Processing device, processing method, and recording medium

Abstract

A processing device according to the present disclosure comprises a detection unit, a calculation unit, an output control unit, and an authentication unit. The detection unit tracks a person detected from an image. The calculation unit calculates a first point in time at which the size of the face of the person in the image reaches a predetermined size suitable for biometric authentication on the basis of the manner of change in the size of the face of the person in the image over time. The output control unit causes an output device to output first information for drawing attention of the person at a second point in time that is earlier than the first point in time by a predetermined period of time. The authentication unit performs biometric authentication on the basis of an image generated after the second point in time.

Description

Processing device, processing method, and recording medium

[0001] The present disclosure relates to a processing device, a processing method, and a program.

[0002] A technology related to this disclosure is disclosed in Patent Document 1. Patent Document 1 discloses an authentication system in which an authentication user does not need to stop when being authenticated.

[0003] When the authentication system detects a person in an image, it tracks that person. If the number of pixels in the current facial region is greater than the number of pixels in the facial region a predetermined time ago, the authentication system determines that the person is approaching the door and performs processing such as extracting facial features. If facial feature information valid for authentication is extracted, the authentication system retains the facial feature information for authentication. On the other hand, if the number of pixels in the current facial region is less than the number of pixels in the facial region a predetermined time ago, the authentication system determines that the person has no intention of entering the room and stops tracking.

[0004] Furthermore, when the authentication system detects with a sensor that a person is in the authentication area, it performs authentication processing using the stored facial feature information. Then, even when a person enters the authentication area, if the facial feature information is not yet stored, the authentication system transmits information to guide the person's line of sight.

[0005] Japanese Patent Application Laid-Open No. 2005-275869

[0006] In biometric authentication using image analysis (face recognition, iris recognition, etc.), there is a demand for walk-through biometric authentication technology that does not require people to stand still. Walk-through biometric authentication technology requires technology that can capture images of a moving person's face in a manner suitable for biometric authentication.

[0007] An example of a purpose of this disclosure is to provide a technology for capturing an image of a moving person's face in a manner suitable for biometric authentication.

[0008] According to this disclosure, a processing device is provided that includes: a detection means for tracking a person detected from an image; a calculation means for calculating a first timing at which the size of the person's face in the image will reach a predetermined size suitable for biometric authentication based on the manner in which the size of the person's face in the image changes over time; an output control means for causing an output device to output first information that attracts a person's attention at a second timing that is a predetermined time before the first timing; and an authentication means for performing biometric authentication based on an image generated after the second timing.

[0009] This disclosure also provides a processing method in which one or more computers track a person detected from an image, calculate a first timing at which the size of the person's face in the image will reach a predetermined size suitable for biometric authentication based on the manner in which the size of the person's face in the image changes over time, output first information that attracts a person's attention to an output device at a second timing that is a predetermined time before the first timing, and perform biometric authentication based on an image generated after the second timing.

[0010] Furthermore, according to this disclosure, a program is provided that causes a computer to function as: a detection means for tracking a person detected from an image; a calculation means for calculating a first timing at which the size of a person's face in an image will reach a predetermined size suitable for biometric authentication based on the manner in which the size of the person's face in the image changes over time; an output control means for causing an output device to output first information that attracts a person's attention at a second timing that is a predetermined time before the first timing; and an authentication means for performing biometric authentication based on an image generated after the second timing.

[0011] According to one aspect of the present disclosure, a technique for capturing an image of a moving person's face in a manner suitable for biometric authentication is realized.

[0012] FIG. 1 is a diagram showing an example of a functional block diagram of a processing device. FIG. 2 is a flowchart showing an example of a processing flow of a processing device. FIG. 3 is a diagram showing an example of a hardware configuration of a processing device. FIG. 4 is a diagram showing an example of a usage scenario of a processing device. FIG. 5 is a flowchart showing another example of a processing flow of a processing device. FIG. 6 is a flowchart showing another example of a processing flow of a processing device. FIG. 7 is a diagram for explaining the functions of the processing device. FIG. 8 is another diagram for explaining the functions of the processing device. FIG. 9 is a diagram showing an example of information displayed by the processing device.

[0013] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In this disclosure, the drawings relate to one or more embodiments. In all drawings, similar components are designated by similar reference numerals, and descriptions thereof will be omitted as appropriate.

[0014] <<First Embodiment>> Fig. 1 is a functional block diagram showing an overview of a processing device 10. The processing device 10 realizes walk-through biometric authentication. That is, the processing device 10 performs biometric authentication by image analysis without requiring a person to stop. Fig. 2 is a flowchart showing an example of the flow of processing executed by the processing device 10.

[0015] 1, the processing device 10 includes a detection unit 11, a calculation unit 12, an output control unit 13, and an authentication unit 14. These functional units execute the process of the flowchart in FIG.

[0016] In S100, the detection unit 11 tracks a person detected from an image. In S101, the calculation unit 12 calculates a first timing at which the size of the person's face in the image becomes a predetermined size suitable for biometric authentication, based on the manner in which the size of the person's face in the image changes over time. In S102, the output control unit 13 causes the output device to output first information that attracts a person's attention at a second timing that is a predetermined time before the first timing. The output device is installed near the camera. In S103, biometric authentication is performed based on an image generated after the second timing.

[0017] However, there is a certain time lag between when the first information is output and when a person's gaze turns to it. If this point is not taken into consideration, it will be impossible to have the output device output the first information that will attract people's attention at the "appropriate timing."

[0018] If the timing of outputting the first information that attracts a person's attention from the output device is inappropriate, the person's gaze will turn toward the output device before the size of the person's face in the image reaches a predetermined size suitable for biometric authentication, and the person's gaze may be averted from the output device by the time the size of the person's face in the image reaches the predetermined size suitable for biometric authentication, which may be an inconvenience.

[0019] Furthermore, if the timing of outputting the first information that attracts a person's attention from the output device is inappropriate, the person's gaze will turn toward the output device after the size of the person's face in the image exceeds a predetermined size suitable for biometric authentication, which may cause an inconvenience in that the person's gaze will not turn toward the output device when the size of the person's face in the image reaches the predetermined size suitable for biometric authentication.

[0020] The processing device 10 calculates a first timing at which the size of a person's face in the image reaches a predetermined size suitable for biometric authentication, and causes the output device to output first information that attracts the person's attention at a second timing that is a predetermined time before the first timing, thereby reducing the inconvenience. The predetermined time is a value that is determined in advance, taking into consideration the time lag between the output of the first information and the person's gaze turning to it.

[0021] With this processing device 10, the person's gaze can be directed toward the output device at the timing when the size of the person's face in the image reaches a predetermined size suitable for biometric authentication. As a result, it is possible to capture an image in which the size of the person's face in the image is the predetermined size suitable for biometric authentication and the orientation of the person's face in the image is suitable for biometric authentication. In other words, it is possible to capture an image in which both the size and orientation of the person's face in the image are suitable for biometric authentication.

[0022] Furthermore, with the processing device 10 described above, the timing of capturing an image when both the size and orientation of the person's face in the image are suitable for biometric authentication occurs after the second timing at which the output device is caused to output the first information that attracts a person's attention. Ideally, this capturing timing occurs at the first timing described above. With the processing device 10 that performs biometric authentication based on an image generated after the second timing, biometric authentication can be performed based on an image where both the size and orientation of the person's face in the image are suitable for biometric authentication. As a result, the accuracy of biometric authentication is improved.

[0023] Furthermore, the processing device 10 calculates a first timing at which the size of a person's face in the image reaches a predetermined size suitable for biometric authentication, based on the manner in which the size of the person's face in the image changes over time. This processing device 10 can accurately calculate the first timing at which the size of the person's face in the image reaches a predetermined size suitable for biometric authentication. Based on this accurately calculated first timing, the processing device 10 can then appropriately determine a second timing at which the output device should output first information that attracts a person's attention.

[0024] <<Second Embodiment>> <Overview> A processing apparatus 10 according to a second embodiment is a specific embodiment of the configuration of the processing apparatus 10 according to the first embodiment. The processing apparatus 10 will be described in detail below.

[0025] <Hardware Configuration> First, an example of the hardware configuration of the processing device 10 will be described. Each functional unit of the processing device 10 is realized by any combination of hardware and software. Those skilled in the art will understand that there are various variations in the realization method and device. The software includes programs that are pre-loaded when the device is shipped, and programs downloaded from recording media such as CDs (Compact Discs) or servers on the Internet.

[0026] FIG. 3 is a block diagram illustrating an example of the hardware configuration of a processing device 10. As shown in FIG. 3, the processing device 10 has a processor 1A, a memory 2A, an input / output interface 3A, a peripheral circuit 4A, and a bus 5A. The peripheral circuit 4A includes various modules. The processing device 10 does not necessarily have to have the peripheral circuit 4A. Note that the processing device 10 may be composed of multiple devices that are physically and / or logically separated. In this case, each of the multiple devices may have the above hardware configuration.

[0027] The bus 5A is a data transmission path for the processor 1A, memory 2A, peripheral circuit 4A, and input / output interface 3A to mutually transmit and receive data. The processor 1A is, for example, a central processing unit (CPU) or a graphics processing unit (GPU). The memory 2A is, for example, a random access memory (RAM) or a read-only memory (ROM). The input / output interface 3A includes interfaces for acquiring information from input devices, external devices, external servers, external sensors, cameras, etc., and interfaces for outputting information to output devices, external devices, external servers, etc. The input / output interface 3A also includes an interface for connecting to a communication network such as the Internet. Examples of input devices include a keyboard, mouse, microphone, physical buttons, and touch panel. Examples of output devices include a display, projection device, speaker, printer, and mailer. The processor 1A can issue commands to each module and perform calculations based on the results of those calculations.

[0028] <Usage Scenarios> The processing device 10 realizes walk-through biometric authentication. That is, the processing device 10 performs biometric authentication by image analysis without requiring a person to stand still.

[0029] The processing device 10 is used, for example, at a passage spot where biometric authentication is required to pass through. The passage spot is, for example, an entrance to a facility, an entrance to a building, an entrance to a room, an entrance to a predetermined area, or the like.

[0030] An example of a usage scenario of the processing device 10 will be described with reference to FIG. 4. In the example of FIG. 4, the passage of a person is controlled by a door D. If the biometric authentication is successful, the door D is unlocked. If the biometric authentication is unsuccessful, the door D is locked. The opening and closing of the door D may be performed automatically under computer control, or may be performed manually. As another example, if the biometric authentication is successful, the door D may be opened. If the biometric authentication is unsuccessful, the door D may be closed. In this example, the opening and closing of the door D is performed automatically under computer control.

[0031] It should be noted that any other technology can be employed to control the passage of people. For example, instead of the door D, a gate that controls the passage of people with an obstacle such as a bar may be employed. If the biometric authentication is successful, the gate will be in an open state. The open state of the gate means that the obstacle such as a bar has been removed from the passage, allowing passage. If the biometric authentication fails, the gate will be in a closed state. The closed state of the gate means that an obstacle such as a bar is positioned in the passage, preventing passage.

[0032] Alternatively, there may be no obstructions to the person's passage, and an attendant may be standing nearby. In this example, the result of the biometric authentication is notified to the attendant via an output device such as a speaker, lamp, display, or projection device. The attendant then takes action according to the notified biometric authentication result. If the biometric authentication is successful, the attendant allows the person to pass through. If the biometric authentication fails, the attendant takes appropriate action, such as speaking to the person.

[0033] 4, a camera C and an output device M are installed at the passing spot. The camera C and the output device M are installed close to each other. The processing device 10 cooperates with the camera C and the output device M to execute predetermined processing.

[0034] The camera C may detect visible light and create an image. Alternatively, the camera C may detect infrared light and create an image. Alternatively, the camera C may detect other electromagnetic waves and create an image.

[0035] Camera C is installed at a position and orientation that allows it to capture an image of person P at the passing spot. Camera C generates a plurality of images in time series. The time interval between the plurality of images is not particularly limited, but may be a general frame rate for moving images. Using the plurality of images in time series generated by camera C, the face of person P is detected, person P is tracked, and the first timing and second timing are calculated. Furthermore, an image to be used for biometric authentication (face authentication, iris authentication, etc.) is selectively extracted from the plurality of images.

[0036] The output device M is configured to output information to a person P. The output device M may have any well-known configuration. For example, the output device M may include at least one of a speaker, a lamp, a display, and a projection device. The output device M may be a general-purpose device such as a tablet terminal, a smartphone, or a personal computer. Alternatively, the output device M may be a dedicated device created for use in combination with the processing device 10.

[0037] 4, the camera C and the output device M are shown separately, but the camera C and the output device M may be configured as physically and / or logically separate devices, or may be configured as an integrated device. Examples of an integrated camera C and output device M include a tablet terminal, a smartphone, a personal computer, etc. that have a camera function and an input / output function.

[0038] The processing device 10 (not shown in FIG. 4 ) is communicatively connected to the camera C and acquires images generated by the camera C in real time. When the processing device 10 detects a person P in the image generated by the camera C, it tracks the detected person P within the image. Next, the processing device 10 calculates a first timing at which the size of the person's face in the image reaches a predetermined size suitable for biometric authentication, based on the manner in which the size of the person's face in the image changes over time.

[0039] Next, at a second timing that is a predetermined time before the first timing, the processing device 10 causes the output device M to output first information that attracts a person's attention, as shown in Fig. 4 . The processing device 10 is communicatively connected to the output device M and can cause the output device M to output predetermined information. In the example of Fig. 4 , the output device M outputs a sound that attracts a person's attention as the first information. Furthermore, the output device M displays information on a display as the first information. Specifically, an arrow that guides the gaze of the person P, who has been guided toward the output device M by the sound output from the output device M, further toward the camera C is displayed as the first information.

[0040] Then, the processing device 10 performs biometric authentication of the person P based on the image generated at the second timing, i.e., after the timing when the output device M outputs the first information.

[0041] <Functional Configuration> Next, a detailed description will be given of the functional configuration of the processing device 10. Fig. 1 shows an example of a functional block diagram of the processing device 10. As shown in the figure, the processing device 10 has a detection unit 11, a calculation unit 12, an output control unit 13, and an authentication unit 14.

[0042] The detection unit 11 tracks a person detected from an image. That is, the detection unit 11 detects a person in an image generated by the camera C and tracks the detected person in the image. For example, the detection unit 11 may detect a person's face in an image generated by the camera C and track the detected person's face in the image. The detection of a person, a person's face, etc. in an image and the tracking of the detected object (a person, a person's face, etc.) in the image can be realized using widely known image analysis techniques.

[0043] The calculation unit 12 calculates a first timing at which the size of the person's face in the image becomes a predetermined size suitable for biometric authentication, based on the manner in which the size of the person's face in the image changes over time. The process executed by the calculation unit 12 will be described in detail below.

[0044] Calculating the size of a person's face in an image The "size of a person's face in an image" can be defined using various methods. An example will be described below, but the method is not limited to these.

[0045] In one example, the distance between two key points detected from a face in an image can be used as the size of the person's face in the image. The two key points are, for example, two eyes. In this case, the distance between the eyes represents the size of the person's face in the image. Note that the distance between key points other than the eyes may also be used as the size of the person's face in the image. The distance between the two key points may be expressed in terms of the number of pixels, or may be expressed as the distance between two points in a coordinate system set for the image.

[0046] In this example, the calculation unit 12 analyzes the image to detect predetermined key points. Then, the calculation unit 12 calculates the distance between the detected key points as a value indicating the size of a person's face in the image. The detection of the predetermined key points can be realized using widely known image analysis techniques.

[0047] In another example, the number of pixels included in a face region in the image can be used as the size of the person's face in the image. In this example, the calculation unit 12 detects the contour of the face using a technique such as semantic segmentation. Then, the calculation unit 12 calculates the number of pixels included in the contour of the face as a value indicating the size of the person's face in the image.

[0048] In another example, the width or height of a facial region in an image can be used as the size of a person's face in the image. In this example, the calculation unit 12 detects the contours of the face using a technique such as semantic segmentation. Then, the calculation unit 12 calculates the width (e.g., the width at the position where the value is maximum) or the height (e.g., the height at the position where the value is maximum) of the facial region as a value indicating the size of the person's face in the image. The width and height may be expressed in terms of the number of pixels or the distance between two points in a coordinate system set in the image.

[0049] The calculation unit 12 calculates the size of a person's face in each of a plurality of time-series images using, for example, the method described above. This allows the calculation unit 12 to calculate time-series data indicating changes in the size of a person's face in the images (a set of values ​​indicating the size of the face in each of a plurality of time-series images). Hereinafter, "time-series data indicating changes in the size of a person's face in the images (a set of values ​​indicating the size of the face in each of a plurality of time-series images)" may be simply referred to as "time-series data of face size."

[0050] Calculation of the manner in which the size of a person's face in an image changes over time The calculation unit 12 calculates the manner in which the size of a person's face in an image changes over time based on the time-series data of the calculated face size.

[0051] The "mode of change" includes at least one of a speed indicating the amount of change in face size per unit time and an acceleration indicating the amount of change in that speed per unit time. Hereinafter, the speed of change in the size of a person's face in an image may be simply referred to as the "speed of change in face size." Also, the acceleration of change in the size of a person's face in an image may be simply referred to as the "acceleration of change in face size."

[0052] In one example, the calculation unit 12 can calculate the average speed and / or average acceleration of the change in face size during a predetermined time period based on time-series data of face size over that time period. The average speed and average acceleration can be calculated using various methods, but one example will be described below.

[0053] For example, the calculation unit 12 divides the predetermined time length into a plurality of time intervals. The plurality of time intervals may be divided so that there are no overlapping time periods. Alternatively, adjacent time intervals may partially overlap each other. The time length of one time interval is a design matter.

[0054] The calculation unit 12 then calculates the amount of change in face size within each time interval (e.g., the difference in face size between the beginning and end of the time interval). The calculation unit 12 then calculates the rate of change in face size within each time interval by dividing the calculated amount of change by the duration of one time interval. The calculation unit 12 then calculates the average rate of change for multiple time intervals based on the calculation results.

[0055] The calculation unit 12 may calculate other statistical values ​​instead of calculating the average value of the speeds over a plurality of time intervals, such as, but not limited to, the maximum value, minimum value, mode, median, etc.

[0056] Furthermore, the calculation unit 12 can calculate the acceleration for each pair of adjacent time intervals by calculating the velocity for each of the plurality of time intervals and then dividing the change in velocity between the adjacent time intervals by the time difference between the adjacent time intervals. Then, the calculation unit 12 calculates the average value of the accelerations for the plurality of pairs based on the calculation results.

[0057] The calculation unit 12 may calculate other statistical values ​​instead of calculating the average value of a plurality of pairs of accelerations. Examples of statistical values ​​include, but are not limited to, maximum values, minimum values, modes, and medians.

[0058] In another example, the calculation unit 12 may calculate an approximation function for determining the size of a person's face in an image from time based on time-series data of face size for a predetermined time period. The approximation function can be calculated using any widely known technology. Note that even if a person is moving at a constant speed, the rate at which the size of the person's face in the image changes is not constant, and the closer the person's face is to the camera, the larger it tends to be. For this reason, the approximation function tends to be a polynomial function. Note that the calculated approximation function may also be a linear function.

[0059] The calculation unit 12 then calculates the speed of change in face size based on the result of differentiating this approximation function (differentiating the face size with respect to time). If the approximation function is a polynomial function, the calculation unit 12 calculates the speed at each of a plurality of predetermined timings based on the function obtained by differentiating the approximation function, and calculates the average value of these. Note that the calculation unit 12 may calculate other statistical values ​​as described above instead of the average value. If the approximation function is a linear function, the calculation unit 12 sets the value obtained by differentiating the approximation function as the average speed.

[0060] Furthermore, the calculation unit 12 calculates the acceleration of the change in face size based on the result of further differentiating the function obtained by differentiating the approximation function (differentiating the velocity with respect to time). When a function is obtained by this differentiation, the calculation unit 12 calculates accelerations at each of a plurality of predetermined timings based on the obtained function and calculates the average value of these. Note that the calculation unit 12 may calculate other statistical values ​​as described above instead of the average value. When a value is obtained by this differentiation, the calculation unit 12 sets the obtained value as the average acceleration. Furthermore, when the above approximation function is a linear function, the calculation unit 12 sets the average acceleration to 0.

[0061] Calculation of the first timing when the size of the person's face in the image becomes a predetermined size suitable for biometric authentication The calculation unit 12 calculates the first timing based on the "change in the size of the person's face in the image over time" calculated as described above, i.e., at least one of the speed of the change in the face size and the acceleration of the change in the face size. The calculation unit 12 calculates the first timing when the size of the person's face in the image becomes a predetermined size suitable for biometric authentication, assuming that the size of the person's face in the image will continue to change in the calculated change manner. There are various methods for this calculation, but one example will be described below.

[0062] First, a method for calculating (predicting) the size y of a person's face in an image at time t will be described.

[0063] In this example, a function is prepared in advance that calculates the size y of a person's face in an image at time t, specifically, a function that calculates the size y of a person's face in an image at time t based on the elapsed time from a reference timing (t=0). The elapsed time is the time from the reference timing to time t. For example, time-series data on the face sizes of multiple people (sampling subjects) for a predetermined period of time is collected through a pre-sampling operation. The function is then prepared by analyzing the sample data. The function may be created by a person and registered in the processing device 10, or it may be created by a computer.

[0064] The function may be a linear function or a polynomial function. Since the actual face size differs for each person, the coefficients of the function and the value of y at time t = 0 are values ​​that differ for each person. That is, the calculation unit 12 calculates these values ​​for each person and uses the calculated values ​​to complete a function for each person that calculates the size y of the person's face in the image at time t.

[0065] The coefficients of the function are predetermined so as to be determined based on at least one of the speed of change in face size and the acceleration of change in face size.

[0066] When the function is a linear function, it is predetermined so that the rate of change in face size is the coefficient.

[0067] When the function is a polynomial function, each of the multiple coefficients is predetermined to be determined by a predetermined formula that uses at least one of the speed of change in face size and the acceleration of change in face size as a variable, and calculates the coefficient. Note that at least one of the multiple coefficients may be given as a fixed value in advance (i.e., a value common to all people).

[0068] The calculation unit 12 calculates the coefficients of a function for each person based on at least one of the speed of change in face size and the acceleration of change in face size calculated for each person. Furthermore, the calculation unit 12 determines, for each person, a predetermined timing in the time-series data of face size for a predetermined length of time as a reference timing (t = 0), and specifies the size of the face in the image at that time. Then, based on the calculation results for each person, the calculation unit 12 completes a function for calculating, for each person, the size y of the person's face in the image at time t based on the elapsed time from the reference timing (t = 0).

[0069] Here, the "reference timing" will be described. As described above, the reference timing is the timing at which t = 0 in the function that calculates the size y of a person's face in an image at time t. This function uses the size of the person's face in the image at the reference timing as a reference. Then, this function calculates the size y of the person's face in the image at time t on the premise that the size of the person's face in the image changes from that reference state at the rate and acceleration of the change in face size described above. The size y of the person's face in the image at time t is calculated based on the elapsed time from the reference timing (t = 0) calculated based on time t.

[0070] By appropriately determining such a reference time, it is possible to accurately calculate the size of a person's face in an image at a time when a predetermined time has elapsed from the reference time.

[0071] For example, the reference timing may be the timing when the size of a person's face in the image becomes equal to or greater than a first threshold. If the size of the person's face in the image is too small, the accuracy of the size of the person's face in the image calculated at that time will be low. If such a calculation result with low accuracy is used as the standard for calculating the size of the person's face in the image at the above-mentioned time t, the accuracy of the calculation result of the size y of the person's face in the image at the time t will naturally be poor. This problem can be prevented by using the timing when the size of the person's face in the image becomes sufficiently large, i.e., the timing when the size becomes equal to or greater than the first threshold, as the reference timing.

[0072] The first threshold here indicates a lower limit of the size of a person's face in an image that is appropriate as a reference for calculating the size of the person's face in the image at time t. The first threshold is a different concept from the lower limit of the size of a person's face in an image that is appropriate for biometric authentication. The first threshold is a value smaller than the lower limit of the size of a person's face in an image that is appropriate for biometric authentication.

[0073] Based on the function for each person thus created, the calculation unit 12 calculates (predicts) the size y of the person's face in the image at time t for each person. Then, based on the calculation result (prediction result), the calculation unit 12 identifies a first timing at which the size y of the person's face in the image becomes a predetermined size suitable for biometric authentication.

[0074] The "predetermined size suitable for biometric authentication" is specified in advance and recorded in the processing device 10. For example, one value may be specified as the predetermined size suitable for biometric authentication and recorded in the processing device 10. Alternatively, a predetermined numerical range may be specified as the predetermined size suitable for biometric authentication and recorded in the processing device 10. In the latter case, the calculation unit 12 identifies a first timing (time t) at which the size y of the person's face in the image becomes a predetermined value within that numerical range. An example of the "predetermined value" is, but is not limited to, the minimum value within that numerical range.

[0075] The output control unit 13 causes the output device M to output first information that attracts a person's attention at a second timing that is a predetermined time before the first timing. The "predetermined time" here is a predetermined value. The predetermined time here is determined in consideration of the time lag between when the first information is output and when a person's gaze turns toward the output device M.

[0076] When the output device M includes a speaker, the output control unit 13 may output a predetermined sound (first information) from the speaker of the output device M. For example, the output control unit 13 may output a sound such as "beep" or a message such as "look at me."

[0077] Furthermore, if the data output device M is equipped with a lamp, the output control unit 13 may turn on or blink the lamp of the data output device M. The lighting or blinking of the lamp serves as the first information.

[0078] Furthermore, when the output device M has a display, the output control unit 13 may display the first information on the display of the output device M. For example, the output control unit 13 may display a message such as "Look towards the camera." Furthermore, the output control unit 13 may display an arrow pointing towards the camera.

[0079] Furthermore, if the output device M has a projection function, the output control unit 13 may cause the output device M to project information. For example, the output control unit 13 may cause the output device M to project the first information onto the floor or ground under a person's feet, or onto an object in the person's line of sight (e.g., door D in FIG. 4 ). For example, the output control unit 13 may cause the output control unit 13 to project a message such as "Look toward the camera." The output control unit 13 may also cause the output control unit 13 to project an arrow pointing toward the camera. The output control unit 13 may also cause the output control unit 13 to project information (such as a landmark) indicating the optimal standing position when capturing an image for biometric authentication onto the floor or ground under a person's feet.

[0080] The output control unit 13 may cause the output device M to output the above-described first information for a relatively short period of time at a second timing that is a predetermined time before the first timing.

[0081] Alternatively, the output control unit 13 may cause the output device M to continue outputting the first information from the second timing to the output end timing. For example, the output control unit 13 may send a signal to the output device M at the second timing to start outputting the first information. The output device M starts outputting the first information in response to receiving the signal. Then, the output control unit 13 may send a signal to the output device M to end output of the first information at the output end timing. The output device M ends output of the first information in response to receiving the signal.

[0082] The output end timing is one of the following: - The time when the biometric authentication result is obtained - The time when biometric authentication is successful - The time when an image of the person's face in the image that meets the specified quality conditions is obtained - The time when the size of the person's face in the image falls within a specified range suitable for biometric authentication and then goes out of the specified range - The time when the person's face in the image goes out of the frame

[0083] "Timing when the biometric authentication result is obtained" The timing when the biometric authentication result is obtained is the timing when the biometric authentication result is obtained by the authentication unit 14, which will be described later. It does not matter what the biometric authentication result (success / failure) is. After the biometric authentication is completed, the person does not need to face the camera C. Therefore, the output control unit 13 can end the output of the first information at this timing.

[0084] In this example, the authentication unit 14 performs biometric authentication based on the image generated after the second timing, and when the result is obtained, notifies the output control unit 13. In response to the notification, the output control unit 13 transmits a signal to the output device M to terminate output of the first information. In response to receiving the signal, the output device M terminates output of the first information.

[0085] "Timing of successful biometric authentication" The timing of successful biometric authentication is the timing at which a "success" result is obtained in biometric authentication by the authentication unit 14, which will be described later. After biometric authentication is "successful," the person does not need to face the camera C. On the other hand, if biometric authentication is "failed," biometric authentication based on another image may continue thereafter. In this case, it is preferable for the person to continue facing the camera C even after a "failed" biometric authentication is obtained. Therefore, the output control unit 13 can end the output of the first information at the timing at which biometric authentication is successful.

[0086] In this example, the authentication unit 14 performs biometric authentication based on the image generated after the second timing, and when a result of "success" is obtained, it notifies the output control unit 13. In response to the notification, the output control unit 13 transmits a signal to the output device M to terminate output of the first information. In response to receiving the signal, the output device M terminates output of the first information.

[0087] "Timing when an image of a person's face in an image that satisfies predetermined quality conditions is obtained" As will be described in detail later, before performing biometric authentication, the authentication unit 14, which will be described later, determines whether a person's face in an image satisfies predetermined quality conditions, and can perform biometric authentication using an image in which the person's face satisfies the predetermined quality conditions. After an image of a quality that allows biometric authentication to be performed is obtained, the person does not need to face the camera C. Therefore, the output control unit 13 can end the output of the first information at this timing.

[0088] In this example, the authentication unit 14 performs the above-described process based on an image generated after the second timing. Then, when the authentication unit 14 obtains an image in which the human face in the image satisfies a predetermined quality condition, it notifies the output control unit 13 of this fact. In response to this notification, the output control unit 13 transmits a signal to the output device M to terminate output of the first information. In response to receiving this signal, the output device M terminates output of the first information.

[0089] "The timing when the size of the person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside the predetermined range" If the size of the person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside the predetermined range, biometric authentication can no longer be performed. Therefore, the person no longer needs to face camera C. Therefore, the output control unit 13 can end the output of the first information when the size of the person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside the predetermined range.

[0090] In this example, the calculation unit 12 analyzes the image generated by the camera C to detect the timing at which the size of the person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside the predetermined range. The calculation unit 12 then notifies the output control unit 13 that such timing has been detected. In response to the notification, the output control unit 13 transmits a signal to the output device M to terminate output of the first information. In response to receiving the signal, the output device M terminates output of the first information.

[0091] "When the person's face in the image goes out of the frame" When the person's face in the image goes out of the frame, biometric authentication can no longer be performed. Therefore, there is no need for the person to face the camera C from that point on. Therefore, the output control unit 13 can end the output of the first information when the person's face in the image goes out of the frame.

[0092] In this example, the calculation unit 12 can detect the timing when the face of a person in the image goes out of frame by analyzing the image generated by the camera C. Then, the calculation unit 12 notifies the output control unit 13 that such timing has been detected. In response to the notification, the output control unit 13 transmits a signal to the output device M to terminate output of the first information. In response to receiving the signal, the output device M terminates output of the first information.

[0093] The timing of framing out may be the timing when the entire person's face disappears from the image, or the timing when part of the person's face disappears from the image, or the timing when the person's face is no longer detected in the image in a face detection process by a computer.

[0094] The authentication unit 14 performs biometric authentication based on an image generated by the camera C after the second timing.

[0095] "Biometric authentication" is an authentication process based on an image of a person's face. Examples of biometric authentication include face authentication and iris authentication. The authentication unit 14 can perform biometric authentication using any well-known technology. For example, biometric information for each of multiple people is registered in a database in advance. The biometric information is information used for biometric authentication. Examples of biometric information include, but are not limited to, face information (face image, facial appearance features, etc.) and iris information (iris image, iris appearance features, etc.). The authentication unit 14 extracts the person's biometric information from the image generated by the camera C after the second timing. The authentication unit 14 then performs biometric authentication by comparing the biometric information extracted from the image with biometric information registered in the database. If biometric information matching the biometric information extracted from the image is registered in the database, the authentication unit 14 determines that the authentication is successful. On the other hand, if biometric information matching the biometric information extracted from the image is not registered in the database, the authentication unit 14 determines that the authentication is unsuccessful.

[0096] The authentication unit 14 may perform biometric authentication based on, for example, an image generated at the first timing calculated by the calculation unit 12 .

[0097] Alternatively, the authentication unit 14 may determine whether the human face in an image generated by the camera C after the second timing satisfies a predetermined quality condition. Then, the authentication unit 14 may perform biometric authentication based on the image that satisfies the quality condition.

[0098] The "quality condition" is, for example, one of the following conditions, or two or more of the following conditions connected by a predetermined logical operator: - The size of the person's face in the image is a predetermined size suitable for biometric authentication - The orientation of the person's face in the image is a predetermined orientation suitable for biometric authentication - The person's face in the image is not blurred

[0099] The size of a person's face in an image can be detected using the method described above, and a predetermined size suitable for biometric authentication is registered in advance as described above.

[0100] The orientation of a person's face in an image can be detected using any well-known image analysis technology. The predetermined orientation suitable for biometric authentication is facing the camera C directly or within a range of slight deviations therefrom. For example, the authentication unit 14 may make this determination using a classifier generated by machine learning that classifies whether the orientation of a person's face in an image is the predetermined orientation suitable for biometric authentication.

[0101] The unblurred nature of a person's face in an image can be detected using any well-known image analysis technique.

[0102] If the biometric authentication is successful, the authentication unit 14 performs a predetermined process to allow the person to pass. For example, the authentication unit 14 may send information indicating that the biometric authentication was successful to a control device that controls a mechanism (door D, gate, etc.) that controls the passage of people. The control device unlocks or opens the door D or the gate in response to the information. Alternatively, if the biometric authentication is successful, the authentication unit 14 may cause an output device that notifies an attendant located at the passage spot to output information indicating that the biometric authentication was successful.

[0103] If the biometric authentication fails, the authentication unit 14 performs error processing. For example, the authentication unit 14 may send information indicating that the biometric authentication failed to a control device that controls a mechanism (door D, gate, etc.) that controls the passage of people. The control device may lock the door D, close the door D, or close the gate in response to the information. Alternatively, the authentication unit 14 may notify the output control unit 13 that the biometric authentication failed and cause the output control unit 13 to output error information. The output control unit 13 notifies the person attempting to pass through the passing spot that the biometric authentication failed via the output device described above. The output control unit 13 may also notify an attendant located at the passing spot that the biometric authentication of the person attempting to pass through the passing spot has failed.

[0104] Next, an example of the processing flow of the processing device 10 will be described using the flowchart in Figure 5. Note that the purpose here is to explain the processing flow. Since the details of each process have been described above, the explanation here will be omitted as appropriate. Note that, although an example of detecting and tracking a human face in an image will be described here, the present invention is not limited to this example.

[0105] The processing device 10 performs a process of detecting a human face in the image generated by the camera C (S200). If a human face is not detected in the image (No in S200), or if there is no instruction to end the process (No in S211), the processing device 10 continues the process of S200.

[0106] If a human face is detected in the image (Yes in S200), the processing device 10 subsequently tracks the detected face in the image (S201). The processing device 10 then calculates the change in size of the human face in the image over time (S202). For example, the processing device 10 calculates the average speed and / or average acceleration of the change in face size. Next, based on the calculated change in face size, the processing device 10 calculates a first timing at which the size of the human face in the image will reach a predetermined size suitable for biometric authentication (S203).

[0107] Thereafter, the processing device 10 monitors the arrival of a second timing, which is a predetermined time before the first timing calculated in S203 (S204). When the second timing arrives (Yes in S204), the processing device 10 causes the output device M to output first information that attracts a person's attention (S205).

[0108] Thereafter, the processing device 10 generates an image for biometric authentication (S206). Specifically, the processing device 10 sets a predetermined image generated by the camera C after the second timing as the image for biometric authentication.

[0109] The processing device 10 then performs biometric authentication based on the image generated in S206 (S207). If the biometric authentication is successful (Yes in S208), the processing device 10 performs a predetermined process to allow the person to pass (S209). On the other hand, if the biometric authentication is unsuccessful (No in S208), the processing device 10 performs error processing (S210).

[0110] "Operational Effects" According to the processing apparatus 10 of the second embodiment, operational effects similar to those of the processing apparatus 10 of the first embodiment are realized.

[0111] Furthermore, the processing device 10 of the second embodiment can perform biometric authentication based on "an image generated at the first timing calculated by the calculation unit 12" among images generated after the second timing at which the output device M is caused to output first information that attracts a person's attention. It is expected that the image generated at the first timing will have a high probability of being suitable for biometric authentication in terms of both the size and orientation of the person's face in the image. Performing biometric authentication based on such an image generated at the first timing improves the accuracy of biometric authentication.

[0112] Furthermore, the processing device 10 of the second embodiment can perform biometric authentication based on an "image in which a human face is determined to satisfy a predetermined quality condition" among images generated after the second timing at which the output device M is caused to output the first information that attracts a person's attention. With this processing device 10, biometric authentication can be performed based on an image in which both the size and orientation of the human face in the image are determined to be suitable for biometric authentication. As a result, the accuracy of biometric authentication is improved.

[0113] Furthermore, the processing device 10 of the second embodiment can continue outputting the first information from the second timing until the output end timing described above. By continuing to output the first information, it is possible to maintain a state in which the person's gaze is directed toward the output device M or the camera C. Furthermore, by ending the output of the first information at the appropriate output end timing described above, it is possible to suppress the inconvenience of unnecessarily maintaining a state in which the person's gaze is directed toward the output device M or the camera C.

[0114] <<Third Embodiment>> A processing device 10 according to a third embodiment identifies the timing at which the blurring of a face in an image satisfies a criterion based on the gait of a person detected in an image, and performs biometric authentication based on an image generated by a camera C at that timing. This will be described in detail below.

[0115] The calculation unit 12 calculates a period at which the degree of blurring of the face satisfies a criterion based on the gait of the person detected in the image. Then, based on the manner in which the size of the face of the person in the image changes over time and the period, the calculation unit 12 calculates, as a first time, the time at which the size of the face of the person in the image becomes a predetermined size suitable for biometric authentication and the degree of blurring of the face satisfies the criterion.

[0116] In the third embodiment, the size of a person's face in an image suitable for biometric authentication is specified within a predetermined numerical range. Therefore, the timing when the person's face detected in the image reaches the predetermined size suitable for biometric authentication is not a pinpoint timing but a time period with a certain duration. The calculation unit 12 calculates, as the first timing, the timing when the blur of the face in the image satisfies a criterion within such a "time period when the person's face detected in the image reaches the predetermined size suitable for biometric authentication."

[0117] In one example, the calculation unit 12 analyzes each of a plurality of time-series images and identifies the timing at which both feet of the person detected in the image touch the ground. Then, based on the identification result, the calculation unit 12 identifies the period at which both feet of the person detected in the image touch the ground.

[0118] In this example, the "timing when both feet of the person detected in the image are on the ground" is the "timing when facial blurring meets the criteria." When both feet of the person detected in the image are on the ground, the person's posture and movement are stable, and facial blurring is reduced. Images generated at such timings are expected to have no or minimal facial blurring.

[0119] In another example, the calculation unit 12 identifies key points of a person in each of a plurality of time-series images. The key points are preferably parts of the face, such as the eyes and nose. Then, the calculation unit 12 calculates the acceleration of the movement speed of the key points of the person based on changes in the positions of the identified key points. Note that the calculation unit 12 may also calculate acceleration of a component in a predetermined direction, such as acceleration in the vertical direction. After calculating the acceleration, the calculation unit 12 identifies a period in which the acceleration reaches a minimum value.

[0120] The above-mentioned up-down direction can be defined in various ways. For example, the above-mentioned up-down direction may be the up-down direction of the image. Alternatively, the direction of gravity may be defined as the down direction, and the direction 180° opposite thereto may be defined as the up-down direction. Alternatively, the up-down direction may be a direction perpendicular to the floor surface in the image. Alternatively, the up-down direction may be set based on the position of a key point in the coordinate system of the image. Note that this premise also applies to the "up-down direction" below.

[0121] In this example, the "timing at which the acceleration reaches a minimum value" is the "timing at which the degree of facial blurring meets the criteria." When the acceleration of the movement speed of a person's key points is relatively small, facial blurring is suppressed compared to when the acceleration is relatively large. Images generated at such times are expected to have little or no facial blurring.

[0122] In another example, the calculation unit 12 identifies a key point of a person in each of a plurality of time-series images. The key point is preferably a part of the face, such as the eyes or nose. The calculation unit 12 then calculates the direction of change in the vertical position of the identified key point over time. The calculation unit 12 then identifies a period of time at which the direction of change in the vertical position changes. Alternatively, the calculation unit 12 may identify a period of time at which the direction of change in the vertical position changes from downward to upward, or a period of time at which the direction of change in the vertical position changes from upward to downward.

[0123] In this example, the "timing at which the direction of change in the vertical position of the key point switches" is the "timing at which the degree of facial blurring meets the criterion." Other examples of the "timing at which the direction of change in the vertical position of the key point switches from downward to upward" or "timing at which the direction of change in the vertical position of the key point switches from upward to downward" are the "timing at which the degree of facial blurring meets the criterion." While the person is moving in the direction of travel, the key point repeats vertical movement. It is expected that images generated at the above-mentioned "timing at which the direction of change in the vertical position of the key point switches" or the like will have reduced facial blurring caused by this vertical movement.

[0124] Calculation of the first timing As described above, the calculation unit 12 calculates, based on the manner in which the size of the person's face in the image changes over time and the above-mentioned period, the timing at which the size of the person's face in the image becomes a predetermined size suitable for biometric authentication and the degree of blurring of the face satisfies a standard, as the first timing.

[0125] First, the calculation unit 12 identifies a time period during which the size of the person's face in the image reaches a predetermined size suitable for biometric authentication, based on the manner in which the size of the person's face in the image changes over time. As described above, in the third embodiment, the size of the person's face in the image suitable for biometric authentication is specified within a predetermined numerical range. Therefore, in one example, the calculation unit 12 identifies a time period during which the size of the person's face in the image reaches the "minimum value within the predetermined numerical range" and the "maximum value within the predetermined numerical range," based on the manner in which the size of the person's face in the image changes over time. Then, the calculation unit 12 identifies the time period between the two identified times as the time period during which the size of the person's face in the image reaches the predetermined size suitable for biometric authentication.

[0126] Next, the calculation unit 12 calculates, as the first timing, the "timing at which the blurring of the face satisfies the criteria" based on the calculated period within the identified "time period during which the size of the person's face in the image becomes a specified size suitable for biometric authentication."

[0127] The calculation unit 12 calculates the first timing on the premise that the "timing at which the degree of facial blurring satisfies the criterion" occurs in the calculated cycle. For example, the calculation unit 12 calculates the "timing at which the degree of facial blurring satisfies the criterion" (actually observed timing) by analyzing an image generated by the camera C. Then, the calculation unit 12 calculates the first timing on the premise that the "timing at which the degree of facial blurring satisfies the criterion" occurs repeatedly in the calculated cycle after the actually observed timing. The calculation unit 12 may calculate one timing as the first timing, or may calculate multiple timings as the first timing.

[0128] Next, an example of the processing flow of the processing device 10 will be described using the flowchart in Figure 6. Note that the purpose here is to explain the processing flow. Since the details of each process have been described above, the explanation here will be omitted as appropriate. Note that, although an example of detecting and tracking a human face in an image will be described here, the present invention is not limited to this example.

[0129] The processing device 10 performs a process of detecting a human face in the image generated by the camera C (S300). If a human face is not detected in the image (No in S300), or if there is no instruction to end the process (No in S312), the processing device 10 continues the process of S300.

[0130] If a human face is detected in the image (Yes in S300), the processing device 10 subsequently tracks the detected face within the image (S301). The processing device 10 then calculates the change in size of the human face in the image over time (S302). For example, the processing device 10 calculates the average speed and / or average acceleration of the change in face size. Furthermore, the processing device 10 calculates the period at which the degree of face blurring satisfies a criterion based on the gait of the person whose face is detected (S303).

[0131] Next, based on the calculated change in face size and the period calculated in S303, the processing device 10 calculates the first timing at which the size of the person's face in the image becomes a predetermined size suitable for biometric authentication and the degree of face blurring meets the criteria (S304).

[0132] Thereafter, the processing device 10 monitors the arrival of a second timing, which is a predetermined time before the first timing calculated in S304 (S305). When the second timing arrives (Yes in S305), the processing device 10 causes the output device M to output first information that attracts a person's attention (S306).

[0133] When the processing device 10 calculates one timing as the first timing, it monitors the arrival of one second timing that is a predetermined time before the one first timing, and when the one second timing arrives, the processing device 10 causes the output device M to output first information that attracts a person's attention.

[0134] On the other hand, when the processing device 10 calculates multiple timings as the first timing, it monitors the arrival of a second timing that is a predetermined time before the earliest timing among the multiple timings. Then, when the second timing arrives, the processing device 10 causes the output device M to output first information that attracts a person's attention. In this case, the processing device 10 may cause the output device M to continue outputting the first information from the second timing until the output end timing. The output end timing is as described in the second embodiment.

[0135] As another example, when the processing device 10 calculates multiple timings as first timings, it may monitor the arrival of multiple second timings that are a predetermined time before each of the multiple timings. Then, the processing device 10 may cause the output device M to output first information that attracts a person's attention each time each of the multiple second timings arrives. In this case, the processing device 10 may cause the output device M to output the first information for only a relatively short time at each of the multiple second timings. For example, the processing device 10 may cause the output device M to output a relatively short sound (first information) such as a "beep" at each of the multiple second timings.

[0136] Thereafter, the processing device 10 generates an image for biometric authentication (S307). Specifically, the processing device 10 sets a predetermined image generated by the camera C after the second timing as the image for biometric authentication.

[0137] The processing device 10 then performs biometric authentication based on the image generated in S307 (S308). If the biometric authentication is successful (Yes in S309), the processing device 10 performs a predetermined process to allow the person to pass (S310). On the other hand, if the biometric authentication is unsuccessful (No in S309), the processing device 10 performs error processing (S311).

[0138] Other configurations of the processing apparatus 10 of the third embodiment are similar to those of the processing apparatus 10 of the first and second embodiments.

[0139] According to the processing apparatus 10 of the third embodiment, the same effects as those of the processing apparatus 10 of the first and second embodiments are achieved.

[0140] Furthermore, the processing device 10 of the third embodiment calculates a period at which the degree of facial blurring meets a criterion based on the gait of a person detected in an image. The processing device 10 then calculates a first timing as the timing at which the size of the person's face in the image reaches a predetermined size suitable for biometric authentication and the degree of facial blurring meets the criterion. The processing device 10 then controls the output device to output first information that attracts a person's attention at a second timing that is a predetermined time before the calculated first timing.

[0141] According to the processing device 10 of the third embodiment, the person's gaze can be directed toward the output device at a timing when the size of the person's face in the image reaches a predetermined size suitable for biometric authentication and the degree of facial blurring meets a standard. As a result, it is possible to capture an image in which the size of the person's face in the image is the predetermined size suitable for biometric authentication, facial blurring is suppressed, and the orientation of the person's face in the image is suitable for biometric authentication. In other words, it is possible to capture an image in which the size, orientation, and state of the person's face in the image are all suitable for biometric authentication. Performing biometric authentication based on such an image improves the accuracy of biometric authentication.

[0142] <<Fourth Embodiment>> In the fourth embodiment, a function for calculating (predicting) the size y of a person's face in an image at time t is prepared for each person's movement trajectory. Then, the processing device 10 calculates (predicts) the size y of a person's face in an image at time t using a function according to the movement trajectory of a person whose face has been detected in the image. This will be described in detail below.

[0143] As shown in Figures 7 to 9, the movement trajectory of a person at a passing spot can take various forms. The way in which the size of the person's face in the image changes over time can vary for each movement trajectory. １ The movement trajectory R １ And, Person P ２ The movement trajectory R ２ And, Person P ３ The movement trajectory R ３ is shown.

[0144] In the fourth embodiment, for example, time-series data on the face sizes of multiple people (sampling subjects) for a predetermined period of time is collected by a preliminary sampling operation. The data is then grouped based on people's movement trajectories. The sample data for each group is then analyzed, and the above-mentioned function for each movement trajectory is prepared. The function may be created by a person and registered in the processing device 10, or may be created by a computer.

[0145] The data grouping may be performed by a person or a computer. The grouping of the trajectories can be calculated using any well-known technique. For example, the similarity between two curves (trajectories) may be calculated, and the data may be grouped based on the calculated similarity.

[0146] The functions for each trajectory may have different orders. For example, the function for one trajectory may be an n-th order function, and the function for another trajectory may be an m-th order function. Note that n and m are different values. Alternatively, the functions for each trajectory may have the same order, but the coefficient values ​​and the contents of the predetermined formulas for calculating the coefficients may be different.

[0147] The calculation unit 12 calculates the first timing based on the manner in which the size of the person's face changes in the image over time and the movement trajectory of the person.

[0148] First, the calculation unit 12 identifies which of the above-mentioned groups of trajectories the trajectory of a person detected in an image corresponds to. This identification can be achieved by various methods.

[0149] In one example, a representative trajectory is determined in advance for each group. The calculation unit 12 then calculates the similarity between the trajectory of a person detected in the image and each representative trajectory of each group. The calculation unit 12 then identifies the group with the highest similarity as the group corresponding to the trajectory of the person detected in the image.

[0150] The movement trajectory of a person detected in an image can be identified using any well-known technology. For example, the position of the person detected in the image in three-dimensional space can be identified and the identified positions can be connected in chronological order to identify the person's movement trajectory. The position of the person detected in the image in three-dimensional space may be identified using a position detection sensor such as a stereo camera or Lidar. Alternatively, a calculation algorithm for calculating a position in three-dimensional space from a two-dimensional image may be used. Alternatively, a camera may be installed that captures passing spots from above to below. Then, the person may be detected in the two-dimensional image generated by this camera and tracked within the image to identify the person's movement trajectory.

[0151] After identifying a group corresponding to the movement trajectory of a person detected in an image, the calculation unit 12 completes a function for the identified group based on the manner in which the size of the face of the person detected in the image changes over time. This process is as described in the second embodiment. That is, the calculation unit 12 calculates the coefficients of the function for the identified group based on the average speed, average acceleration, etc. of each person. The calculation unit 12 also identifies the value of y (the size of the face in the image) at t = 0 for each person. Then, the calculation unit 12 completes the function based on these results.

[0152] Other configurations of the processing apparatus 10 of the fourth embodiment are similar to those of the processing apparatus 10 of the first to third embodiments.

[0153] According to the processing apparatus 10 of the fourth embodiment, the same effects as those of the processing apparatus 10 of the first to third embodiments are realized.

[0154] Furthermore, the processing device 10 of the fourth embodiment calculates (predicts) the size y of a person's face in an image at time t using a function corresponding to the person's movement trajectory. Then, based on the calculation result, the processing device 10 calculates a first timing at which the size of the person's face in the image will reach a predetermined size suitable for biometric authentication. This processing device 10 can accurately calculate the first timing at which the size of the person's face in the image will reach a predetermined size suitable for biometric authentication. Then, based on the accurately calculated first timing, the processing device 10 can appropriately determine a second timing at which the output device should output first information that attracts a person's attention.

[0155] <<Fifth Embodiment>> A processing device 10 according to a fifth embodiment can notify a person of various characteristic pieces of information. This will be described in detail below.

[0156] The output control unit 13 executes a process of notifying a person of the arrival of at least one of the following timings. This process is a process of causing the output device M to output information indicating the arrival of a predetermined timing. - The timing when the size of a person's face in an image falls within a predetermined range suitable for biometric authentication - The timing when the size of a person's face in an image falls within a predetermined range suitable for biometric authentication, but then goes outside the predetermined range - The timing when the size of a person's face in an image falls within a predetermined range suitable for biometric authentication, but then goes outside the predetermined range, without the biometric authentication being successful - The timing when a person's face in an image falls out of the frame - The timing when a person's face in an image falls out of the frame, without the biometric authentication being successful - The timing when an image of a person's face in an image that satisfies predetermined quality conditions is obtained - The timing when the size of a person's face in an image falls within a predetermined range suitable for biometric authentication, but then goes outside the predetermined range, without the image satisfying the predetermined quality conditions being obtained - The timing when a person's face in an image falls out of the frame, without the image satisfying the predetermined quality conditions being obtained

[0157] "The moment when the size of a person's face in the image falls within a predetermined range suitable for biometric authentication." By notifying people of the arrival of this moment, they can be made aware of the timing to take an image suitable for biometric authentication. Then, when people take an action suitable for biometric authentication (such as opening their eyes wide), it is expected that the accuracy of biometric authentication will improve.

[0158] In this example, the calculation unit 12 can detect the timing when the size of a person's face in the image falls within a predetermined range suitable for biometric authentication by analyzing the image generated by the camera C. The calculation unit 12 then notifies the output control unit 13 that such timing has been detected. In response to the notification, the output control unit 13 causes the output device M to output timing arrival information.

[0159] "The timing when the size of a person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside that range" If the size of a person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside that range, biometric authentication can no longer be performed. By notifying the person of the arrival of this timing, it is possible to make the person aware that there is no need to take any action suitable for biometric authentication from that point on (such as opening their eyes wide or looking towards camera C).

[0160] In this example, the calculation unit 12 analyzes the image generated by the camera C to detect the timing at which the size of the person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside the predetermined range. The calculation unit 12 then notifies the output control unit 13 that such timing has been detected. In response to the notification, the output control unit 13 causes the output device M to output timing arrival information.

[0161] "The moment when the size of a person's face in the image falls within a predetermined range suitable for biometric authentication, but then falls outside the predetermined range without biometric authentication being successful" If the size of a person's face in the image falls within a predetermined range suitable for biometric authentication, but then falls outside the predetermined range, biometric authentication can no longer be performed. If biometric authentication is successful at that point, there is no particular problem. However, if biometric authentication is not successful at that point, the person must return to a position where the size of the person's face in the image falls within the predetermined range suitable for biometric authentication, and then try biometric authentication again. By notifying the person of the arrival of such a moment, it is possible to encourage the person to take the action described above.

[0162] In this example, the calculation unit 12 analyzes the image generated by the camera C to detect the timing at which the size of the person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside the predetermined range. The calculation unit 12 then notifies the output control unit 13 that such timing has been detected. Furthermore, if the biometric authentication is successful, the authentication unit 14 notifies the output control unit 13 of this fact.

[0163] Then, if the output control unit 13 receives the above notification from the calculation unit 12 without receiving the above notification from the authentication unit 14, it causes the output device M to output timing arrival information in response to the above notification from the calculation unit 12.

[0164] "The moment when the person's face in the image goes out of frame" When the person's face goes out of frame in the image, biometric authentication can no longer be performed. By notifying the person of this timing, it is possible to make the person aware that there is no need to take any action appropriate for biometric authentication from that point on (such as opening their eyes wide or looking towards camera C).

[0165] In this example, the calculation unit 12 can detect the timing at which the face of a person in the image goes out of frame by analyzing the image generated by the camera C. Then, the calculation unit 12 notifies the output control unit 13 that such timing has been detected. In response to the notification, the output control unit 13 causes the output device M to output timing arrival information.

[0166] The timing of framing out may be the timing when the entire person's face disappears from the image, or the timing when part of the person's face disappears from the image, or the timing when the person's face is no longer detected in the image in a face detection process by a computer.

[0167] "The moment when the person's face in the image goes out of the frame without successful biometric authentication" When the person's face in the image goes out of the frame, biometric authentication can no longer be performed. If biometric authentication is successful at that point, there is no particular problem. However, if biometric authentication is not successful at that point, the person must return to the shooting area of ​​camera C, for example, and try biometric authentication again. By notifying the person of the arrival of such a moment, it is possible to encourage the person to take the action described above.

[0168] In this example, the calculation unit 12 can detect the timing at which the face of a person in the image goes out of frame by analyzing the image generated by the camera C. Then, the calculation unit 12 notifies the output control unit 13 that such timing has been detected. Furthermore, if the biometric authentication is successful, the authentication unit 14 notifies the output control unit 13 of that fact.

[0169] Then, if the output control unit 13 receives the above notification from the calculation unit 12 without receiving the above notification from the authentication unit 14, it causes the output device M to output timing arrival information in response to the above notification from the calculation unit 12.

[0170] "The timing at which an image of a person's face that satisfies predetermined quality conditions is obtained" After an image of a person's face that satisfies predetermined quality conditions, that is, an image of a quality that allows biometric authentication to be performed, is obtained, the person does not need to take any action appropriate for biometric authentication (such as opening their eyes wide or turning toward camera C). By notifying the person of the arrival of such a timing, it is possible to make the person aware that there is no need to take any action appropriate for biometric authentication from that point on.

[0171] In this example, when an image in which the human face satisfies a predetermined quality condition is obtained, the authentication unit 14 notifies the output control unit 13. In response to the notification, the output control unit 13 causes the output device M to output timing arrival information.

[0172] "The timing when the size of a person's face in an image falls outside the specified range after falling within the specified range suitable for biometric authentication, without an image satisfying the specified quality conditions being obtained." If the size of a person's face in an image falls outside the specified range after falling within the specified range suitable for biometric authentication, biometric authentication can no longer be performed. If an image of the person's face in the image that meets the specified quality conditions has been obtained at that point, there is no particular problem. However, if an image of the person's face in the image that meets the specified quality conditions has not been obtained at that point, the person must return to a position where the size of the person's face in the image falls within the specified range suitable for biometric authentication, and then repeat the biometric authentication. By notifying a person of the arrival of such a timing, it is possible to encourage the person to take the action described above.

[0173] In this example, the calculation unit 12 analyzes the image generated by the camera C to detect the timing at which the size of the person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside the predetermined range. The calculation unit 12 then notifies the output control unit 13 that such timing has been detected. Furthermore, when an image is obtained in which the person's face in the image satisfies predetermined quality conditions, the authentication unit 14 notifies the output control unit 13 of this fact.

[0174] Then, if the output control unit 13 receives the above notification from the calculation unit 12 without receiving the above notification from the authentication unit 14, it causes the output device M to output timing arrival information in response to the above notification from the calculation unit 12.

[0175] "The moment when a person's face in an image goes out of frame while an image of the person's face that meets the specified quality conditions has not been obtained" When a person's face goes out of frame in an image, biometric authentication can no longer be performed. If an image of the person's face that meets the specified quality conditions has been obtained at that point, there is no particular problem. However, if an image of the person's face that meets the specified quality conditions has not been obtained at that point, the person must return to the shooting area of ​​camera C, for example, and repeat biometric authentication. By notifying the person of the arrival of such a moment, it is possible to encourage the person to take the action described above.

[0176] In this example, the calculation unit 12 can detect the timing at which a person's face in the image goes out of frame by analyzing the image generated by the camera C. Then, the calculation unit 12 notifies the output control unit 13 that such timing has been detected. Furthermore, when an image in which the person's face in the image satisfies a predetermined quality condition is obtained, the authentication unit 14 notifies the output control unit 13 to that effect.

[0177] Then, if the output control unit 13 receives the above notification from the calculation unit 12 without receiving the above notification from the authentication unit 14, it causes the output device M to output timing arrival information in response to the above notification from the calculation unit 12.

[0178] The output control unit 13 may notify a person of the arrival of one of the multiple timings described above. Alternatively, the output control unit 13 may notify a person of the arrival of two or more of the multiple timings described above.

[0179] When notifying a person of the arrival of two or more timings, the output device M may output different timing arrival information in response to the arrival of each timing. For example, the timing arrival information corresponding to the arrival of each of the two or more timings may be different sounds. Furthermore, the timing arrival information corresponding to the arrival of each of the two or more timings may be the lighting or blinking of lamps of different colors. Furthermore, the timing arrival information corresponding to the arrival of each of the two or more timings may be the lighting or blinking of lamps in different ways. Furthermore, the timing arrival information corresponding to the arrival of each of the two or more timings may be images containing different information (messages, figures, etc.).

[0180] In this case, when causing the output device M to output timing arrival information, the output control unit 13 may notify the output device M of the content of the timing arrival information to be output (such as the sound to be output, the color of the lamp, how the lamp lights up or flashes, and the content of the information to be output).The output device M may then output the timing arrival information based on the content notified by the output control unit 13.

[0181] Alternatively, the contents of timing arrival information corresponding to each of a plurality of timings may be stored in advance in the output device M. Then, when the output control unit 13 causes the output device M to output timing arrival information, it may transmit timing identification information that identifies the detected timing to the output device M. The output device M may read out the contents of the timing arrival information corresponding to the received timing identification information, and output the timing arrival information based on the read out contents.

[0182] Other configurations of the processing apparatus 10 of the fifth embodiment are similar to those of the processing apparatus 10 of the first to fourth embodiments.

[0183] According to the processing apparatus 10 of the fifth embodiment, the same effects as those of the processing apparatus 10 of the first to fourth embodiments are realized.

[0184] Furthermore, the processing device 10 of the fifth embodiment can notify a person of the arrival of the characteristic timing described above. With such a processing device 10, as described above, a person can take appropriate action at an appropriate time based on the notification.

[0185] <<Modification>> The output control unit 13 may control the projection device to project information indicating a movement trajectory suitable for biometric authentication while walking onto the ground. An example is shown in FIG. 10. In the figure, information N indicating a movement trajectory suitable for biometric authentication is １ and N ２ is projected onto the ground. ２ The arrow indicates a movement trajectory suitable for biometric authentication. １ The message encourages people to move along the path.

[0186] When the configuration of the fourth embodiment is adopted, as described above, the calculation unit 12 identifies a group corresponding to the movement trajectory of a person detected in an image. The determination unit 13 may project the representative movement trajectory of the identified group on a projection device as shown in Fig. 10. An image of the representative movement trajectory is generated in advance for each group and stored in a storage device accessible from the projection device.

[0187] According to this modification, it is possible to notify a person of a movement trajectory suitable for walk-through biometric authentication. By having a person move along a movement trajectory suitable for walk-through biometric authentication, it is possible to improve the accuracy of biometric authentication.

[0188] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0189] In addition, in the flowcharts used in the above description, multiple steps (processes) are described in order. However, the order of the steps performed in each embodiment is not limited to the order described. In each embodiment, the order of the steps shown in the drawings can be changed as long as it does not cause any problems in terms of the content.

[0190] Some or all of the above embodiments may be described as, but are not limited to, the following supplementary notes: 1. A processing device comprising: a detection means for tracking a person detected from an image; a calculation means for calculating a first timing at which the size of the person's face in the image reaches a predetermined size suitable for biometric authentication based on a change in the size of the person's face in the image over time; an output control means for causing an output device to output first information that attracts the person's attention at a second timing that is a predetermined time before the first timing; and an authentication means for performing biometric authentication based on an image generated after the second timing. 2. The processing device described in 1, wherein the output control means executes processing to notify a person of the arrival of at least one of the following timings: a timing at which the size of the person's face in the image falls within a predetermined range suitable for biometric authentication; a timing at which the size of the person's face in the image falls outside the predetermined range after falling within the predetermined range; a timing at which the size of the person's face in the image falls outside the predetermined range without successful biometric authentication; a timing at which the person's face in the image leaves the frame; and a timing at which the person's face in the image leaves the frame without successful biometric authentication. 3. The processing device according to 1 or 2, wherein the output control means causes the output device to continue outputting the first information from the second timing to an output end timing, and the output end timing is a timing when a biometric authentication result is obtained, a timing when biometric authentication is successful, a timing when the size of a person's face in an image falls within a predetermined range suitable for biometric authentication and then falls outside the predetermined range, or a timing when the person's face in an image is framed out.4. The processing device according to any one of 1 to 3, wherein the authentication means, before performing biometric authentication, determines whether a human face in an image satisfies predetermined quality conditions, and performs biometric authentication using an image in which the human face satisfies the predetermined quality conditions, and the output control means executes processing to notify a person of the arrival of at least one of the following timings: timing when an image that satisfies the predetermined quality conditions is obtained, timing when the size of the human face in the image falls within a predetermined range suitable for biometric authentication and then goes out of the predetermined range without an image satisfying the predetermined quality conditions being obtained, and timing when the human face in the image frames out without an image satisfying the predetermined quality conditions being obtained. 5. The processing device according to any one of 1 to 4, wherein the authentication means, before performing biometric authentication, determines whether a human face in an image satisfies predetermined quality conditions, and performs biometric authentication using an image in which the human face satisfies the predetermined quality conditions, and the output control means causes the output device to continue outputting the first information from the second timing to an output end timing, and the output end timing is the timing when an image that satisfies the predetermined quality conditions is obtained. 6. The processing device according to any one of 1 to 5, wherein the output control means controls a projection device to project information indicating a movement trajectory suitable for biometric authentication while walking onto the ground. 7. The processing device according to any one of 1 to 6, wherein the calculation means calculates the first timing based on a change in size of a person's face in the image over time and the movement trajectory of the person. 8. The processing device according to any one of 1 to 7, wherein the calculation means calculates a period at which a timing will arrive when the degree of blurring of the face satisfies a criterion based on the change in size of the person's face in the image over time and the period, and calculates as the first timing the timing when the size of the person's face in the image becomes a predetermined size suitable for biometric authentication and the degree of blurring of the face satisfies the criterion based on the change in size of the person's face in the image over time and the period.9. A processing method in which one or more computers track a person detected from an image, calculate a first timing at which the size of the person's face in the image will reach a predetermined size suitable for biometric authentication based on the manner in which the size of the person's face in the image changes over time, cause an output device to output first information that attracts the person's attention at a second timing that is a predetermined time before the first timing, and perform biometric authentication based on an image generated after the second timing. 10. A program that causes a computer to function as: detection means for tracking a person detected from an image, calculation means for calculating the first timing at which the size of the person's face in the image will reach a predetermined size suitable for biometric authentication based on the manner in which the size of the person's face in the image changes over time, output control means for causing an output device to output first information that attracts the person's attention at the second timing that is a predetermined time before the first timing, and authentication means for performing biometric authentication based on an image generated after the second timing.

[0191] Some or all of Supplements 2 to 8 that are dependent on the processing device of Supplement 1 described above may also be dependent on the processing method of Supplement 9 and the program of Supplement 10 in the same dependent relationship as Supplement 1 and Supplements 2 to 8. Furthermore, within the scope of each of the above-described embodiments, some or all of the configurations described as Supplements can be realized in various hardware, software, various recording means for recording software, or systems.

[0192] This application claims priority based on Japanese Patent Application No. 2024-046114, filed March 22, 2024, the disclosure of which is incorporated herein by reference in its entirety.

[0193] 10 Processing device 11 Detection unit 12 Calculation unit 13 Output control unit 14 Authentication unit 1A Processor 2A Memory 3A Input / output I / F 4A Peripheral circuit 5A Bus

Claims

1. A processing device having: a detection means for tracking a person detected from an image; a calculation means for calculating a first timing at which the size of a person's face in the image will reach a predetermined size suitable for biometric authentication based on the manner in which the size of the person's face in the image changes over time; an output control means for causing an output device to output first information that attracts a person's attention at a second timing that is a predetermined time before the first timing; and an authentication means for performing biometric authentication based on an image generated after the second timing.

2. The processing device according to claim 1, wherein the output control means executes processing to notify a person of the arrival of at least one of the following timings: the timing when the size of a person's face in the image falls within a predetermined range suitable for biometric authentication; the timing when the size of a person's face in the image falls within the predetermined range but then goes outside the predetermined range; the timing when the size of a person's face in the image falls within the predetermined range but then goes outside the predetermined range without biometric authentication being successful; the timing when a person's face in the image falls out of the frame; and the timing when a person's face in the image falls out of the frame without biometric authentication being successful.

3. The processing device according to claim 1 or 2, wherein the output control means causes the output device to continue outputting the first information from the second timing to an output end timing, and the output end timing is the timing when the biometric authentication result is obtained, the timing when biometric authentication is successful, the timing when the size of the person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside the predetermined range, or the timing when the person's face in the image falls out of the frame.

4. A processing device according to any one of claims 1 to 3, wherein the authentication means determines whether the human face in the image satisfies predetermined quality conditions before performing biometric authentication, and performs biometric authentication using an image in which the human face satisfies the predetermined quality conditions, and the output control means executes processing to notify a person of the arrival of at least one of the following times: the time when an image satisfying the predetermined quality conditions is obtained, the time when the size of the human face in the image falls within a predetermined range suitable for biometric authentication and then goes outside the predetermined range without an image satisfying the predetermined quality conditions being obtained, and the time when the human face in the image goes out of frame without an image satisfying the predetermined quality conditions being obtained.

5. A processing device according to any one of claims 1 to 4, wherein the authentication means determines whether a human face in an image satisfies predetermined quality conditions before performing biometric authentication, and performs biometric authentication using an image in which the human face satisfies the predetermined quality conditions, and the output control means causes the output device to continue outputting the first information from the second timing to an output end timing, and the output end timing is the timing when an image that satisfies the predetermined quality conditions is obtained.

6. A processing device according to any one of claims 1 to 5, wherein the output control means controls a projection device to project onto the ground information indicating a movement trajectory suitable for biometric authentication while walking.

7. A processing device according to any one of claims 1 to 6, wherein the calculation means calculates the first timing based on the manner in which the size of a person's face in the image changes over time and the person's movement trajectory.

8. A processing device according to any one of claims 1 to 7, wherein the calculation means calculates the period at which the degree of blurring of the face satisfies a criterion based on the gait of the person detected in the image, and calculates as the first timing the time at which the size of the person's face in the image reaches a predetermined size suitable for biometric authentication and the degree of blurring of the face satisfies a criterion based on the manner in which the size of the person's face in the image changes over time and the period.

9. A processing method in which one or more computers track a person detected from an image, calculate a first timing at which the size of the person's face in the image will reach a predetermined size suitable for biometric authentication based on the way the size of the person's face changes over time, output first information that attracts the person's attention to an output device at a second timing that is a predetermined time before the first timing, and perform biometric authentication based on an image generated after the second timing.

10. The processing method according to claim 9, wherein the computer executes a process to notify a person of the arrival of at least one of the following timings: the timing when the size of a person's face in the image falls within a predetermined range suitable for biometric authentication; the timing when the size of a person's face in the image falls within the predetermined range but then goes outside the predetermined range; the timing when the size of a person's face in the image falls within the predetermined range but then goes outside the predetermined range without biometric authentication being successful; the timing when a person's face in the image falls out of the frame; and the timing when a person's face in the image falls out of the frame without biometric authentication being successful.

11. The processing method according to claim 9 or 10, wherein the computer causes the output device to continue outputting the first information from the second timing to an output end timing, and the output end timing is the timing when the biometric authentication result is obtained, the timing when biometric authentication is successful, the timing when the size of the person's face in the image falls within a predetermined range suitable for biometric authentication and then goes out of the predetermined range, or the timing when the person's face in the image falls out of the frame.

12. A processing method according to any one of claims 9 to 11, wherein the computer determines whether a human face in an image satisfies predetermined quality conditions before performing biometric authentication, performs biometric authentication using an image in which the human face satisfies the predetermined quality conditions, and notifies a person of the arrival of at least one of the following times: when an image that satisfies the predetermined quality conditions is obtained, when the size of the human face in the image falls within a predetermined range suitable for biometric authentication and then goes outside the predetermined range without an image that satisfies the predetermined quality conditions being obtained, and when the human face in the image goes out of frame without an image that satisfies the predetermined quality conditions being obtained.

13. A processing method according to any one of claims 9 to 12, wherein the computer determines whether a human face in an image satisfies predetermined quality conditions before performing biometric authentication, performs biometric authentication using an image in which the human face satisfies the predetermined quality conditions, and causes the output device to continue outputting the first information from the second timing to an output end timing, the output end timing being the timing at which an image that satisfies the predetermined quality conditions is obtained.

14. A processing method according to any one of claims 9 to 13, wherein the computer controls a projection device to project onto the ground information indicating a movement trajectory suitable for biometric authentication while walking.

15. A recording medium having recorded thereon a program that causes a computer to function as: a detection means for tracking a person detected from an image; a calculation means for calculating a first timing at which the size of a person's face in an image will reach a predetermined size suitable for biometric authentication based on the manner in which the size of the person's face in the image changes over time; an output control means for causing an output device to output first information that attracts a person's attention at a second timing that is a predetermined time before the first timing; and an authentication means for performing biometric authentication based on an image generated after the second timing.

16. The recording medium according to claim 15, wherein the output control means executes a process of notifying a person of the arrival of at least one of the following timings: the timing when the size of a person's face in the image falls within a predetermined range suitable for biometric authentication; the timing when the size of a person's face in the image falls outside the predetermined range after falling within the predetermined range; the timing when the size of a person's face in the image falls outside the predetermined range after falling within the predetermined range without biometric authentication being successful; the timing when a person's face in the image falls out of the frame; and the timing when a person's face in the image falls out of the frame without biometric authentication being successful.

17. The recording medium of claim 15 or 16, wherein the output control means causes the output device to continue outputting the first information from the second timing to an output end timing, and the output end timing is the timing when the biometric authentication result is obtained, the timing when biometric authentication is successful, the timing when the size of the person's face in the image falls within a predetermined range suitable for biometric authentication and then falls outside the predetermined range, or the timing when the person's face in the image falls out of the frame.

18. A recording medium according to any one of claims 15 to 17, wherein the authentication means determines whether the human face in the image satisfies predetermined quality conditions before performing biometric authentication, and performs biometric authentication using an image in which the human face satisfies the predetermined quality conditions, and the output control means executes a process to notify a person of the arrival of at least one of the following times: the time when an image satisfying the predetermined quality conditions is obtained, the time when the size of the human face in the image falls within a predetermined range suitable for biometric authentication and then goes outside the predetermined range without an image satisfying the predetermined quality conditions being obtained, and the time when the human face in the image frames out without an image satisfying the predetermined quality conditions being obtained.

19. A recording medium described in any one of claims 15 to 18, wherein the authentication means determines whether the human face in the image satisfies predetermined quality conditions before performing biometric authentication, and performs biometric authentication using an image in which the human face satisfies the predetermined quality conditions, and the output control means causes the output device to continue outputting the first information from the second timing to an output end timing, and the output end timing is the timing when an image that satisfies the predetermined quality conditions is obtained.

20. A recording medium according to any one of claims 15 to 19, wherein the output control means controls a projection device to project onto the ground information indicating a movement trajectory suitable for biometric authentication while walking.

Images