Facial recognition method and facial recognition device

Abstract

To provide a face authentication method and a face authentication apparatus, configured to perform quick and accurate face authentication by imaging the face of an occupant outside a vehicle, using an imaging apparatus mounted on the vehicle.SOLUTION: A method includes: acquiring face image data by imaging the face of an occupant outside a vehicle, using an imaging apparatus mounted on a vehicle, and acquiring information indicating the body height of the occupant (S10); controlling, when a distance between a current position of the vehicle and a boarding position of the occupant reaches a predetermined distance, the height and attitude of the vehicle, on the basis of the acquired body height of the occupant, so that an imaging direction of the imaging apparatus is pointed to the face of the occupant (S60); and acquiring, after controlling the height and attitude of the vehicle (S90), face image data by imaging the face of the occupant using the imaging apparatus, to authenticate the occupant based on the face image data.SELECTED DRAWING: Figure 4A

Description

[Technical Field]

[0001] The present invention relates to a face authentication method and a face authentication device. [Background technology]

[0002] Conventionally, there has been known a face recognition device that captures an image of a person's face with a camera and verifies whether the person is the person to be authenticated (personal authentication) by comparing the captured image with a pre-registered image (Patent Document 1). The face recognition device described in Patent Document 1 comprises a main body provided with a face capturing unit that captures an image of the face of the person to be authenticated and a face position detection unit that detects the height of the person's face, and a movement mechanism that can move the main body vertically. By moving the main body up and down vertically according to the height of the face detected by the face position detection unit, the face recognition device can move the face capturing unit to a height appropriate for authentication processing, enabling quick and accurate face recognition. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2019-133275 Summary of the Invention [Problem to be solved by the invention]

[0004] However, when authentication is performed using facial images of passengers outside the vehicle captured using an imaging device installed in the vehicle, the vehicle height adjustment mechanism generally has a slow operating speed, which delays the start of authentication.

[0005] The present invention has been made in consideration of the above-mentioned problems, and aims to provide a facial recognition method that can perform fast and accurate facial recognition in a facial recognition method that authenticates passengers using facial image data obtained by capturing images of the faces of passengers outside the vehicle using an imaging device mounted on the vehicle. [Means for solving the problem]

[0006] A facial recognition method uses an imaging device mounted on a vehicle to capture an image of a passenger's face outside the vehicle, acquire facial image data, and authenticate the passenger based on the facial image data, by acquiring information indicating the passenger's height, and when the distance from the current position of the vehicle to the passenger's boarding position reaches a predetermined distance, controlling the height or attitude of the vehicle based on the acquired passenger's height so that the imaging direction of the imaging device is directed toward the passenger's face. After controlling the height or attitude of the vehicle, the imaging device is used to capture an image of the passenger's face, acquire facial image data, and authenticate the passenger based on the facial image data. [Effects of the Invention]

[0007] According to the present invention, it is possible to perform fast and accurate face authentication using an imaging device mounted on a vehicle. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram showing an example of the configuration of a face authentication device according to the first embodiment. [Figure 2] FIG. 2 is a schematic diagram showing the mounting position of the camera and the angle of view of the camera in the face authentication device according to the first embodiment. [Figure 3A] FIG. 3A is a schematic diagram showing a state in which the face recognition device according to the first embodiment controls the height of the vehicle. [Figure 3B] FIG. 3B is a schematic diagram showing a state in which the facial recognition device according to the first embodiment controls the attitude of the vehicle (first example). [Figure 3C] FIG. 3C is a schematic diagram showing a state in which the face recognition device according to the first embodiment controls the attitude of the vehicle (second example). [Figure 4A] FIG. 4A is a flowchart (part 1) showing an example of the first process of the face recognition device according to the first embodiment. [Figure 4B] FIG. 4B is a flowchart (part 2) showing an example of the first process of the face recognition device according to the first embodiment. [Figure 5] FIG. 5 is a flowchart showing an example of the second process of the face recognition device according to the first embodiment. [Figure 6] FIG. 6 is a block diagram showing an example of the configuration of a face authentication device according to the second embodiment. [Figure 7A] FIG. 7A is a flowchart showing an example of processing performed by the face recognition device according to the second embodiment (part 1). [Figure 7B] FIG. 7B is a flowchart showing an example of the processing of the face recognition device according to the second embodiment (part 2). DETAILED DESCRIPTION OF THE INVENTION

[0009] The embodiments will be described with reference to the drawings. In the description of the drawings, the same parts are designated by the same reference numerals and the description thereof will be omitted.

[0010] (First embodiment) [Configuration of face recognition device] An example of the configuration of a face recognition device 1 according to the first embodiment will be described with reference to Fig. 1. The face recognition device 1 includes an imaging unit 10, a vehicle position acquisition unit 20, a communication unit 30, an object detection unit 40, a light irradiation unit 50, and a control unit 60.

[0011] The facial recognition device 1 is mounted on a vehicle that provides transportation services. The vehicle that provides transportation services may be a vehicle that runs autonomously using an automatic driving function, or a vehicle that runs manually by a driver. The facial recognition device 1 acquires facial image data by capturing an image of the face of a passenger outside the vehicle using an imaging unit 10 mounted on the vehicle, and authenticates whether the passenger is the person in question based on the acquired facial image data.

[0012] The imaging unit 10 is composed of multiple cameras (imaging devices: 11, 12, and 13). The multiple cameras (11, 12, and 13) constituting the imaging unit 10 have imaging elements such as CCDs (charge-coupled devices) and CMOSs ​​(complementary metal oxide semiconductors). The imaging unit 10 is mounted on a vehicle. More specifically, the imaging unit 10 is mounted on the side of the exterior of the vehicle and is composed of three cameras: camera 11, which is mounted forward of the passenger boarding / alighting doors on the side of the vehicle, camera 12, which is mounted above the passenger boarding / alighting doors, and camera 13, which is mounted rearward of the passenger boarding / alighting doors. Each camera is mounted at a different height. The imaging direction of each camera faces a different direction toward the side of the vehicle where the passenger boarding / alighting doors are installed. This allows for the acquisition of multiple image data of passengers' faces outside the vehicle from different positions, based on their heights. Each camera is fixed to the vehicle body, and the imaging direction cannot be changed by the imaging unit 10 alone. When changing the imaging direction of each camera, the vehicle height or attitude is changed by a vehicle height adjustment mechanism provided on the vehicle. The mounting position of each camera and the angle of view of each camera will be described later with reference to Fig. 2. The imaging direction and angle of view of each camera when the vehicle height or attitude is controlled by the vehicle height adjustment mechanism will be described later with reference to Figs. 3A to 3C.

[0013] The imaging unit 10 starts imaging upon receiving a signal instructing it to start imaging, and stops imaging upon receiving a signal instructing it to end imaging. Signals instructing it to start and end imaging are transmitted from the control unit 60. Each camera continuously captures images at a predetermined cycle and acquires multiple image data. The imaging unit 10 outputs the acquired image data to the control unit 60. In this embodiment, the imaging unit 10 is composed of three cameras (11, 12, and 13), but the imaging unit 10 may be composed of one or two cameras, or four or more cameras. In this embodiment, the face recognition device 1 includes the imaging unit 10, but a camera included in an ADAS (Advanced Driver-Assistance Systems) can be used. Therefore, the mounting position of the camera is not limited to the side of the vehicle, and it may be mounted on the front of the vehicle.

[0014] The vehicle position acquisition unit 20 includes a receiving device that receives GPS (Global Positioning System) signals and measures the vehicle's position (absolute position) and attitude (absolute attitude) in Earth coordinates. Furthermore, the vehicle position acquisition unit 20 includes a processing circuit that performs odometry or dead reckoning. The processing circuit measures the vehicle's relative position, yaw angle, and speed relative to a predetermined reference point based on information acquired from wheel speed sensors that detect the wheel speed of each wheel of the vehicle and a steering angle sensor that detects the steering angle of the steering wheels. The processing circuit can measure the vehicle's absolute position and attitude from the vehicle's relative position and yaw angle relative to the predetermined reference point (absolute position). The vehicle position acquisition unit 20 outputs the vehicle's absolute position and attitude to the control unit 60.

[0015] The communication unit 30 includes a receiver that receives a signal instructing a vehicle to be dispatched from the data center and a transmitter that transmits a signal indicating passenger information acquired during passenger authentication to the data center. The signal instructing a vehicle to be dispatched includes passenger information. The passenger information includes at least the passenger's boarding location (absolute location) and verification image data for passenger authentication. Furthermore, the information on passengers who have already used the transportation service further includes information indicating the passenger's height. The verification image data is image data that records a facial image of the passenger who requested a vehicle dispatch. The verification image data is used to verify whether the passenger who requested a vehicle dispatch and the passenger at the boarding location are the same person, i.e., whether the passenger is the person in question. The data center is a facility that receives a signal regarding a passenger's vehicle dispatch request transmitted from a communication terminal and transmits a signal instructing a vehicle to be dispatched to a vehicle near the passenger based on the received information. The data center is installed within a company that provides a vehicle dispatch service or transportation service. The communication terminal may be a dedicated terminal installed in a predetermined location for requesting a vehicle dispatch, or a mobile terminal owned by the person requesting a vehicle dispatch. Information indicating the passenger's boarding location can be acquired from a location acquisition unit (GPS signal) equipped in a mobile terminal such as a smartphone owned by the passenger. Alternatively, the passenger may be configured to input the boarding location when transmitting a signal requesting a vehicle dispatch from the communication terminal. The verification image data can be acquired by a camera equipped in the communication terminal, and can be acquired by configuring the information to be transmitted simultaneously with the vehicle dispatch request. Information indicating the passenger's height may be input into the mobile terminal and configured to be transmitted together with the vehicle dispatch request. The data center stores the passenger's verification image data and the passenger's height.

[0016] The object detection unit 40 is composed of multiple different types of object detection sensors, such as laser radar, millimeter-wave radar, cameras, and LiDAR (Light Detection and Ranging) sensors mounted on the vehicle, that detect objects around the vehicle. The object detection unit 40 can detect moving objects such as other vehicles, including motorcycles, and pedestrians (people), as well as stationary objects such as road markings, road signs, curbs, guardrails, parked vehicles, and people. The object detection unit 40 integrates multiple detection results obtained from the multiple object detection sensors and outputs a single detection result for each object. Specifically, the object detection unit 40 calculates the most reasonable object position, size, and behavior that minimizes error from the object position, size, and behavior obtained from each object detection sensor, taking into account the error characteristics of each object detection sensor. The object detection unit 40 uses known sensor fusion technology to comprehensively evaluate the detection results obtained from multiple types of sensors and obtain more accurate detection results. The object detection unit 40 outputs information indicating the position, size, and behavior of an object to the control unit 60. The size of an object includes at least the height of the object. In this embodiment, the object detection unit 40 includes multiple object detection sensors, but may also be configured with, for example, a single object detection sensor.

[0017] The light irradiation unit 50 is a lamp that emits light to the outside of the vehicle. The light irradiation unit 50 emits light in the same direction as the imaging direction of each camera of the imaging unit 10. In this embodiment, the light irradiation unit 50 includes three lamps corresponding to the multiple cameras (11, 12, 13). The light irradiation unit 50 may be installed in a position that allows it to emit light in the same direction as the imaging direction of each camera of the imaging unit 10. For example, the light irradiation unit 50 is mounted adjacent to each camera mounted on the side of the exterior of the vehicle. Note that the light irradiation unit 50 may be configured as a single lamp if it has an illumination range that includes the entire angle of view of each camera. For example, an LED (light-emitting diode) light can be used as the lamp. The light irradiation unit 50 turns on the lamp upon receiving a signal from the control unit 60 instructing it to turn on, and turns off the lamp upon receiving a signal instructing it to turn off. The light irradiation unit 50 can illuminate the faces of passengers outside the vehicle by emitting light.

[0018] The control unit 60 acquires information indicating the passenger's height, and when the distance from the vehicle's current position to the passenger's boarding position reaches a predetermined distance, controls the vehicle height or attitude based on the acquired passenger's height so that the imaging direction of the imaging unit 10 faces the passenger. After controlling the vehicle height or attitude, the control unit 60 captures an image of the passenger's face using the imaging unit 10 to acquire facial image data, and authenticates the passenger based on the facial image data. The control unit 60 is a general-purpose microcomputer equipped with a CPU (Central Processing Unit), memory (storage units) such as RAM and ROM, and input / output units. A computer program for functioning as the face recognition device 1 is installed in the microcomputer. By executing the computer program, the microcomputer functions as multiple information processing circuits (61-66) included in the face recognition device 1. Note that this embodiment illustrates an example in which the multiple information processing circuits (61-66) included in the face recognition device 1 are realized by software, but it is also possible to configure the information processing circuits by providing dedicated hardware for executing each information processing. Furthermore, the multiple information processing circuits may be configured as separate hardware. The control unit 60 includes a distance measurement unit 61, a passenger detection unit 62, a face position estimation unit 63, a posture control unit 64, a face detection unit 65, and an authentication determination unit 66 as a plurality of information processing circuits.

[0019] The distance measurement unit 61 measures the distance from the current position of the vehicle to the passenger boarding position. Specifically, when a travel route from the current position of the vehicle to the passenger boarding position is set by a navigation function provided in the vehicle, the distance measurement unit 61 acquires the travel route from the navigation function and measures the distance from the current position of the vehicle to the passenger boarding position on the travel route. Note that the distance measurement unit 61 may estimate a travel route from the current position of the vehicle to the passenger boarding position without acquiring the route from the navigation function provided in the vehicle, and measure the distance from the current position of the vehicle to the passenger boarding position on the estimated travel route.

[0020] When the distance from the vehicle's current location to the passenger's boarding location reaches a predetermined distance, the passenger detection unit 62 detects as a passenger a person present at the boarding location whose height is within a predetermined range from the acquired passenger height. Specifically, the passenger detection unit 62 acquires information indicating the passenger's boarding location and the passenger's height (hereinafter also referred to as "acquired height") received by the communication unit 30, and information indicating the position and size of an object detected by the object detection unit 40. The passenger detection unit 62 extracts people present within a predetermined area centered on the passenger's boarding location based on the information indicating the object's location, and determines the height of the extracted person based on the information indicating the object's size. More specifically, the passenger detection unit 62 detects the vertical length (height) of the object included in the information indicating the object's size as the person's height (hereinafter also referred to as "measured height"). The object detection unit 40 detects as a passenger a person among the extracted people whose height (measured height) is within a predetermined range from the acquired height. The predetermined area is, for example, an area within a radius of 3 m from the passenger's boarding location. The set value of the predetermined distance will be described later.

[0021] The face position estimation unit 63 estimates the position of the passenger's face based on the passenger's height. Specifically, the face position corresponding to the passenger's height is acquired from a memory (not shown) provided in the control unit 60 in which standard face positions for each height are pre-stored, and the acquired face position is estimated as the passenger's face position. The face position is the height from the ground to the center of the facial contour. The face position estimation unit 63 estimates the position of the passenger's face from the acquired height. The face position estimation unit 63 also estimates the position of the passenger's face from the measured height. The face position estimation unit 63 outputs two face positions: one estimated from the acquired height and the other estimated from the measured height.

[0022] When the distance from the current position of the vehicle to the passenger's boarding position reaches a predetermined distance, the attitude control unit 64 controls the height or attitude of the vehicle based on the passenger's height so that the imaging direction of the imaging unit 10 faces the passenger's face. Specifically, the attitude control unit 64 controls the vehicle's height or attitude by controlling a height adjustment mechanism provided in the vehicle. The attitude control unit 64 calculates the vehicle height or attitude at which the imaging direction of the imaging unit 10 faces the passenger's face based on the face position estimated from the acquired height and the face position estimated from the measured height, and transmits the calculated vehicle height or attitude to a height adjustment controller provided in the height adjustment mechanism. This allows the attitude control unit 64 to orient the imaging direction of the imaging unit 10 toward the passenger's face, and to capture the passenger's face at the center of the camera's angle of view until the vehicle stops at the boarding position. The vehicle height is the vertical distance from the ground where the vehicle's tires are in contact with the ground to the highest point of the vehicle's roof, and the vehicle attitude is the vehicle's roll angle and pitch angle. The attitude control unit 64 can also control the vehicle height and attitude in combination. The attitude control unit 64 controls the vehicle height or attitude by executing the first control or the second control described below depending on the situation. The third control is an example of the case where the first control and the second control are executed in combination.

[0023] (First control) When the distance from the current position of the vehicle to the passenger's boarding position reaches a predetermined distance, the attitude control unit 64 controls the height or attitude of the vehicle based on the acquired passenger's height (acquired height) so that the imaging direction of the imaging unit 10 is directed toward the passenger's face. Specifically, the attitude control unit 64 calculates the vehicle height or attitude such that the center position of the vertical angle of view (hereinafter referred to as the "vertical angle of view") of any of the multiple cameras (11, 12, 13) is the position of the passenger's face estimated from the acquired height, and transmits the calculated vehicle height or attitude to the vehicle height adjustment controller. In other words, directing the imaging direction of the camera toward the passenger's face means aligning the center position of the vertical angle of view of the camera with the passenger's face. (Second control) The attitude control unit 64 controls the height or attitude of the vehicle based on the detected passenger height (measured height) so that the imaging direction of the imaging unit 10 faces the passenger's face. Specifically, the attitude control unit 64 calculates the vehicle height or attitude such that the center position of the vertical angle of view of any of the multiple cameras (11, 12, 13) is the position of the passenger's face estimated from the measured height, and transmits the calculated vehicle height or attitude to the vehicle height adjustment controller. If the attitude control unit 64 cannot detect a passenger at the boarding position, it cannot acquire the measured height. If the attitude control unit 64 cannot detect a passenger at the boarding position, it executes only the first control. Note that the attitude control unit 64 may execute only the second control without executing the first control. (Third control) When the distance from the boarding position of the vehicle to the boarding position of the passenger reaches a predetermined distance, the attitude control unit 64 starts controlling the height or attitude of the vehicle based on the acquired height of the passenger (acquired height) so that the imaging direction of the imaging unit 10 is directed toward the face of the passenger. Then, if the attitude control unit 64 detects a passenger at the boarding position after starting to control the height or attitude of the vehicle, it controls the height or attitude of the vehicle based on the detected height of the passenger (measured height) so that the imaging direction of the imaging unit 10 is directed toward the face of the passenger. Therefore, when the attitude control unit 64 detects a passenger at the boarding position, it switches from the first control to the second control. Note that a method for selecting a camera from the multiple cameras (11, 12, 13) whose center position of the vertical angle of view is aligned with the face position estimated from the acquired height or measured height will be described in detail below with reference to FIG. 2.

[0024] The attitude control unit 64 controls the height or attitude of the vehicle so that the imaging direction of the imaging unit 10 faces the passengers by the time the vehicle stops at the passenger boarding position. The timing at which the control of the vehicle's height or attitude ends can be adjusted by setting a predetermined distance. Specifically, the predetermined distance is a distance that allows the vehicle to travel for a time longer than the time required to change the vehicle's height from the standard position to the highest position or the time required to change the vehicle's height from the standard position to the lowest position. This allows the height or attitude of the vehicle to be controlled so that the imaging direction of the imaging unit 10 faces the passengers by the time the vehicle stops at the passenger boarding position. Therefore, by increasing the predetermined distance, the imaging direction of the imaging unit 10 can be directed toward the passengers' faces at an earlier timing. Note that the height when the vehicle is traveling is the standard position. Furthermore, after starting control of the vehicle's height or attitude, if the attitude control unit 64 determines that the vehicle's behavior is equal to or greater than a predetermined value, the attitude control unit 64 suspends control of the vehicle's height and attitude, and if it determines that the vehicle's behavior is less than the predetermined value, the attitude control unit 64 resumes control of the vehicle's height or attitude. The predetermined values ​​are, for example, allowable limits of acceleration in the roll direction, pitch direction, and vertical direction that do not cause discomfort to passengers inside the vehicle. When the posture control unit 64 determines that the center position of the vertical angle of view of any camera is the position of the passenger's face estimated from the acquired height or the position of the passenger's face estimated from the measured height, it outputs a signal to the face detection unit 65 notifying that acquisition of passenger's facial image data is possible. The posture control unit 64 not only outputs the signal notifying that acquisition of passenger's facial image data is possible to the face detection unit 65, but also acquires information indicating the brightness outside the vehicle from a light sensor provided in the vehicle, and, when it determines that the brightness outside the vehicle is less than a predetermined brightness, outputs a signal to the light irradiation unit 50 instructing the light to turn on the illumination lamps. The predetermined brightness is, for example, the brightness at which a light controller that controls the on / off of the vehicle's sidelights and headlights turns on the sidelights.

[0025] After the attitude control unit 64 controls the vehicle height or attitude, the face detection unit 65 captures the passenger's face using the imaging unit 10 to acquire the passenger's facial image data. Specifically, when the face detection unit 65 receives a signal notifying that the passenger's facial image data can be acquired, i.e., when the center position of the vertical angle of view of any camera coincides with the position of the passenger's face estimated from the acquired height or the position of the face estimated from the measured height, the face detection unit 65 outputs a signal to the imaging unit 10 to instruct the imaging unit 10 to start capturing the image. The face detection unit 65 receives image data acquired by the multiple cameras (11, 12, 13). The face detection unit 65 detects the face of a person from the received image data and extracts the image in which the face is detected as the passenger's facial image data. The face detection unit 65 can detect the face of a person captured in the image data by using pattern matching, a known image analysis technique.

[0026] The authentication determination unit 66 authenticates passengers based on the facial image data of the passengers acquired by the multiple cameras (11, 12, 13). Specifically, the authentication determination unit 66 compares the facial image data of the passengers with the collation image data to determine whether the person appearing in the facial image data of the passengers and the person appearing in the collation image data are the same person. If the authentication determination unit 66 determines that the person appearing in the facial image data of the passengers and the person appearing in the collation image data are the same person, it authenticates the passenger as the passenger. The authentication determination unit 66 extracts facial features such as the eyes, nose, mouth, and facial contours from the faces of the passengers and the person appearing in the collation image data, and authenticates the passenger as the passenger if it determines that the match rate of each extracted feature is equal to or greater than a predetermined match rate. Note that the method of comparing two facial images to determine whether they are the same person is not limited to the above method. The authentication determination unit 66 can use known face recognition technology to determine whether the person appearing in the facial image data and the person appearing in the collation image data are the same person. The authentication determination unit 66 may authenticate the passenger as the passenger when it determines that the person captured in the passenger face image data acquired by at least two of the multiple cameras (11, 12, 13) is the same person as the person captured in the verification image data. If the authentication determination unit 66 authenticates the passenger as the passenger, it outputs a signal to an interface provided in the vehicle or a door controller that controls the boarding / alighting doors to permit the boarding / alighting doors to be opened. This allows the passenger outside the vehicle to be confirmed as the person who requested the vehicle dispatch, and then the passenger can be allowed to board.

[0027] If the imaging direction of the imaging unit 10 is controlled to face the passenger's face before the vehicle stops at the passenger's boarding position, the authentication determination unit 66 ends the process related to passenger authentication before the vehicle stops. In this case, if the authentication determination unit 66 determines that the passenger's facial image data matches the verification image data, it outputs a signal to instruct the vehicle to stop. Furthermore, if the authentication determination unit 66 determines that the passenger's facial image data does not match the verification image data, it outputs a signal to prohibit the vehicle from stopping.

[0028] Next, the camera mounting positions and camera angles of view will be described with reference to FIG. 2. Multiple cameras (11, 12, 13) are mounted in different positions on vehicle 100, and the imaging direction of each camera faces a different direction. Specifically, camera 11 is mounted at a height of 170 cm from the ground and in front of boarding / alighting door D. Camera 12 is mounted at a height of 180 cm from the ground above boarding / alighting door D. Camera 13 is mounted at a height of 150 cm from the ground and behind boarding / alighting door D. All of the multiple cameras (11, 12, 13) are identical cameras, and the angle of view Av1 of camera 11, the angle of view Av2 of camera 12, and the angle of view Av3 of camera 13 are the same. The imaging direction of camera 11 is directed horizontally in the vehicle width direction, the imaging direction of camera 12 is tilted downward from the horizontal direction, and the imaging direction of camera 13 is tilted forward and upward from the vehicle width direction. The imaging direction of camera 13 is tilted toward the front of the vehicle because vehicle 100 stops so that passengers face directly opposite boarding / alighting door D. Therefore, by mounting camera 13 so that its imaging direction is tilted toward the front of the vehicle, passengers' faces can be included within the angle of view of camera 13 even when the vehicle stops so that passengers face directly opposite boarding / alighting door D. As described above, in this embodiment, multiple cameras (11, 12, 13) are mounted at different positions, and the imaging directions of each camera are oriented in different directions. This makes it possible to set the center position of the vertical angle of view of any camera to the position of the passenger's face from the state before controlling the vehicle height or attitude. The attitude control unit 64 selects the camera whose center position of the vertical angle of view is closest to the position of the passenger's face estimated from the acquired height or the position of the face estimated from the measured height as the camera whose center position of the vertical angle of view is closest to the position of the passenger's face. For example, if the face position is 155 cm, camera 13 mounted at a height of 150 cm from the ground is selected as the camera whose vertical center position of the angle of view is aligned with the face position. This minimizes the amount of control of the vehicle height adjustment mechanism, and enables the vehicle height or attitude to be controlled to the target vehicle height or attitude in the shortest time. Note that the camera mounting position, angle of view, and imaging direction are not limited to those described above and can be set appropriately according to the vehicle dimensions and the expected height of passengers.

[0029] Next, with reference to FIGS. 3A to 3C, the imaging direction and angle of view of each camera when the vehicle height or attitude is controlled by the height adjustment mechanism will be described. There are several types of height adjustment mechanisms, but the height adjustment mechanism installed on vehicle 100 is an air suspension that can expand and contract by itself using air pressure. Note that air suspension is a well-known height adjustment mechanism, and detailed description will be omitted. The height adjustment mechanism can control all suspensions in the same way, or can control each suspension individually. Note that the camera angles of view (Av1, Av1h, Av1u, Av1d) shown in FIGS. 3A to 3C all indicate the angle of view of camera 11. The following description will be given assuming that the center position of the vertical angle of view of camera 11 is aligned with the position of the face. Note that the height adjustment mechanism is not limited to a configuration that expands or contracts the length of the suspension, and may be configured to control the vehicle height or attitude.

[0030] FIG. 3A shows a state in which the vehicle height is controlled. Specifically, it shows a state in which the vehicle height adjustment mechanism extends all of the suspensions to raise the vehicle height. By raising the vehicle height, the angle of view Av1h of camera 11 and the center position c1h of the vertical angle of view of the angle of view Av1h move in the upward direction U relative to the angle of view Av1 of camera 11 and the center position c1 of the vertical angle of view of the angle of view Av1 in the standard state. This allows the center position of the vertical angle of view of camera 11 to be aligned with the position of the passenger's face, which is higher than the center position c1 of the vertical angle of view of camera 11 in the standard state.

[0031] FIG. 3B shows a state in which the vehicle posture is changed (first example). Specifically, the vehicle height adjustment mechanism compresses the suspension on the right side R of the vehicle, thereby relatively raising the left side L of the vehicle. By relatively raising the left side L of the vehicle, the angle of view Av1u of the camera 11 and the center position c1u of the vertical angle of view of the angle of view Av1u are tilted toward the upper left of the vehicle with respect to the angle of view Av1 of the camera 11 in the standard state and the center position c1 of the vertical angle of view of the angle of view Av1. This allows the center position of the vertical angle of view of the camera 11 to be aligned with the position of the passenger's face, which is higher than the center position c1 of the vertical angle of view in the standard state. Note that while FIG. 3B shows an example in which the suspension on the right side R of the vehicle is compressed, it is also possible to control the suspension on the left side L of the vehicle to be extended and the suspension on the right side R of the vehicle to be compressed. This allows the imaging direction of the camera 11 to be directed toward the position of the passenger's face, which is higher.

[0032] FIG. 3C shows a state in which the vehicle posture is changed (second example). Specifically, the vehicle height adjustment mechanism compresses the suspension on the left side L of the vehicle, thereby relatively raising the right side R of the vehicle. By relatively raising the right side R of the vehicle, the angle of view Av1d of the camera 11 and the center position c1d of the vertical angle of view of the angle of view Av1d are tilted toward the lower left of the vehicle with respect to the angle of view Av1 of the camera 11 in the standard state and the center position c1 of the vertical angle of view of the angle of view Av1. This allows the center position of the vertical angle of view of the camera 11 to be aligned with the position of the passenger's face, which is lower than the center position c1 of the vertical angle of view of the camera 11 in the standard state. Note that while FIG. 3C shows an example in which the suspension on the left side L of the vehicle is compressed, it is also possible to control the suspension on the right side R of the vehicle to be extended and the suspension on the left side L of the vehicle to be compressed. This allows the imaging direction of the camera 11 to be directed toward the position of the passenger's face, which is lower. As described above, by controlling the height or attitude of the vehicle to align the center position of the vertical angle of view of one of the cameras with the position of the passenger's face, facial image data of the passenger's face captured at the center position of the angle of view of one of the cameras can be obtained before the vehicle stops.

[0033] [Facial recognition method] (First process) Next, an example of a first process of the face recognition device shown in Fig. 1 will be described with reference to Fig. 4A and Fig. 4B. The operation of the face recognition device 1 shown in the flowchart of Fig. 4 starts at the same time as receiving a dispatch instruction from the data center, and ends when it is determined whether or not the boarding door can be opened.

[0034] In step S10, the control unit 60 acquires the passenger's riding position, height, and verification image data received by the communication unit 30. The process proceeds to step S20, where the control unit 60 acquires the vehicle's position measured by the vehicle position acquisition unit 20. The process proceeds to step S30, where the passenger detection unit 62 determines whether the distance from the vehicle's current position to the passenger's riding position has reached a predetermined distance. Specifically, in step S30, the distance measurement unit 61 acquires a route to the passenger's riding position from the vehicle's navigation function and calculates the distance from the vehicle's current position to the passenger's riding position on the acquired route. Note that the distance measurement unit 61 may also estimate a route from the vehicle's current position to the passenger's riding position and calculate the distance from the vehicle's current position to the passenger's riding position on the estimated route. If the passenger detection unit 62 determines that the distance from the vehicle's current position to the passenger's riding position has reached the predetermined distance (YES in step S30), the process proceeds to step S40. If the passenger detection unit 62 determines that the distance from the current position of the vehicle to the passenger boarding position has not reached the predetermined distance (NO in step S30), the process returns to step S20.

[0035] In step S40, if the passenger detection unit 62 detects a passenger at the boarding position (YES in step S40), the process proceeds to step S50. In step S40, if the passenger detection unit 62 fails to detect a passenger at the boarding position (NO in step S40), the process proceeds to step S70. Specifically, in step S40, the passenger detection unit 62 acquires information indicating the passenger's boarding position and the passenger's height received by the communication unit 30, and information indicating the position and size of the object detected by the object detection unit 40. Based on the information indicating the object's position, the passenger detection unit 62 extracts a person present within a predetermined area centered on the passenger's boarding position, and determines the height of the extracted person based on the information indicating the object's size. More specifically, the passenger detection unit 62 detects the vertical length (height) of the object included in the information indicating the object's size as the person's height (measured height). Of the extracted people, the passenger detection unit 62 detects as passengers any person whose height (measured height) is within a predetermined range from the acquired height. The predetermined area is, for example, an area within a radius of 3 m from the passenger's boarding position. In step S50, the passenger detection unit 62 acquires the height of the detected passenger, and the process proceeds to step S60.

[0036] In step S60, the attitude control unit 64 calculates the vehicle height or attitude at which the imaging direction of the imaging unit 10 is directed toward the passenger's face based on the measured height. Specifically, the attitude control unit 64 calculates the vehicle height or attitude at which the center position of the vertical angle of view of any one of the multiple cameras (11, 12, 13) is the position of the passenger's face estimated from the measured height. In step S70, the attitude control unit 64 calculates the vehicle height or attitude at which the imaging direction of the imaging unit 10 is directed toward the passenger's face based on the acquired height. Specifically, the attitude control unit 64 calculates the vehicle height or attitude at which the center position of the vertical angle of view of any one of the multiple cameras (11, 12, 13) is the position of the passenger's face estimated from the acquired height. The process proceeds to step S80, and if the attitude control unit 64 determines that the vehicle's minimum ground clearance at the calculated vehicle height or attitude is equal to or greater than a predetermined ground clearance (YES in step S80), the process proceeds to step S90. If the attitude control unit 64 determines that the vehicle's minimum ground clearance at the calculated vehicle height or attitude is less than a predetermined ground clearance (NO in step S80), the process ends. In step S90, the attitude control unit 64 transmits the vehicle height or attitude calculated in step S60 or step S70 to the vehicle height adjustment controller, and controls the vehicle height or attitude. The process proceeds to step S100, where the attitude control unit 64 acquires information indicating the brightness outside the vehicle from a light sensor provided in the vehicle. If the attitude control unit 64 determines that the brightness outside the vehicle is less than a predetermined brightness (YES in step S100), the process proceeds to step S110. If the attitude control unit 64 determines in step S100 that the brightness outside the vehicle is equal to or greater than the predetermined brightness (NO in step S100), the process proceeds to step S120. In step S110, the attitude control unit 64 outputs a signal to the light irradiation unit 50 to instruct the light irradiation unit 50 to turn on the illumination lamp.

[0037] In step S120, if the attitude control unit 64 determines that the vehicle behavior is equal to or greater than a predetermined value (predetermined behavior) (YES in step S120), the process proceeds to step S130. If the attitude control unit 64 determines that the vehicle behavior is less than the predetermined value (NO in step S120), the process proceeds to step S140. In step S130, the attitude control unit 64 suspends control of the vehicle height and attitude. The process proceeds to step S170, and if the attitude control unit 64 determines that the vehicle behavior has become less than the predetermined value (YES in step S170), the process returns to step S90. In step S170, if the attitude control unit 64 determines that the vehicle behavior is not less than the predetermined value (NO in step S170), the process returns to step S130. The predetermined values ​​are allowable limit values ​​of acceleration in the roll direction, pitch direction, and vertical direction that do not cause discomfort to passengers inside the vehicle. In step S140, the face detection unit 65 captures an image of the passenger's face to obtain facial image data of the passenger, and the process proceeds to step S150.

[0038] In step S150, if the authentication determination unit 66 determines that the passenger's facial image data matches the verification image data (YES in step S150), the process proceeds to step S160. If the authentication determination unit 66 determines that the passenger's facial image data does not match the verification image data (NO in step S150), the process ends. Specifically, if the authentication determination unit 66 determines that the person captured in the passenger's facial image data acquired by at least two of the multiple cameras (11, 12, 13) is the same person as the person captured in the verification image data, the authentication determination unit 66 authenticates the passenger as the person. The authentication determination unit 66 extracts facial features such as the eyes, nose, mouth, and facial contour from the passenger's facial image data and the face of the person captured in the verification image data, and determines whether the match rate for each of the extracted features is equal to or greater than a predetermined match rate. The process proceeds to step S160, where the authentication determination unit 66 outputs a signal to an interface provided in the vehicle or a door controller controlling the boarding / alighting door D to permit the boarding / alighting door D to be opened.

[0039] If the passenger authentication (step S150) is set to end before the vehicle stops at the boarding position, and if it is determined that the passenger's facial image data matches the collation image data (YES in step S150), a signal instructing the vehicle to stop is output. Also, if the passenger authentication (step S150) is set to end before the vehicle stops at the boarding position, and if it is determined that the passenger's facial image data does not match the collation image data (NO in step S150), a signal prohibiting the vehicle from stopping is output.

[0040] (Second process) Next, an example of the second processing of the face recognition device 1 shown in Fig. 1 will be described with reference to Fig. 5. The operation of the face recognition device 1 shown in the flowchart of Fig. 5 starts at the same time as receiving a dispatch instruction from the data center, and ends when the acquired passenger height is transmitted to the data center. Note that the processing of the face recognition device 1 shown in Fig. 5 is executed when the passenger information does not include information indicating the passenger's height.

[0041] In step S200, the distance measurement unit 61 determines whether the vehicle is stopped at the passenger boarding position based on the distance from the vehicle's current position to the passenger boarding position. If the distance measurement unit 61 determines that the vehicle is stopped at the passenger boarding position (YES in step S200), the process proceeds to step S210. If the distance measurement unit 61 determines that the vehicle is not stopped at the passenger boarding position (NO in step S200), the process returns to step S200. In step S210, the passenger detection unit 62 detects, as passengers, people who are present within a predetermined area centered on the passenger boarding position based on the information indicating the object position. The face detection unit 65 captures the passenger's face using the imaging unit 10 to acquire facial image data of the passenger. Note that the acquired facial image data of the passenger is acquired without controlling the height or attitude of the vehicle. The predetermined area is, for example, an area within a radius of 3 m from the passenger boarding position.

[0042] The process proceeds to step S220. If the authentication determination unit 66 determines that the passenger's facial image data and the collation image data match (YES in step S220), the process proceeds to step S230. If the authentication determination unit 66 determines that the passenger's facial image data and the collation image data do not match (NO in step S220), the process ends. Specifically, if the authentication determination unit 66 determines that the person captured in the passenger's facial image data acquired by at least two of the multiple cameras (11, 12, 13) is the same person as the person captured in the collation image data, the passenger is authenticated. The process proceeds to step S230, where the passenger detection unit 62 acquires the passenger's height. Specifically, the passenger detection unit 62 acquires the height of the person corresponding to the passenger from information indicating the size of the object. The process proceeds to step S240, where the authentication determination unit 66 outputs a signal to an interface provided in the vehicle or a door controller controlling the boarding / alighting door to permit the boarding / alighting door to be opened. The process proceeds to step S250, where the passenger detection unit 62 transmits the passenger's height to the data center via the communication unit 30.

[0043] [Action and effect] As described above, the present embodiment provides the following advantageous effects.

[0044] The facial recognition device 1 acquires information indicating the passenger's height, and when the distance from the current vehicle position to the passenger's boarding position reaches a predetermined distance, the facial recognition device 1 controls the vehicle's height or attitude based on the acquired passenger's height so that the imaging direction of the imaging device 10 faces the passenger's face. After controlling the vehicle's height or attitude, the facial recognition device 1 captures an image of the passenger's face using the imaging device 10 to acquire facial image data, and authenticates the passenger based on the facial image data. This allows the facial recognition device 1 to pre-orient the imaging direction of the imaging device 10 toward the passenger's face and acquire facial image data in which the passenger's face is captured in the center of the imaging device 10's angle of view. Therefore, the facial recognition device 1 can perform fast and accurate facial recognition. When the distance from the current vehicle position to the passenger's boarding position reaches a predetermined distance, the facial recognition device 1 detects as a passenger a person present at the boarding position who has a height within a predetermined range of the acquired passenger's height, and controls the vehicle's height or attitude based on the detected passenger's height so that the imaging direction faces the passenger's face. This allows the facial recognition device 1 to detect passengers at the boarding position based on the acquired passenger heights. Therefore, the facial recognition device 1 can more accurately orient the imaging direction of the imaging device 10 toward the passenger's face, enabling more accurate facial recognition.

[0045] When the distance from the vehicle's current position to the passenger's boarding position reaches a predetermined distance, the facial recognition device 1 starts controlling the vehicle's height or attitude based on the acquired passenger's height so that the imaging direction is directed toward the passenger's face. If the facial recognition device 1 detects a passenger after starting to control the vehicle's height or attitude, it controls the vehicle's height or attitude based on the detected passenger's height so that the imaging direction is directed toward the passenger's face. This allows the facial recognition device 1 to control the vehicle's height and attitude at an early stage based on the acquired passenger's height, thereby enabling rapid facial recognition. Furthermore, when a passenger is detected, the facial recognition device 1 controls the vehicle's height or attitude based on the detected passenger's actual height so that the imaging direction is directed toward the passenger's face, thereby enabling more accurate facial recognition. The facial recognition device 1 controls the vehicle's height or attitude so that the imaging direction is directed toward the passenger's face by the time the vehicle stops at the boarding position. This allows the facial recognition device 1 to start facial recognition at the latest when the vehicle stops, enabling more rapid facial recognition compared to controlling the vehicle's height or attitude after stopping at the boarding position.

[0046] If the facial recognition device 1 determines that the vehicle's behavior is equal to or greater than the predetermined behavior, it suspends control of the vehicle's height and attitude. If the facial recognition device 1 determines that the vehicle's behavior is less than the predetermined behavior, it resumes control of the vehicle's height or attitude. This allows the facial recognition device 1 to quickly perform facial recognition while preventing uncomfortable shaking for passengers already in the vehicle. The facial recognition device 1 controls the vehicle's height or attitude by controlling the vehicle's height adjustment mechanism. This allows the facial recognition device 1 to orient the imaging direction of the imaging device 10 toward the passenger's face. The facial recognition device 1 completes passenger authentication before the vehicle stops at the boarding position. This allows the facial recognition device 1 to quickly perform facial recognition and allow passengers to board the vehicle as soon as the vehicle stops. If the facial recognition device 1 determines that the person captured in the facial image data and the person captured in the verification image data are the same person, it outputs a signal to instruct the vehicle to stop. This allows the facial recognition device 1 to stop the vehicle only if it determines that the passenger is the intended passenger, thereby avoiding unnecessary stops.

[0047] The facial recognition device 1 includes at least two imaging devices 10, one for the front and one for the rear of the vehicle. If the facial recognition device 1 determines that a person captured in facial image data acquired by each of the at least two imaging devices is the same person as a person captured in the verification image data, it outputs a signal to instruct the vehicle to stop. This allows the facial recognition device 1 to perform more accurate and rapid facial recognition and to stop the vehicle only if it determines that the passenger is the person in question. This prevents unnecessary stops. If the facial recognition device 1 is unable to acquire information indicating the passenger's height, it detects a person present at the boarding position as a passenger, captures an image of the passenger's face using the imaging device 10 to acquire facial image data, and acquires verification image data in which the passenger's facial image is recorded. If the facial recognition device 1 determines that the person captured in the facial image data is the same person as the person captured in the verification image data, it acquires the passenger's height. This allows the facial recognition device 1 to acquire the passenger's height, enabling rapid and accurate facial recognition the next time passengers are authenticated. The facial recognition device 1 includes a light irradiation unit that emits light to the outside of the vehicle. When imaging begins, the light is emitted in the imaging direction. This allows the face recognition device 1 to acquire face image data with high accuracy even at night, and to perform accurate face recognition.

[0048] (Second embodiment) [Configuration of face recognition device] With reference to FIG. 6, an example of the configuration of a face authentication device 1 according to the second embodiment will be described. The face authentication device 1 according to the second embodiment differs from the first embodiment in that it further includes a center of gravity estimation unit 67, but the other configurations are the same. Therefore, only the differences will be described, and descriptions of other common parts will be omitted. The center of gravity estimation unit 67 acquires collation image data in which a passenger's face image is recorded, and calculates a first deviation between the center of gravity of the passenger's facial contour and the center of gravity of both eyes from the collation image data. The center of gravity estimation unit 67 also calculates a second deviation between the center of gravity of the passenger's facial contour and the center of gravity of both eyes from at least one piece of face image data among the acquired face image data. The center of gravity estimation unit 67 extracts face image data in which the difference between the first deviation and the second deviation is less than a predetermined difference, and outputs the extracted face image data to the authentication determination unit 66.

[0049] [Facial recognition method] Next, an example of the processing of the face recognition device 1 shown in FIG. 6 will be described with reference to FIGS. 7A and 7B. The processing of the face recognition device 1 according to the second embodiment differs from the processing of the face recognition device 1 shown in FIG. 1 in that it further includes processing of steps S310 to S330, but the other processing is the same. Therefore, only the differences will be described, and a description of the other common processing will be omitted. In step S310, the center of gravity estimation unit 67 calculates a first deviation between the center of gravity of the passenger's facial contour and the center of gravity of both eyes from the collation image data. The center of gravity of both eyes is, for example, the midpoint of the line segment connecting the center positions of the contours of the left and right eyes. The deviation between the center of gravity of the face contour and the center of gravity of both eyes is the distance between the center of gravity of the face contour and the center of gravity of both eyes. The processing proceeds to step S320, where the center of gravity estimation unit 67 calculates a second deviation between the center of gravity of the face contour and the center of gravity of both eyes from at least one piece of face image data. The process proceeds to step S330, where the center of gravity estimation unit 67 extracts face image data for which the difference between the first deviation and the second deviation is less than a predetermined difference, and outputs the extracted face image data to the authentication determination unit 66.

[0050] [Action and effect] As described above, this embodiment achieves the following advantages in addition to the advantages of the first embodiment. The facial recognition device 1 acquires collation image data in which a facial image of a passenger is recorded, calculates a first deviation between the center of gravity of the passenger's facial contour and the center of gravity of both eyes from the collation image data, and calculates a second deviation between the center of gravity of the passenger's facial contour and the center of gravity of both eyes from at least one facial image data. The facial recognition device 1 extracts facial image data in which the difference between the first deviation and the second deviation is less than a predetermined difference, and authenticates the passenger based on the extracted facial image data and the collation image data. As a result, if a person shown in the facial image data is not a passenger, the facial recognition device 1 can eliminate the facial image data before executing passenger authentication processing, thereby reducing the processing load required for authentication. This enables faster and more accurate facial recognition. [Explanation of symbols]

[0051] 1. Facial recognition device 10 Imaging device (imaging unit) 20 Vehicle position acquisition unit 30 Communications Department 50 Light irradiation unit 60 Control Unit

Claims

1. A facial recognition method in which a computer captures an image of a face of a passenger outside a vehicle using an imaging device mounted on the vehicle to obtain facial image data, and authenticates the passenger based on the facial image data, wherein the computer: acquiring information indicative of the passenger's height; when the distance from the current position of the vehicle to the boarding position of the passenger reaches a predetermined distance, controlling the height or attitude of the vehicle based on the acquired height of the passenger so that the imaging direction of the imaging device is directed toward the face of the passenger; After controlling the height or the attitude, the imaging device is used to capture an image of the passenger's face to obtain the face image data. authenticating the passenger based on the facial image data; Face recognition method.

2. The computer When the distance reaches the predetermined distance, a person who is present at the boarding position and has a height within a predetermined range from the acquired height of the passenger is detected as the passenger; Based on the detected height of the passenger, the height or attitude of the vehicle is controlled so that the imaging direction is directed toward the face of the passenger. The face authentication method according to claim 1 .

3. The computer When the distance reaches the predetermined distance, control of the height or posture is started based on the acquired height of the passenger so that the imaging direction is directed toward a face of the passenger; If the passenger is detected after the control is started, the height or posture is controlled based on the detected height of the passenger so that the imaging direction is directed toward the face of the passenger. The face authentication method according to claim 2 .

4. The computer By the time the vehicle stops at the boarding position, the height or attitude of the vehicle is controlled so that the imaging direction is directed toward the face of the passenger. The face authentication method according to claim 1 .

5. The computer When it is determined that the behavior of the vehicle is equal to or greater than a predetermined behavior, the height and attitude control is suspended, and when it is determined that the behavior is less than the predetermined behavior, the control is resumed. The face authentication method according to claim 1 .

6. The computer The vehicle height or attitude is controlled by controlling a vehicle height adjustment mechanism provided in the vehicle. The face authentication method according to claim 1 .

7. The computer acquiring collation image data in which a facial image of the passenger is recorded, and calculating a first deviation between a centroid of a facial contour of the passenger and a centroid of both eyes from the collation image data; calculating a second deviation between the center of gravity of the facial contour and the center of gravity of the eyes from at least one of the face image data; extracting face image data in which the difference between the first deviation and the second deviation is less than a predetermined difference; The passenger is authenticated based on the extracted face image data and the collation image data. The face authentication method according to claim 1 .

8. The computer The passenger authentication is completed before the vehicle stops at the boarding location. The face authentication method according to claim 7 .

9. The computer If it is determined through the authentication that the person appearing in the face image data and the person appearing in the verification image data are the same person, a signal to instruct the vehicle to stop is output. The face authentication method according to claim 8.

10. At least two of the imaging devices are provided in the front-rear direction of the vehicle, The computer When it is determined that the person appearing in the facial image data acquired by at least two of the imaging devices is the same person as the person appearing in the collation image data, a signal to instruct the vehicle to stop is output. The face authentication method according to claim 9 .

11. The computer If the information indicating the height of the passenger cannot be acquired, a person present at the boarding position is detected as the passenger; Using the imaging device, an image of the passenger's face is captured to obtain facial image data; Acquire verification image data in which a facial image of the passenger is recorded; If it is determined that the person in the face image data and the person in the verification image data are the same person, the height of the passenger is acquired. The face authentication method according to claim 1 or 7.

12. a light emitting unit that emits light to the outside of the vehicle; The computer When starting imaging, light is emitted in the imaging direction. The face authentication method according to claim 1 .

13. A facial recognition device that captures an image of a face of a passenger outside the vehicle using an imaging device mounted on the vehicle to obtain facial image data, and authenticates the passenger based on the facial image data, a vehicle position acquisition unit that acquires a current position of the vehicle; a communication unit that acquires information indicating the passenger's riding position and the passenger's height; a control unit, The control unit when the distance from the current position of the vehicle to the boarding position of the passenger reaches a predetermined distance, controlling the height or attitude of the vehicle based on the acquired height of the passenger so that the imaging direction of the imaging device is directed toward the face of the passenger; After controlling the height or the attitude, the imaging device is used to capture an image of the passenger's face to obtain the face image data. authenticating the passenger based on the facial image data; Facial recognition device.

Images