18, and a control unit (not shown) are mounted on the on-board device of the vehicle, and the rest are provided in a server connected to the on-board device via a network, with the on-board device, the server, and the image capture device 2. In addition, the image acquisition unit 11, detection unit 12, provisional fraud judgment unit 13, fraud judgment result confirmation unit 14, test result acquisition unit 15, drinking judgment unit 16, interlock unit 17, personal authentication unit 18, and a control unit not shown may all be provided on the server.

[0109] As described above, according to embodiment 2, the anti-fraud device 1a is equipped with a personal authentication unit 18 that performs personal authentication of the occupant detected in the first area when the provisional fraud judgment unit 13 provisionally judges that there is no fraud, and the fraud judgment result confirmation unit 14 is configured to confirm that there is no fraud when the personal authentication unit 18 authenticates the occupant detected in the first area and detects the face of the test-performing occupant who is performing the operation of conducting an alcohol test based on the captured image acquired by the image acquisition unit 11, and compares the face of the test-performing occupant with the face of the occupant detected in the first area, and if it is presumed that the test-performing occupant and the occupant detected in the first area are the same person. Therefore, when the fraud prevention device 1a performs a fraud provisional judgment process and provisionally judges that there is no fraud, it further performs personal authentication of the occupant detected in the first area, and if the personal authentication is successful and it is inferred that the occupant conducting the test and the occupant detected in the first area are the same person, it determines that there is no fraud.This more reliably prevents fraud in the breathalyzer tests administered to both drivers, in which an occupant other than the driver impersonates the driver and the driver attempts to evade the breathalyzer test, compared to when personal authentication is not performed.

[0110] In the anti-fraud device 1 according to the first embodiment and the anti-fraud device 1a according to the second embodiment, the condition for provisional judgment used by the fraud provisional judgment unit 13 when performing the fraud provisional judgment process is set to the above-mentioned condition (1). That is, in the first and second embodiments, the fraud provisional judgment unit 13 performs the fraud provisional judgment process to provisionally judge whether or not there is fraud based on the presence or absence of an occupant detected in the first area or the second area, and provisionally judges that there is no fraud if one occupant is detected in the first area and no occupant is detected in the second area. However, this is merely an example, and a condition other than the above-mentioned condition (1) may be set as the condition for provisional judgment.

[0111] For example, the provisional determination condition may be set as the following condition (2): Condition (2) One occupant is detected in the first area, no occupant is detected in the second area, and an image of a seat belt is captured in the second area.

[0112] That is, in the above-described first or second embodiment, the fraud provisional determination unit 13 may perform a fraud provisional determination process and provisionally determine that there is no fraud if one occupant is detected in the first area, no occupant is detected in the second area, and a seat belt is captured in the second area. For example, the fraud provisional determination unit 13 may perform a known image recognition process on the captured image of the second area based on the second detection result obtained from the second detection unit 122 to determine whether a seat belt is captured in the second area. In this case, the second detection unit 122 may output, as the second detection result, the detection result indicating whether an occupant is detected in the second area and coordinates indicating the second area in the captured image to the fraud provisional determination unit 13. The coordinates indicating the second area in the captured image are, for example, the coordinates of the four corners of the second area. For example, the fraud provisional determination unit 13 may determine whether a seat belt is captured in the second area by performing pattern matching or by using a machine learning model that inputs a captured image and outputs information indicating whether a seat belt is captured in the second area of ​​the captured image.

[0113] In this case, the fraud provisional determination unit 13 determines whether or not an image of a seat belt is captured in the second area after, for example, step ST112 becomes "YES" in the operation described using the flowchart shown in Fig. 5. If the fraud provisional determination unit 13 determines that an image of a seat belt is captured in the second area, it provisionally determines that no fraud has occurred (step ST113), and if it determines that an image of a seat belt has not been captured in the second area, it provisionally determines that there has been fraud (step ST114).

[0114] By setting condition (2) in the provisional judgment condition, the fraud prevention device 1 or 1a can make a stricter provisional judgment against, for example, impersonation of the driver by a passenger other than the driver than when condition (1) is set in the provisional judgment condition. In other words, the fraud prevention device 1 or 1a can make a provisional judgment on the presence or absence of fraud with greater accuracy.

[0115] In the first or second embodiment, the provisional determination condition may be set to a condition such as the following condition (3): Condition (3) One occupant is detected in the first area, no occupant is detected in the second area, and the seat belt of the driver's seat is fastened.

[0116] That is, in the above-described first or second embodiment, the fraud provisional determination unit 13 may perform a fraud provisional determination process, and provisionally determine that there is no fraud if one occupant is detected in the first area, no occupant is detected in the second area, and the driver's seat belt is fastened. The fraud provisional determination unit 13 may, for example, acquire a status indicating the fastening state of the driver's seat belt from a seat belt sensor (not shown) mounted in the vehicle, and determine that the driver's seat belt is fastened.

[0117] In this case, the fraud provisional determination unit 13 determines whether or not the seat belt of the driver's seat is fastened, for example, after the result of step ST112 becomes "YES" in the operation described using the flowchart shown in Fig. 5. If the fraud provisional determination unit 13 determines that the seat belt of the driver's seat is fastened, it provisionally determines that there is no fraud (step ST113), and if it determines that the seat belt has not been imaged in the second area, it provisionally determines that there is fraud (step ST114).

[0118] Generally, determining whether a seat belt is fastened based on information detected by a seat belt sensor provides a more reliable determination result than determining whether a seat belt is fastened based on whether an image of the seat belt is captured based on an image. By setting condition (3) in the provisional determination condition, the fraud prevention device 1 or the fraud prevention device 1a can make a stricter provisional determination against driver impersonation by a passenger other than the driver than when condition (2) is set in the provisional determination condition. In other words, the fraud prevention device 1 or the fraud prevention device 1a can make a provisional determination of whether fraud is occurring with greater accuracy.

[0119] Furthermore, in the above-mentioned embodiment 1 or embodiment 2, the anti-fraud device 1 or anti-fraud device 1a may be configured to include a re-examination execution unit 19 (see Figure 13 described below) that, when predetermined requirements (hereinafter referred to as "re-examination execution requirements") are met, causes the detection unit 12 to detect an occupant in the first area, the detection unit 12 to detect an occupant in the second area, the fraud provisional judgment process by the fraud provisional judgment unit 13, and the fraud judgment result confirmation process by the fraud judgment result confirmation unit 14 to be performed again.

[0120] Here, Figure 13 is a diagram showing an example configuration of the anti-fraud device 1 in embodiment 1 that includes a re-examination execution unit 19. Note that Figure 13 shows an example configuration of the anti-fraud device 1 in embodiment 1 that includes a re-examination execution unit 19, in other words, an example configuration of the anti-fraud device 1 in which the re-examination execution unit 19 has been added to the example configuration shown in Figure 1, but in embodiment 2, the anti-fraud device 1a may also be configured to include the re-examination execution unit 19. In this case, the example configuration of the anti-fraud device 1a is the example configuration of the anti-fraud device 1a shown in Figure 10 in which the re-examination execution unit 19 has been added.

[0121] The reinspection execution requirement is set, for example, as the following reinspection execution requirement (1) or reinspection execution requirement (2). The reinspection execution requirement may have multiple requirements set, such as the following reinspection execution requirement (1) and reinspection execution requirement (2). The reinspection execution requirement is set in advance as appropriate, for example, by an administrator, and is stored in a location that can be referenced by the fraud prevention device 1 or fraud prevention device 1a. Reinspection execution requirement (1): The vehicle door is opened Reinspection execution requirement (2): The driver's seat belt is released

[0122] The re-inspection execution unit 19 may detect the opening and closing of the vehicle door by acquiring information on the opening and closing of the door from a door sensor (not shown) mounted on the vehicle, for example. The re-inspection execution unit 19 may also detect the unfastening of the driver's seat belt by acquiring information indicating the seat belt fastening state from a seat belt sensor on the driver's seat, for example.

[0123] In this case, in the anti-fraud device 1 according to embodiment 1, the retest execution unit 19 monitors whether the retest execution requirements have been met while performing the operation shown in the flowchart of Fig. 3 or the operation shown in the flowchart of Fig. 7, and if it determines that the retest execution requirements have been met, causes the detection unit 12 to again detect an occupant in the first area, the detection unit 12 to again detect an occupant in the second area, the fraud provisional determination unit 13 to again perform the fraud provisional determination process, and the fraud determination result determination unit 14 to again perform the fraud prevention device 1 repeats the operation shown in the flowchart of Fig. 3 or the operation shown in the flowchart of Fig. 7, starting from step ST10.

[0124] In this case, in the anti-fraud device 1a according to the second embodiment, the retest execution unit 19 monitors whether the retest execution requirements are met while performing the operations shown in the flowchart of Fig. 11, and if it determines that the retest execution requirements are met, causes the detection unit 12 to again detect an occupant in the first area, the detection unit 12 to again detect an occupant in the second area, the provisional fraud determination unit 13 to again perform the provisional fraud determination process, the personal authentication unit 18 to again perform the personal authentication process, and the fraud determination result determination process by the fraud determination result determination unit 14. The anti-fraud device 1a repeats the operations shown in the flowchart of Fig. 11 from step ST10.

[0125] In this way, the anti-fraud device 1 may be configured to include a retest execution unit 19 that, when the retest execution requirements are met, causes the detection unit 12 to detect an occupant in the first area, the detection unit 12 to detect an occupant in the second area, the fraud provisional determination process by the fraud provisional determination unit 13, and the fraud determination result determination process by the fraud determination result determination unit 14 to be performed again. The anti-fraud device 1a may also be configured to include a retest execution unit 19 that, when the retest execution requirements are met, causes the detection unit 12 to detect an occupant in the first area, the detection unit 12 to detect an occupant in the second area, the fraud provisional determination process by the fraud provisional determination unit 13, the personal authentication process by the personal authentication unit 18, and the fraud determination result determination process by the fraud determination result determination unit 14 to be performed again. In this way, the anti-fraud device 1 or the anti-fraud device 1a can detect when it is suspected that fraud has occurred, such as an occupant other than the driver impersonating the driver, during the breathalyzer test, and cause the breathalyzer test to be rerun. As a result, the fraud prevention device 1 or the fraud prevention device 1a can more accurately prevent fraud in a breathalyzer test administered to a driver of a vehicle. In detail, the fraud prevention device 1 or the fraud prevention device 1a can more accurately prevent fraud in a breathalyzer test administered to a driver of a vehicle, in which a passenger other than the driver impersonates the driver and the driver tries to evade the breathalyzer test.

[0126] The hardware configuration of the anti-fraud device 1 or anti-fraud device 1a including the recheck execution unit 19 is similar to the hardware configuration shown in Figures 9A and 9B, and is therefore not shown. The function of the recheck execution unit 19 is realized by a processing circuit 1001. The processing circuit 1001 executes the function of the recheck execution unit 19 by reading and executing a program stored in a memory 1005. That is, the anti-fraud device 1 or anti-fraud device 1a includes a memory 1005 for storing a program that, when executed by the processing circuit 1001, results in the execution of steps ST10 to ST50 of Figure 3 described above, steps ST10 to ST50 of Figure 7 described above, or steps ST10 to ST50 of Figure 11 described above. It can also be said that the program stored in the memory 1005 causes a computer to execute the processing procedure or method of the recheck execution unit 19.

[0127] It should be noted that the embodiments may be freely combined, or any of the components in each embodiment may be modified, or any of the components in each embodiment may be omitted.

[0128] It is possible to prevent fraud in breathalyzer tests administered to drivers in vehicles.

[0129] 1, 1a Anti-fraud device, 11 Image acquisition unit, 12 Detection unit, 121 First detection unit, 122 Second detection unit, 13 Provisional fraud determination unit, 14 Fraud determination result confirmation unit, 15 Test result acquisition unit, 16 Drinking determination unit, 17 Interlock unit, 18 Personal authentication unit, 19 Re-test execution unit, 100 Anti-fraud system, 2 Imaging device, 3 Output device, 1001 Processing circuit, 1002 Input interface device, 1003 Output interface device, 1004 Processor, 1005 Memory.

Claims

1. an image acquisition unit that acquires an image captured by an imaging device that is provided in a vehicle and is capable of capturing an image of at least the periphery of a driver's seat including a foot area; a detection unit that detects, based on the captured image acquired by the image acquisition unit, an occupant who is imaged in a first region around a headrest of the driver's seat in the captured image, and an occupant who is imaged in a second region below the headrest of the driver's seat in the captured image and including an area at the feet of the driver's seat; a fraud provisional determination unit that performs a fraud provisional determination process for provisionally determining the presence or absence of fraud based on the presence or absence of the occupant detected in the first area or the presence or absence of the occupant detected in the second area, and provisionally determines that there is no fraud when one occupant is detected in the first area and no occupant is detected in the second area; a fraud determination result confirmation unit that performs a fraud determination result confirmation process to confirm the result of the determination of the presence or absence of fraud, and when the fraud provisional determination unit has provisionally determined that there is no fraud and, as a result of detecting a face of a test-performing occupant who is performing an operation of performing an alcohol test based on the captured image acquired by the image acquisition unit and comparing the face of the test-performing occupant with the face of the occupant detected in the first area, determines that there is no fraud when it is presumed that the test-performing occupant and the occupant detected in the first area are the same person, The detection unit detects the occupant whose image is captured in the second area by a method different from a method for detecting the occupant whose image is captured in the first area.

13. An anti-fraud device comprising:

2. The detection unit detects that the occupant is present when the head of the occupant is detected in the second area.

2. The fraud prevention device according to claim 1.

3. The detection unit detects the face of the occupant captured in the first area and the face of the occupant captured in the second area. Detecting the face or head of the occupant being imaged 2. The fraud prevention device according to claim 1.

4. The fraud provisional determination unit performs the fraud provisional determination process, and provisionally determines that there is no fraud if one occupant is detected in the first area, no occupant is detected in the second area, and a seat belt is imaged in the second area.

2. The fraud prevention device according to claim 1.

5. The fraud provisional determination unit performs the fraud provisional determination process, and provisionally determines that there is no fraud when one occupant is detected in the first area, no occupant is detected in the second area, and the seat belt of the driver's seat is fastened.

2. The fraud prevention device according to claim 1.

6. a drinking judgment unit that judges whether the test-taking occupant is in a drunk state based on test result information indicating an alcohol test result obtained by the alcohol test taken by the test-taking occupant; an interlock unit that determines whether or not to release an interlock in the vehicle based on the determination result of the presence or absence of fraud by the fraud determination result determination unit and the determination result of the drinking alcohol determination unit as to whether or not the test occupant is in a drinking state; 2. The fraud prevention device according to claim 1, further comprising:

7. The interlock portion is release the interlock when the fraud determination result determination unit determines that there is no fraud and the drinking alcohol determination unit determines that the test-carrying occupant is not drinking alcohol; If the fraud determination result determination unit does not determine that there is no fraud, or if the drinking alcohol determination unit determines that the test-carrying occupant is in a drunk state, the interlock is continued.

7. The fraud prevention device according to claim 6.

8. a personal authentication unit that performs personal authentication of the occupant detected in the first area when the fraud provisional determination unit provisionally determines that there is no fraud, The fraud determination result confirmation unit, when the personal authentication unit has authenticated the occupant detected in the first area, and when the face of the test-performing occupant performing the operation of conducting the alcohol test is detected based on the captured image acquired by the image acquisition unit and the face of the test-performing occupant is compared with the face of the occupant detected in the first area, determines that there is no fraud when it is presumed that the test-performing occupant and the occupant detected in the first area are the same person.

2. The fraud prevention device according to claim 1.

9. a re-inspection execution unit that detects the opening and closing of a door of the vehicle, and when it detects that the door is opened, causes the detection unit to detect the occupant in the first area, the detection unit to detect the occupant in the second area, the fraud provisional judgment unit to perform the fraud provisional judgment process, and the fraud judgment result confirmation unit to perform the fraud judgment result confirmation process again.

2. The fraud prevention device according to claim 1, further comprising:

10. a re-inspection execution unit that detects a fastening state of a seat belt in the driver's seat, and when it detects that the seat belt in the driver's seat has been released, causes the detection unit to detect the occupant in the first area, the detection unit to detect the occupant in the second area, the fraud provisional judgment unit to perform the fraud provisional judgment process, and the fraud judgment result confirmation unit to perform the fraud judgment result confirmation process again.

2. The fraud prevention device according to claim 1, further comprising:

11. An image acquisition unit acquires an image captured by an imaging device provided in a vehicle so as to be able to capture an image of at least the periphery of a driver's seat including a foot area; a detection unit detecting, based on the captured image acquired by the image acquisition unit, an occupant who is imaged in a first region around a headrest of the driver's seat in the captured image and an occupant who is imaged in a second region below the headrest of the driver's seat in the captured image and including an area at the feet of the driver's seat; a fraud provisional determination unit performing a fraud provisional determination process for provisionally determining the presence or absence of fraud based on the presence or absence of the occupant detected in the first area or the presence or absence of the occupant detected in the second area, and provisionally determining that there is no fraud when one occupant is detected in the first area and no occupant is detected in the second area; a fraud determination result determination unit performing a fraud determination result determination process for determining the result of the determination of the presence or absence of fraud, and when the fraud provisional determination unit provisionally determines that there is no fraud and detects a face of a test-performing occupant performing an action to perform an alcohol test based on the captured image acquired by the image acquisition unit and compares the face of the test-performing occupant with the face of the occupant detected in the first area, when it is estimated that the test-performing occupant and the occupant detected in the first area are the same person, determining that there is no fraud, The detection unit detects the occupant whose image is captured in the second area by a method different from a method for detecting the occupant whose image is captured in the first area. A fraud prevention method comprising:

12. An anti-tamper device according to any one of claims 1 to 10; an imaging device provided in the vehicle, the imaging device imaging at least the area around the driver's seat including the foot area in the vehicle; and a fraud prevention system.