Identity verification system, identity verification method, and program

Abstract

The personal authentication system comprises a camera (260) and an information processing device (200) that performs personal authentication based on images of a subject captured by the camera (260). The information processing device (200) includes a first processing unit (510) that determines the movement of the subject's face based on first captured images (251) of the subject, which are continuously captured by the camera (260) as the subject changes the direction of their face from the front to a direction different from the front, and performs facial authentication of the subject based on the first captured images (251) and a registered image (256), and a second processing unit (520) that performs facial authentication of the subject based on a second captured image (252) of the subject captured by the camera (260) and a registered image (256).

Description

Technical Field

[0001] This invention relates to a personal authentication system, a personal authentication method, and a program.

Background Art

[0002] Japanese Patent Application Laid-Open No. 2022-62533 (Patent Document 1) discloses an NC operation panel having a face authentication function for an operator and a function for managing the body temperature of the operator.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] As disclosed in the above Patent Document 1, in various scenarios, authentication (personal authentication) of whether a target person is the genuine person is performed using face authentication. In this case, it is required that the personal authentication of the target person be performed more accurately.

[0005] For example, in a scenario where maintenance of a machine tool is performed, face authentication is performed on the operator who performs the maintenance of the machine tool, thereby executing the personal authentication of the operator. It is required that the personal authentication of the operator be accurately performed so that the maintenance of the machine tool is performed by a person having appropriate knowledge and skills.

[0006] An object of this invention is to provide a personal authentication system, a personal authentication method, and a program capable of more accurately performing authentication of whether a target person is the genuine person.

Means for Solving the Problems

[0007] ​The personal authentication system according to this invention comprises a camera for photographing a subject and an information processing device that performs authentication of the subject's identity based on the image of the subject captured by the camera. The information processing device includes a first processing unit that determines the movement of the subject's face based on a first image of the subject, taken continuously by the camera, in which the subject changes the orientation of their face from the front to a direction different from the front, and performs facial authentication of the subject based on the first image and a pre-registered image of the subject, and a second processing unit that performs facial authentication of the subject based on a second image of the subject taken by the camera and the registered image.

[0008] The personal authentication method according to this invention is a personal authentication method that performs authentication of the subject's identity based on a captured image of the subject. The personal authentication method comprises the steps of: obtaining a first captured image by continuously photographing a subject from the front and changing the direction of their face to a direction different from the front; determining the movement of the subject's face based on the first captured image and performing facial authentication of the subject based on the first captured image and a pre-registered image of the subject; obtaining a second captured image by photographing the subject; and performing facial authentication of the subject based on the second captured image and the registered image.

[0009] The program according to this invention is a program for an information processing device that performs authentication of the identity of a subject based on an image of the subject taken by the device. The program causes the information processing device to perform the following steps: determine the movement of the subject's face based on a first image of the subject, in which the subject changes the orientation of their face from the front to a direction different from the front; perform facial authentication of the subject based on the first image and a pre-registered image of the subject; and perform facial authentication of the subject based on a second image of the subject and the registered image. [Effects of the Invention]

[0010] According to this invention, it is possible to provide an identity verification system, an identity verification method, and a program that can more accurately verify that a subject is who they claim to be. [Brief explanation of the drawing]

[0011] [Figure 1] This diagram schematically illustrates the relationship between the user authentication system and the machine tool in an embodiment of this invention. [Figure 2] This is a functional block diagram showing the worker's PC and the facial recognition server during identity theft detection. [Figure 3] This diagram shows the display unit when impersonation is detected. [Figure 4] This is a functional block diagram showing the worker's PC and the facial recognition server used during identity verification. [Figure 5] This diagram shows the display unit used during identity verification. [Figure 6] This is a flowchart showing the steps for detecting impersonation. [Figure 7] This is a flowchart showing the steps for identity verification. [Modes for carrying out the invention]

[0012] Embodiments of this invention will be described with reference to the drawings. In the drawings referred to below, the same or equivalent components are given the same numbers.

[0013] Figure 1 is a schematic diagram illustrating the relationship between the user authentication system and the machine tool in an embodiment of this invention. Referring to Figure 1, the worker performing maintenance on the machine tool 600 carries a worker PC 200. Once the worker's identity is authenticated, the worker can communicate maintenance information regarding various states of the machine tool 600 between the worker PC 200 and the control panel 610 of the machine tool 600. In this embodiment, we will describe the case in which the user authentication system, user authentication method, and program of the present invention are applied to the user authentication of a worker (corresponding to the target person in this invention) performing maintenance on the machine tool 600.

[0014] The user authentication system 100 in this embodiment includes a worker PC 200, a camera 260, a user authentication management server 300, and a face matching server 400. The worker PC 200 (camera 260), the user authentication management server 300, and the face matching server 400 communicate with each other via a network such as a wireless LAN.

[0015] The worker PC200 corresponds to the information processing device in the present invention. Each component of the worker PC200 is realized by hardware including arithmetic units such as a CPU (Central Processing Unit) and various computer processors, storage devices such as memory and storage, and wired or wireless communication lines connecting them, and software stored in the storage device that supplies processing instructions to the arithmetic units. The computer program may consist of device drivers, an operating system, various application programs located at a higher layer, or libraries that provide common functions to these programs.

[0016] The worker PC200 performs authentication (identity verification) to confirm that the worker is who they claim to be, based on the image of the worker captured by the camera 260.

[0017] The main functions of the operator PC 200 are spoof detection and operator authentication. Spoof detection is to detect spoofs using photos. Operator authentication is to confirm that the operator is the genuine person using face authentication. The operator PC 200 finally executes the authentication that the operator is the genuine person through the steps of spoof detection and operator authentication.

[0018] The operator PC 200 is a mobile computer that can be carried by the operator. The operator PC 200 has a display unit 210 and an operation unit 220. The display unit 210 is composed of a liquid crystal display, an organic EL display, or other display devices. The display unit 210 can display the captured image of the operator taken by the camera 260, etc. The operation unit 220 accepts various operations by the operator.

[0019] The camera 260 is configured to be able to photograph the operator. The camera 260 is provided on the operator PC 200. The camera 260 is configured to be able to photograph the face of the operator who operates the operation unit 220 of the operator PC 200. In the present invention, the camera may be provided separately from the information processing device.

[0020] The personal authentication management server 300 is installed by the administrator of the personal authentication system 100. The personal authentication management server 300 is installed, for example, in the company building of the manufacturer of the machine tool 600.

[0021] The main function of the personal authentication management server 300 is the management of information related to operator authentication. The personal authentication management server 300 registers the face photo of the operator, verifies the face photos of the operator before and after registration, monitors the subsequent calculated personal reliability during the personal authentication process, and stores various information associated with the operator's personal authentication in the database.

[0022] More specifically, the user authentication management server 300 stores registered worker images 256. The registered worker images 256 are registered in the database of the user authentication management server 300 by the administrator of the user authentication system 100, who registers the worker's facial photograph provided by the worker. The user authentication management server 300 stores multiple registered images 256 for each worker. These multiple registered images 256 include registered images where the worker's face is facing forward. The multiple registered images 256 may further include registered images where the worker's face is facing in a direction other than forward.

[0023] The user authentication management server 300 transmits the registered worker images 256 stored in its database to the face matching server 400. The face matching server 400 extracts facial feature quantities from the registered images 256, which are numerical representations of features such as the position of the eyes, the shape of the nose, and the size of the mouth. The face matching server 400 stores the facial feature quantities of the worker extracted from the registered images 256.

[0024] The user authentication management server 300 also stores threshold information. This threshold information includes a threshold t for the face angle used for impersonation detection, and a first threshold Ta and a second threshold Tb for the level of trust in the user, used for impersonation detection and user verification, respectively.

[0025] When an operator operates the control unit 220, the operator PC 200 receives the operator's user ID and password. If the received password matches the password registered with the user ID, the operator PC 200 sends the user ID to the personal authentication management server 300. The personal authentication management server 300 sends personal information 561 corresponding to the user ID to the operator PC 200. The personal information 561 includes at least the threshold information described above.

[0026] Figure 2 is a functional block diagram showing the worker's PC and face matching server during impersonation detection. Figure 3 is a diagram showing the display unit during impersonation detection.

[0027] Referring to Figures 1 to 3, the operator PC200 has a first processing unit 510. The first processing unit 510 performs processing for detecting scams.

[0028] The first processing unit 510 determines the movement of the worker's face based on the first captured images 251 of the worker, which are continuously captured by the camera 260 and show the worker changing the direction of their face from the front to a direction different from the front. It also performs face recognition of the worker based on the first captured images 251 and the registered images 256 of the worker that have been registered in advance.

[0029] More specifically, the first processing unit 510 includes a display control unit 531. The display control unit 531 controls the display unit 210. When impersonation is detected, the display control unit 531 controls the display unit 210 so that the first display screen 210A is displayed.

[0030] As shown in Figure 3, the first display screen 210A has a first instruction image section 216A. The first instruction image section 216A displays instructions for the worker when impersonation is detected. The first instruction image section 216A includes a first comment section 218A and a first illustration section 217A. The first comment section 218A displays a comment instructing the worker to change the direction of their face from facing forward to a direction other than facing forward. The first comment section 218A displays the comment, "Please slowly turn in the following direction: Right (Please keep it there for a few seconds)." The first illustration section 217A displays an illustration instructing the worker to change the direction of their face from facing forward to a direction other than facing forward. The first illustration section 217A displays an illustration of a face instructing the worker to turn to the right, along with an arrow.

[0031] In the first indicator image section 216A, one of the following directions—left, right, up, or down—is randomly indicated as a direction different from the front. The direction different from the front indicated in the first indicator image section 216A is not particularly limited and may be, for example, a diagonal direction to the up, down, left, or right directions.

[0032] The first display screen 210A further includes a cautionary comment section 212. The cautionary comment section 212 displays precautions for workers when impersonation is detected. For example, the cautionary comment section 212 displays precautions such as, "Please remove the following: masks, sunglasses, and safety goggles," "Avoid bright lighting (including backlighting)," and "Do not cover your eyes or mouth."

[0033] As shown in Figure 2, the first processing unit 510 further includes a camera control unit 532. The camera control unit 532 controls the camera 260. When impersonation is detected, the camera control unit 532 controls the camera 260 to continuously photograph the worker as they change the orientation of their face from the front to a direction different from the front. The shooting interval set by the camera control unit 532 is, for example, 500 msec / frame. The first captured image 251 of the worker as they change the orientation of their face from the front to a direction different from the front is obtained by the shooting by the camera 260.

[0034] The operator PC 200 further includes a transmission unit 230. The transmission unit 230 transmits the first captured image 251 to the face matching server 400 via the network.

[0035] As shown in Figure 3, the first display screen 210A further includes a first captured image section 211A. The first captured image section 211A continuously displays the first captured image 251 captured by the camera 260.

[0036] As shown in Figures 1 and 2, the face matching server 400 is installed by a provider of cloud services. The main functions of the face matching server 400 are to calculate the angle of the worker's face based on the captured image and to calculate the confidence level of the person by comparing the facial features extracted from the captured image with the facial features extracted from the registered image.

[0037] The face matching server 400 has an angle calculation unit 410. When impersonation is detected, the angle calculation unit 410 acquires a first captured image 251 from the worker PC 200. The angle calculation unit 410 calculates the angle of the worker's face in the first captured image 251 (with the angle of the face when facing a direction other than facing forward as the reference 0°).

[0038] For example, the angle calculation unit 410 recognizes the contour of the face and feature points such as eyebrows, eyes, nose, and mouth, and calculates the angle of the worker's face in the first captured image 251 by verifying the change in the position of the feature points relative to the recognized contour.

[0039] The face matching server 400 further comprises a confidence calculation unit 420 and a feature quantity storage unit 430. The feature quantity storage unit 430 stores the facial features of the worker extracted from the registered image 256 from the personal authentication management server 300.

[0040] The confidence calculation unit 420, upon detecting impersonation, acquires a first captured image 251 from the worker PC 200 and reads the worker's facial features from the feature memory unit 430. The confidence calculation unit 420 compares the facial features extracted from the first captured image 251 of the worker, which is facing a direction other than the front, with the facial features read from the feature memory unit 430 to calculate the confidence level (identity confidence) that the worker's face in the first captured image 251 is the face of the worker in the registered image 256. When the identity confidence level is 1, the probability that the worker in the first captured image 251 is the worker in the registered image 256 is the highest, and when the identity confidence level is 0, the probability that the worker in the first captured image 251 is the worker in the registered image 256 is the lowest.

[0041] More specifically, the confidence calculation unit 420 extracts the worker's facial features from the first captured image 251, performs face matching by comparing the worker's facial features with the facial features read from the feature memory unit 430, and calculates the confidence level of the person (corresponding to the degree of agreement between the facial features extracted from the first captured image 251 and the facial features read from the feature memory unit 430). Face matching is not limited to this; for example, images may be directly compared, or a matching method using machine learning may be applied. Furthermore, instead of using the face matching server 400 on the cloud, for example, the worker's PC 200 may be equipped with the functions performed by the face matching server 400.

[0042] The angle of the worker's face and the confidence level of their identity, calculated by the face recognition server 400, are transmitted to the worker PC 200. The worker PC 200 further includes a receiving unit 240. The receiving unit 240 receives the angle of the worker's face and the confidence level of their identity from the face recognition server 400.

[0043] The worker PC 200 further has a storage unit 560. The storage unit 560 stores personal information 561 from the personal authentication management server 300. The storage unit 560 further stores a program that causes the worker PC 200 to perform personal authentication of the worker.

[0044] The first processing unit 510 further includes a motion determination unit 533. The motion determination unit 533 obtains the angle of the worker's face from the face matching server 400 and reads a threshold value t for the face angle from the storage unit 560. The motion determination unit 533 compares the angle of the worker's face from the face matching server 400 with the threshold value t for the face angle, and determines that the worker has moved if the face angle is greater than or equal to the threshold value t. For example, the threshold value t is 30°.

[0045] The first processing unit 510 further includes a first facial recognition determination unit 534. When the motion determination unit 533 determines that there is movement of the worker, the first facial recognition determination unit 534 obtains the person confidence level from the face matching server 400 and reads a first threshold Ta of the person confidence level at the time of impersonation detection from the storage unit 560. The first facial recognition determination unit 534 compares the person confidence level from the face matching server 400 with the first threshold Ta of the person confidence level at the time of impersonation detection, and affirms the face recognition of the worker if the person confidence level is equal to or greater than the first threshold Ta.

[0046] When the first facial recognition determination unit 534 confirms the identification of the worker's face, the impersonation detection step is completed and the process moves on to the identity verification step.

[0047] Figure 4 is a functional block diagram showing the worker's PC and facial recognition server during identity verification. Figure 5 is a diagram showing the display unit during identity verification.

[0048] Referring to Figures 4 and 5, the worker PC 200 has a second processing unit 520. The second processing unit 520 performs processing for verifying the identity of the worker.

[0049] The second processing unit 520 performs facial recognition of the worker based on facial features extracted from the second captured image 252 of the worker taken by the camera 260 and the registered image 256.

[0050] More specifically, the second processing unit 520 includes a display control unit 531. The display control unit 531 controls the display unit 210 so that the second display screen 210B is displayed when verifying the identity of the worker.

[0051] As shown in Figure 5, the second display screen 210B has a second instruction image section 216B and a caution comment section 212. The second instruction image section 216B displays instructions for the worker during identity verification. The second instruction image section 216B includes a second comment section 218B and a second illustration section 217B. The second comment section 218B displays a comment instructing the worker to return their face to facing forward. The second comment section 218B displays the comment, "Please slowly turn in the following direction: Forward (move within the range where both eyes can see)." The second illustration section 217B displays an illustration instructing the worker to return their face to facing forward. The second illustration section 217B displays an illustration of a face facing forward. The caution comment section 212 is the same as the caution comment section 212 in the first display screen 210A.

[0052] As shown in Figure 4, the second processing unit 520 further includes a camera control unit 532. The camera control unit 532 controls the camera 260 to photograph the worker when verifying the worker's identity. The camera 260 captures a second image 252 of the worker facing forward.

[0053] The transmission unit 230 transmits the second captured image 252 to the face matching server 400 via the network.

[0054] As shown in Figure 5, the second display screen 210B further includes a second captured image section 211B. The second captured image section 211B displays the second captured image 252 captured by the camera 260.

[0055] As shown in Figure 4, the confidence calculation unit 420 acquires the second captured image 252 from the worker's PC 200 when verifying the worker's identity, and reads the worker's facial features extracted from the registered image 256 from the feature memory unit 430. The confidence calculation unit 420 compares the facial features extracted from the second captured image 252 with the facial features read from the feature memory unit 430 to calculate the confidence level (identity confidence) that the worker's face in the second captured image 252 is the face of the worker in the registered image 256.

[0056] The confidence level of the person calculated by the face recognition server 400 is transmitted to the worker's PC 200. The receiving unit 240 receives the confidence level of the person from the face recognition server 400.

[0057] The second processing unit 520 further includes a second facial recognition determination unit 535. The second facial recognition determination unit 535 obtains the confidence level of the person from the face matching server 400 and reads a second threshold Tb of the confidence level of the person at the time of identity verification from the storage unit 560. The second facial recognition determination unit 535 compares the confidence level of the person from the face matching server 400 with the second threshold Tb of the confidence level of the person at the time of identity verification, and affirms the facial recognition of the worker if the confidence level of the person is equal to or greater than the second threshold Tb.

[0058] The second threshold Tb for identity verification is greater than the first threshold Ta for identity detection. For example, the first threshold Ta is 0.5 and the second threshold Tb is 0.8.

[0059] The second facial recognition determination unit 535 confirms the recognition of the worker's face, completing the identity verification step. The worker PC 200 then performs final authentication to confirm that the worker is indeed the person in question. This enables communication of maintenance information for the machine tool 600 between the worker PC 200 and the control panel 610 of the machine tool 600.

[0060] Figure 6 is a flowchart showing the steps for detecting impersonation. Referring to Figures 1 to 3 and Figure 6, the operator PC 200 receives the ID code and password (S101).

[0061] In this step, if the received password matches the password registered with the user ID, the worker PC 200 sends the user ID to the personal authentication management server 300. The personal authentication management server 300 sends the personal information 561 corresponding to the user ID to the worker PC 200.

[0062] Next, the worker PC 200 acquires personal information 561 (S102). In this step, the storage unit 560 stores the personal information 561 from the personal authentication management server 300.

[0063] Next, the operator PC 200 (display control unit 531) displays the first display screen 210A on the display unit 210. In this step, the operator receives an instruction displayed on the first instruction image unit 216A and changes the direction of their face from facing forward to a direction different from facing forward.

[0064] Next, the worker PC 200 (camera control unit 532) photographs the worker as they change the orientation of their face from facing forward to a direction different from facing forward. The worker PC 200 (transmitter 230) transmits the obtained first captured image 251 to the face matching server 400 (S104).

[0065] The face matching server 400 (angle calculation unit 410) calculates the angle of the worker's face in the first captured image 251 from the worker's PC 200.

[0066] Next, the worker PC 200 (receiving unit 240) receives the face angle calculated by the face matching server 400 (S105). Next, the worker PC 200 (motion determination unit 533) compares the face angle of the worker calculated by the face matching server 400 with the face angle threshold t stored in the storage unit 560, and determines whether the face angle is greater than or equal to the threshold t (S106).

[0067] If it is determined in step S106 that the angle of the face is less than the threshold t, the process returns to step S104.

[0068] If it is determined in step S106 that the angle of the face is greater than or equal to threshold t, the face matching server 400 (confidence calculation unit 420) calculates the confidence level (identity confidence level) that the worker's face in the first captured image 251 is the face of the worker in the registered image 256.

[0069] Next, the worker PC 200 (receiving unit 240) receives the confidence level calculated by the face matching server 400 (S107). Next, the worker PC 200 (first face recognition determination unit 534) compares the confidence level calculated by the face matching server 400 with the first threshold Ta of the confidence level when impersonation is detected, and determines whether the confidence level is equal to or greater than the first threshold Ta (S108).

[0070] If, in step S108, it is determined that the confidence level of the person is below the first threshold Ta, the operator PC200 determines whether a certain amount of time has elapsed since the start of the impersonation detection step (S109).

[0071] If it is determined that a certain amount of time has not elapsed in step S109, the process returns to step S104. If it is determined that a certain amount of time has elapsed in step S109, the process returns to step S103, and the operator PC200 (display control unit 531) displays a "Not Approved" comment on the first display screen 210A.

[0072] If, in step S108, the trustworthiness of the person is determined to be equal to or greater than the first threshold Ta, the impersonation detection step is completed.

[0073] Figure 7 is a flowchart showing the steps for identity verification. Referring to Figures 1, 4 to 5, and 7, the worker PC 200 (display control unit 531) displays the second display screen 210B on the display unit 210 (S111). In this step, the worker receives the instruction displayed on the second instruction image unit 216B and turns their face from a direction other than facing forward back to facing forward.

[0074] Next, the worker PC 200 (camera control unit 532) takes a picture of the worker. The worker PC 200 (transmitter 230) transmits the obtained second captured image 252 to the face matching server 400 (S112).

[0075] The face matching server 400 (confidence calculation unit 420) calculates the confidence level (identity confidence level) that the worker's face in the second captured image 252 is the face of the worker in the registered image 256.

[0076] Next, the worker PC 200 (receiving unit 240) receives the confidence level of the person calculated by the face matching server 400 (S113). Next, the worker PC 200 (second face recognition determination unit 535) compares the confidence level of the person calculated by the face matching server 400 with the second threshold Tb of the confidence level of the person at the time of identity verification, and determines whether the confidence level of the person is equal to or greater than the second threshold Tb (S114).

[0077] If, in step S114, it is determined that the confidence level of the person is less than the second threshold Tb, the operator PC200 determines whether a certain amount of time has elapsed since the start of the identity verification step (S115).

[0078] If it is determined that a certain amount of time has not elapsed in step S115, the process returns to step S112. If it is determined that a certain amount of time has elapsed in step S115, the process returns to step S111, and the operator PC 200 (display control unit 531) displays a "Not Approved" comment on the second display screen 210B.

[0079] If the confidence level of the person is determined to be equal to or greater than the second threshold Tb in step S114, the identity verification step is completed. The worker PC200 then performs the final authentication to confirm that the worker is indeed the person in question.

[0080] Note that the order of the impersonation detection step and the identity verification step described above can be reversed.

[0081] The configuration of the user authentication system, user authentication method, and program in the embodiments of this invention described above is summarized below.

[0082] The personal authentication system 100 in this embodiment includes a camera 260 for photographing a worker as the subject, and a worker PC 200 as an information processing device that performs authentication of the worker's identity based on the images of the worker captured by the camera 260. The worker PC 200 includes a first processing unit 510 that determines the movement of the worker's face based on first captured images 251 of the worker, which are continuously captured by the camera 260, showing the worker changing the direction of their face from the front to a direction different from the front, and performs facial authentication of the worker based on the first captured images 251 and a pre-registered registered image 256 of the worker, and a second processing unit 520 that performs facial authentication of the worker based on a second captured image 252 of the worker captured by the camera 260 and the registered image 256.

[0083] Furthermore, the personal authentication method in this embodiment is a personal authentication method that performs authentication of the worker's identity based on captured images of the worker as the subject. The personal authentication method comprises the steps of obtaining a first captured image 251 by continuously photographing the worker from the front and changing the direction of their face to a direction different from the front (S104), determining the movement of the worker's face based on the first captured image 251 and performing face authentication of the worker based on the first captured image 251 and a pre-registered image 256 of the worker (S106, S108), obtaining a second captured image 252 by photographing the worker (S112), and performing face authentication of the worker based on the second captured image 252 and the registered image 256 (S114).

[0084] Furthermore, the program in this embodiment is a program for the worker PC 200, which is an information processing device that performs authentication of the worker's identity based on a captured image of the worker as the target. The program causes the worker PC 200 to perform the following steps: determine the movement of the worker's face based on a first captured image 251 of the worker, in which the face is turned from the front to a direction different from the front, and perform face authentication of the worker based on the first captured image 251 and a pre-registered image 256 of the worker (S106, S108); and perform face authentication of the worker based on a second captured image 252 of the worker and the registered image 256 (S114).

[0085] The program may be recorded on a computer-readable storage medium. The computer-readable storage medium may be a non-transitory computer-readable storage medium.

[0086] With this configuration, by determining the movement of the worker's face based on the first captured image 251 of the worker, in which the worker changes the direction of their face from the front to a direction different from the front, it is possible to confirm that the worker who is the subject of the first captured image 251 is not an imposter using a photograph. When imposter detection occurs, by performing facial recognition of the worker based on the first captured image 251 and a pre-registered image 256 of the worker, it is possible to confirm that the action of changing the facial movement described above is being performed by the actual person, not someone else. Furthermore, by performing facial recognition of the worker based on the second captured image 252 of the worker facing forward and the registered image 256, it is possible to confirm that the worker is the actual person. As a result, accurate authentication of the worker's identity can be performed.

[0087] Furthermore, the second threshold Tb used for facial recognition during identity verification is greater than the first threshold Ta used for facial recognition during impersonation detection. With this configuration, when using the first captured image 251 which is facing a direction other than the front for impersonation detection, a lower first threshold Ta is used, while when using the second captured image 252 for identity verification, a higher second threshold Tb is used, thereby enabling more accurate identity verification.

[0088] Furthermore, when impersonation is detected, the first processing unit 510 displays on the display unit 210 the direction in which the worker should change the orientation of their face. With this configuration, the impersonation detection step can be carried out smoothly by prompting the worker to change the orientation of their face through the display on the display unit 210.

[0089] In this embodiment, the personal authentication system, personal authentication method, and program of the present invention were described in the case where they are applied to the personal authentication of an operator performing maintenance on a machine tool 600, but the present invention is not limited to this. The present invention can be applied, for example, to personal authentication when unlocking various doors, or to personal authentication when granting access to a computer.

[0090] The embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of the invention is indicated by the claims rather than the foregoing description, and all modifications within the meaning and scope of equivalents of the claims are intended. [Explanation of symbols]

[0091] 100 User authentication system, 200 Worker PC, 210 Display unit, 210A First display screen, 210B Second display screen, 211A First captured image unit, 211B Second captured image unit, 212 Warning comment unit, 216A First instruction image unit, 216B Second instruction image unit, 217A First illustration unit, 217B Second illustration unit, 218A First comment unit, 218B Second comment unit, 220 Operation unit, 230 Transmission unit, 240 Receiving unit, 251 First captured image, 252 Second captured image, 256 Registered image, 260 Camera, 300 User authentication management server, 400 Face matching server, 410 Angle calculation unit, 420 Confidence calculation unit, 430 Feature quantity storage unit, 510 First processing unit, 520 Second processing unit, 531 Display control unit, 532 Camera control unit, 533 motion detection unit, 534 first facial recognition detection unit, 535 second facial recognition detection unit, 560 storage unit, 561 personal information, 600 machine tool, 610 control panel, 620 recording medium, Ta first threshold, Tb second threshold, t threshold.

Claims

1. An information processing device that performs authentication of whether the subject is the person in question based on an image of the subject, The aforementioned information processing device is A first processing unit determines the movement of the subject's face based on a first image of the subject, which is continuously captured and changes the direction of the subject's face from the front to a direction different from the front, and performs facial recognition of the subject based on the first image and a pre-registered image of the subject. It includes a second processing unit that performs facial recognition of the subject based on a second captured image of the subject and the registered image, The first processing unit affirms facial recognition of the subject if the confidence level that the subject in the first captured image is the same subject in the registered image is equal to or greater than a first threshold. The second processing unit affirms facial recognition of a subject when the confidence that the subject in the second captured image is the same subject in the registered image is greater than or equal to a second threshold which is greater than the first threshold.

2. Further comprising a display unit capable of displaying the subject being photographed, The personal authentication system according to claim 1, wherein the first processing unit causes the display unit to display the direction in which the subject changes the orientation of their face.

3. A method for authenticating an individual by performing authentication of the individual's identity based on a photograph of the individual, The first step is to obtain a first image by continuously photographing a subject from the front, while they change the direction of their face to a direction different from the front. The steps include: determining the facial movements of the subject based on the first captured image, and performing facial recognition of the subject based on the first captured image and a pre-registered image of the subject; The steps include obtaining a second image by photographing the subject, The system includes a step of performing facial recognition of the subject based on the second captured image and the registered image, The step of performing facial recognition of a subject based on the first captured image and the registered image includes a step of affirming facial recognition of the subject if the confidence that the subject in the first captured image is the same subject as in the registered image is equal to or greater than a first threshold, A method for authenticating a person, comprising the step of performing facial recognition of a person based on the second captured image and the registered image, wherein the confidence that the person in the second captured image is the same person as the person in the registered image is greater than or equal to a second threshold which is greater than the first threshold.

4. A program for an information processing device that performs authentication of the subject's identity based on a captured image of the subject, The program is configured on the information processing device. The process involves determining the facial movement of a subject based on a first captured image of the subject as they change their facial orientation from a frontal view to a direction different from the front, and performing facial recognition of the subject based on the first captured image and a pre-registered image of the subject. The system performs a step of facial recognition of the subject based on a second captured image of the subject and the registered image. The step of performing facial recognition of a subject based on the first captured image and the registered image includes a step of affirming facial recognition of the subject if the confidence that the subject in the first captured image is the same subject as in the registered image is equal to or greater than a first threshold, A program that performs facial recognition of a subject based on the second captured image and the registered image, and includes a step of affirming facial recognition of the subject if the confidence that the subject in the second captured image is the same subject in the registered image is greater than or equal to a second threshold which is greater than the first threshold.

Images