JP2022084586A - Entrance control apparatus and entrance control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an entrance control apparatus which easily executes a face authentication of a user, who utilizes a facility, at the time of entrance while safely managing a face image of the user, and an entrance control method.

SOLUTION: In a face authentication system where a face authentication machine, a management terminal, a registration device, a face management server 5 and a face collation server 6 are connected over a network such as the Internet, an entrance control apparatus (face authentication machine 1) comprises: a face image acquisition unit (camera 11) which acquires a face image of a user for authentication by unlocking a lock device (external device 16) of a door and imaging the user before entering a predetermined area; a transmission unit (communication unit 13) which transmits a processing request of a face authentication to the face collation server 6 on the basis of the face image for authentication; a reception unit (communication unit 13) which receives an authentication result returned from the face collation server 6; and a control unit 15 which reflects an unlock signal of the lock device with the authentication device.

SELECTED DRAWING: Figure 6

COPYRIGHT: (C)2022,JPO&INPIT

Description

The present disclosure relates to an entry control device and a room entry control method for controlling the opening and closing of a gate device through which the subject passes, based on the result of face recognition of the subject.

Conventionally, for example, a reading unit that receives an identifier transmitted from a wireless tag (remote type IC card) possessed by a user and a facial feature amount registered corresponding to the identifier received from the wireless tag are acquired. The acquisition unit, the imaging unit that images the user, the extraction unit that extracts facial features from the image data captured by the imaging unit, the facial features extracted by the extraction unit, and the acquisition unit. It is provided with a face matching unit that collates whether or not the acquired facial features match, and an open / close control unit that sets the exit of the gate to an open state or a closed state according to the collation result in the face matching unit. A gate device is known (see Patent Document 1).

International Publication No. WO2018 / 181968

By the way, in the above-mentioned conventional technique, it is possible to narrow down the target person of face authentication (that is, the registered face feature amount) by the identifier transmitted from the user's wireless tag, but the user passing through the gate Since it is necessary to carry an IC card at all times, it becomes difficult to safely manage personal information (that is, information that can identify a specific individual) contained in the IC card. Further, in the above-mentioned conventional technique, no consideration is given to a mechanism in which both face image data and its feature amount data are placed on a cloud server or the like and the data is safely managed.

Further, in the above-mentioned prior art, the gate device executes a process of collating whether the facial feature amount extracted by the extraction unit matches the facial feature amount acquired by the acquisition unit, but it is used in a facility. No consideration is given to the mechanism for simplifying the process related to face recognition when a person enters the room.

Therefore, the present disclosure provides an entry control device and an entry control method capable of easily performing face recognition at the time of entering a room while safely managing the face image of the user who uses the facility. The main purpose.

The room entry control device of the present disclosure unlocks a lock device such as a door configured as an external system of the facility to the face authentication server that executes face authentication of the user of the facility, and enters the predetermined area. It is an entry control device that requests the processing of the face authentication when the user enters the room and controls the unlocking of the lock device based on the authentication result, and photographs the user before entering the predetermined area. , A face image acquisition unit that acquires a face image for authentication of the user, a transmission unit that transmits a processing request for face authentication to the face authentication server based on the face image for authentication, and the face. A receiving unit that receives an authentication result returned from an authentication server and a control signal transmitting unit that transmits the authentication result as a control signal to the external system are provided, and a processor of the external system receives the control signal. The configuration is such that it is reflected in the unlock signal of the lock device.

Further, the room entry control method of the present disclosure unlocks a lock device such as a door configured as an external system of the facility for the face authentication server that executes face authentication of the user of the facility, and enters the predetermined area. This is an entry control method in which the face authentication processing request is made when the user enters the room, and the unlocking of the lock device is controlled based on the authentication result. A photograph is taken, a face image for authentication of the user is acquired, a processing request for the face authentication is transmitted to the face authentication server based on the face image for authentication, and the face authentication server returns. The authentication result is received and the authentication result is transmitted to the external system as a control signal, and the processor of the external system receives the control signal and reflects it in the unlock signal of the lock device.

According to the present disclosure, it is possible to easily perform face authentication when the user enters the room while safely managing the face image of the user who uses the facility.

Overall configuration diagram of the face recognition system according to this embodiment Explanatory diagram showing an example of an event that occurs in a facility Explanatory diagram showing the outline of the collation group Explanatory diagram showing the outline of the inquiry of face matching performed by the face recognition machine 1. Explanatory diagram showing the outline of the face matching unit Block diagram showing the schematic configuration of the face recognition machine 1 Block diagram showing the schematic configuration of the management terminal 2 A block diagram showing a schematic configuration of the registration device 4. A block diagram showing a schematic configuration of the face management server 5. Block diagram showing the schematic configuration of the face matching server 6 Block diagram showing a schematic configuration of an external device Explanatory diagram showing the outline of the face image extraction process Explanatory diagram showing the outline of the inquiry of face recognition by the collation group information Explanatory drawing showing an example of the registration screen displayed on the registration device 4. Sequence diagram showing the procedure of registration process for users A sequence diagram showing a first example of face recognition processing for a user A sequence diagram showing a second example of face recognition processing for a user Explanatory diagram showing an example of the authentication screen displayed on the face recognition machine 1. A sequence diagram showing a third example of face recognition processing for a user Explanatory diagram showing the login screen displayed on the management terminal 2. The screen displayed on the management terminal 2 when registering, referencing, updating, and deleting the association between the face recognition machine 1 and the matching group. The screen displayed on the management terminal 2 when registering, referencing, updating, and deleting the association between the face recognition machine 1 and the matching group. Explanatory diagram showing an example of a user's management screen displayed on the management terminal 2. Explanatory diagram showing an example of the conference room search screen displayed on the management terminal 2. Explanatory diagram showing an example of the attendee display screen displayed on the management terminal 2. Explanatory diagram showing an example of the personal connection search screen displayed on the management terminal 2.

The first invention made to solve the above-mentioned problems unlocks a lock device such as a door configured as an external system of the facility for a face recognition server that executes face recognition of a user of the facility. This is an entry control device that requests the face recognition processing when the user enters the predetermined area and controls the unlocking of the lock device based on the authentication result, and before entering the predetermined area. A face image acquisition unit that photographs the user and acquires a face image for authentication of the user, and a face recognition processing request is transmitted to the face recognition server based on the face image for authentication. The processor of the external system includes a transmission unit for receiving the authentication result returned from the face recognition server, a control signal transmission unit for transmitting the authentication result as a control signal to the external system, and the processor of the external system. The control signal is received and reflected in the lock release signal of the lock device.

According to this, the face authentication of the user is executed by the face authentication server, and the authentication result is reflected in the unlock signal of the door lock device, so that the face image of the user who uses the facility is safely managed. , Face recognition when the user enters the room can be easily executed.

Further, the second invention is configured to include a power feeding unit that receives power supply from the gate device.

According to this, it becomes possible to receive electric power with a simple configuration. In addition, if the gate device is connected to an uninterruptible power supply (or emergency power supply), the gate device can operate normally even if a power failure occurs in the facility, and the security of the facility can be achieved. Can be maintained.

Further, the third invention further includes a storage unit for storing the face image for authentication, and the face image for authentication stored in the storage unit is deleted after the face authentication process is completed. It is configured.

According to this, even when the gate open / close control device is attached to the gate device used by an unspecified person, it is possible to prevent the face image from being stored in the gate open / close control device for a long period of time, and to store the user's face image. It becomes possible to manage more safely.

Further, the fourth aspect of the present invention is to unlock a lock device such as a door configured as an external system of the facility for a face recognition server that executes face recognition of a facility user and enter a predetermined area. It is a room entry control method that requests the processing of the face recognition when the user enters the room and controls the unlocking of the lock device based on the authentication result, and photographs the user before entering the predetermined area. , The face image for authentication of the user is acquired, the processing request for the face recognition is transmitted to the face recognition server based on the face image for authentication, and the authentication result returned from the face recognition server. Is received and the authentication result is transmitted to the external system as a control signal, and the processor of the external system receives the control signal and reflects it in the unlock signal of the lock device.

According to this, the face authentication of the user is executed by the face authentication server, and the authentication result is reflected in the unlock signal of the door lock device, so that the face image of the user who uses the facility is safely managed. , Face recognition when the user enters the room can be easily executed.

Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

FIG. 1 is an overall configuration diagram of a face recognition system according to the present embodiment.

This face recognition system includes a face recognition machine 1, a management terminal 2, a face recognition server 3 (cloud server), and a registration device 4 (face recognition registration device). The face recognition server 3 includes a face management server 5 (face recognition management server) and a face matching server 6.

The face recognition machine 1, the management terminal 2, the registration device 4, the face management server 5, and the face matching server 6 are connected to each other via a network such as the Internet. The face recognition system is built for each provider (business operator) of the face recognition service. A plurality of face recognition machines 1 are installed at each place where face recognition is required, such as the entrance / exit of a building or the entrance / exit of a room. The registration device 4 is arranged at a reception counter or the like that accepts visitors to the facility. A required number of face matching servers 6 are provided according to the number of face recognition machines 1 and the like.

The face recognition machine 1 includes a camera 11, and the camera 11 acquires a photographed image of a user (a visitor of a facility, a worker at a company, a store, or the like in the facility). Further, the face recognition machine 1 has a display 12, which displays the face recognition result acquired from the face matching server 6 and notifies the user of the face recognition result.

The management terminal 2 is operated by the administrator of the face recognition system, and is composed of a PC, and a management application that manages the operations of the face recognition machine 1, the registration device 4, the face management server 5, and the face matching server 6 is installed. Has been done. This management application allows the administrator to perform various management tasks. The management application is implemented as a Web application. As for the administrator, a plurality of administrator groups having different access authority for each information in the face recognition system are set.

The registration device 4 acquires information used for processing the user's face recognition (hereinafter referred to as authentication information). The authentication information may include not only the face image for registration of each user but also specific information other than the face image used for identifying the user.

Specific information includes information displayed on the user's business card, information displayed on the payment card (for example, credit card or cash card) possessed by the user, and the place of stay of the user in the facility. Information and the like can be used. The specific information may include information that can permanently identify the user, such as the name of the user. However, the specific information may be, for example, the user number given to the user of the facility, the number of the conference room (including any space where interviews are possible) used by the user in the facility, the accommodation facility, etc. Information that can temporarily identify the user may be used, such as the number of the room in which the user stays.

The face management server 5 centrally manages user information (including personal information). Specifically, the face management server 5 stores and stores face image data for registration and specific information for each user in association with each other, and manages them. The face management server 5 acquires a face image for registration and specific information from the registration device 4 at the time of user registration. Further, the face management server 5 may acquire at least a part of the information included in the face image for registration and the specific information from the management terminal 2 (that is, the administrator).

At the time of face authentication, the face matching server 6 acquires face image data for authentication of the target person of face authentication from the face authentication machine 1, and generates face feature amount data of the target person from the face image data. , Whether or not the target person is a registrant by performing face matching by comparing the face feature amount data of the target person with the face feature amount data of the registrant (registered user) stored in the own device. Perform face authentication to determine whether or not.

Further, the face matching server 6 acquires face image data for user registration from the face management server 5 (or registration device 4) at the time of user registration prior to face recognition, and the face image data. The user's facial feature amount data is generated from the data and stored in the own device. In some cases, the face matching server 6 may acquire the photographed image of the user from the management terminal 2 (that is, the administrator) and acquire the face image data from the photographed image.

In the present embodiment, the face management server 5 (face image management unit) and the face matching server 6 (face image matching unit) are assumed to exist in physically different information processing devices, but a single one. It may exist in the information processing apparatus of.

Further, in the present embodiment, the management terminal 2 and the face management server 5 are provided, but the management terminal 2 and the face management server 5 can be configured by a single information processing device. For example, by installing a management application on the face management server 5, the face management server 5 can also serve as the management terminal 2. In addition, an administrator (for example, a worker at a company or store in a facility) whose access authority to information of the face authentication system is limited can use the face management server 5 from his / her own information processing terminal (PC, tablet, etc.). It is possible to access information that is less necessary to be kept secret than the personal information of the user (for example, information about the meeting attended by the user, information about the personal network of the user, etc., which will be described later). ..

Further, in the present embodiment, the face feature amount matching is performed, but the face matching is not limited to the face feature amount matching, and a matching method to which machine learning or the like is applied may be adopted. Further, the present embodiment can be applied to biometric authentication other than face authentication.

Next, an example of an event generated by the action of the user of the facility to which the face recognition system is applied will be described. FIG. 2 is an explanatory diagram showing an example of an event occurring in a facility.

As shown in FIG. 2, the facility user 10 performs an operation of registration as a facility user (that is, registration of information used for face matching processing) by the registration device 4 arranged at the reception. Can be done. For registered users who have revisited the facility within a predetermined period, registration by the registration device 4 can be omitted.

At the facility, face recognition is performed for each of various events (see inside the two-dot chain line in FIG. 2) generated by the actions of a person (including a registered user) in the facility.

For example, the registered user 10 can pass through the security gate by receiving face recognition in front of the security gate that manages the passage of the user 10 in the facility. A face recognition machine 1 that cooperates with the gate device 16 (that is, controls the open / closed state of the door portion) can be installed in the security gate.

Further, for example, the registered user 10 can enter the conference room by receiving face recognition in front of the conference room in the facility. In front of the conference room, a face recognition machine 1 that cooperates with a lock device that locks the door of the conference room (that is, controls the locking or unlocking of the door) can be arranged. In an accommodation facility such as a hotel, a face recognition machine 1 that cooperates with a lock device for the door of the room in which the user stays can be installed instead of the conference room.

Further, for example, the registered user 10 can automatically settle the usage fee of the restaurant by receiving face recognition in the restaurant in the facility. A camera 11 capable of photographing the face of the user 10 is installed in the restaurant, and the face recognition machine 1 cooperates with a payment terminal device (or a payment system) used for payment of a usage fee at the restaurant.

Further, for example, the registered user 10 can automatically settle the usage fee of the gym by receiving face recognition at the gym in the facility. A camera 11 capable of photographing a user's face is installed in the gym, and the face recognition machine 1 cooperates with a payment terminal device (or a payment system) used for payment of a usage fee at the gym.

In addition, the event that occurs in the facility is not limited to the above example, for example, the settlement of the usage fee in the parking lot of the facility, the settlement of the usage fee of the laundry service in the facility, and the usage fee of the bathhouse in the facility. Various events such as settlement of payments may be included.

Further, for example, when an employee of a company that provides a service in the facility is registered as a user, the face recognition machine 1 is placed at the entrance / exit of the backyard (for example, a warehouse, a workshop, etc.) of the facility. It is possible to ensure security in the backyard.

Next, the collation group will be described. FIG. 3 is an explanatory diagram showing an outline of a collation group.

In the present embodiment, the users are grouped according to the installation location of the face recognition machine 1 to which the user receives face authentication, and a collation group is set for each user. Further, similarly to the user, the face recognition machine 1 is divided into groups according to the installation location of the face recognition machine 1, and a collation group is set for each face recognition machine 1. Further, the face matching server 6 corresponding to the face recognition machine 1, that is, the face matching server 6 that receives a face matching inquiry (request) from the face recognition machine 1, is grouped in the same manner as the face recognition machine 1, and each face is divided into groups. A collation group is set in the collation server 6.

For example, in the face recognition machine 1 installed at the entrance of a multi-tenant building in which a plurality of companies are occupying, users belonging to the plurality of companies receive face recognition. On the other hand, the information of users who belong to different organizations cannot be managed in the same database. Therefore, in such a case, a collation group for each of a plurality of companies is formed, and the user registration information is stored in the collation group unit. In the face recognition machine 1, the face verification server 6 for the collation group for each company is used. Inquire about face matching.

The user can be authenticated only by the face recognition machine 1 corresponding to the collation group to which he / she belongs. Therefore, even a registered user cannot be authenticated by the face recognition machine 1 that does not belong to his / her collation group, for example, the face recognition machine 1 of a building that he / she is not allowed to enter. For example, if a user working at one business office goes to another business office, the face recognition machine 1 at that business office cannot be authenticated.

Next, an inquiry for face matching from the face recognition machine 1 to the face matching server 6 will be described. FIG. 4 is an explanatory diagram showing an outline of an inquiry for face matching performed by the face recognition machine 1.

The face recognition machine 1 makes a face matching inquiry to the face matching server 6 belonging to the same matching group as the own device, and makes a face matching request (face recognition processing) to the face matching server 6 which is the inquiry destination (request destination). Request) is sent. Further, the face authentication machine 1 holds the network address (IP address) of the face matching server 6 as the matching group information, and based on this network address, the face matching server 6 corresponding to the own device is face-matched. Make an inquiry.

The inquiry method for face recognition includes a simultaneous inquiry shown in FIG. 4A, a random inquiry shown in FIG. 4B, and a sequential inquiry (inquiry destination switching) shown in FIG. 4C.

As shown in FIG. 4A, in the case of a simultaneous inquiry, the face recognition machine 1 simultaneously makes a face matching inquiry to the face matching server 6 corresponding to each of the plurality of matching groups. For example, when the face recognition machine 1 is installed at the entrance of a multi-tenant building and users of a plurality of matching groups receive face recognition, the face recognition machine 1 simultaneously applies to the face matching server 6 of each group. Send a face matching request.

In order to reduce the load on one face matching server 6, it is also possible to divide and store the face feature amount data of the user belonging to one matching group into a plurality of face matching servers 6. Also in this case, the face recognition machine 1 simultaneously inquires about face matching to all face matching servers 6 having different registered contents in the same matching group as itself.

As shown in FIG. 4B, in the case of a random query, the face recognition machine 1 selects the face matching server 6 to be the contact for face matching from among the plurality of face matching servers 6 in the same matching group as itself. Select randomly. The registered contents of the plurality of face matching servers 6 are the same. That is, the same user is targeted for collation, and the facial feature amount data of the same user is stored. In such an inquiry method, since the inquiry from the face authentication machine 1 is distributed to each face matching server 6, the load of each face matching server 6 can be reduced and the load can be distributed.

As shown in FIG. 4C, in the case of sequential inquiry, the face recognition machine 1 sequentially selects a face matching inquiry destination from a plurality of face matching servers 6 in the same matching group as itself. Specifically, when an order (priority) is given to the face matching server 6, the face recognition machine 1 selects the face matching server 6 according to the order, and there is no response from the high-ranking face matching server 6. To switch the destination to the face matching server 6. That is, the face recognition machine 1 selects the first-ranked face matching server 6 in the matching group, and requests the face recognition server 6 for face recognition. Then, when there is no response from the face matching server 6, the face recognition machine 1 selects another face matching server 6 having the next rank in the same matching group, and performs face recognition to the face matching server 6. Request. Redundancy (backup) can be achieved by such control.

Next, the face matching unit will be described. FIG. 5 is an explanatory diagram showing an outline of the face matching unit.

The face matching server 6 may be provided with a plurality of face matching units for performing face matching. This face matching unit has a face matching process that executes a face matching program, and a face matching database in which information (face feature amount data) of users to be compared with face matching is registered.

This face matching unit is provided for each matching group, and performs face matching in response to a face matching request from the face recognition machine 1 belonging to the matching group. Therefore, one face matching server 6 can support a plurality of matching groups.

Here, the face authentication machine 1 holds the network address (IP address) of the face matching server 6 as the matching group information, and based on this network address, the face matching server 6 corresponding to the own device is face-matched. Can be inquired about. Further, in the present embodiment, the face authentication machine 1 holds a network address (IP address) for each face matching unit, and based on this network address, inquires about the face matching unit corresponding to its own device for face matching. It can be performed.

Further, since the face collation database is provided for each collation group, the face feature amount data of the user can be stored in the collation group unit. The face matching process executes a matching process between the face feature amount data stored in the group unit and the face feature amount data generated from the face image data acquired from the face authentication machine 1 of the same group.

Here, there are various aspects of the correspondence between the face recognition machine 1 and the face matching unit.

In the example shown in FIG. 5A, the face matching server 6 is provided with one face matching unit of the matching group to which the face recognition machine 1 belongs, and the face recognition machine 1 and the face matching server 6 are one-to-one. It corresponds with.

In the example shown in FIG. 5B, the face recognition machine 1 and the face matching server 6 have a one-to-one correspondence, but the face matching server 6 has two face matching units of the matching group to which the face recognition machine 1 belongs. It is provided.
In this case, in the case of the face recognition machine 1 installed at the entrance of a multi-tenant building, the information of users (for example, user A and user B) to which the organization belongs is managed by different collation databases. Can be done.

In the example shown in FIG. 5C, the face recognition machine 1 and the face matching server 6 have a one-to-two correspondence, and the face recognition machine 1 inquires about face matching to the two face matching servers 6 (user). Collation request). In this case, the load distribution of the face matching server 6 is provided by providing the same face matching unit, that is, two or more face matching units that perform face matching using the face feature amount data of the same user, in different face matching servers 6. It is possible to deal with life and death.

Next, the schematic configuration of the face recognition machine 1 will be described. FIG. 6 is a block diagram showing a schematic configuration of the face recognition machine 1.

The face recognition machine 1 includes a camera 11 (face image acquisition unit), a display 12 (display unit), a communication unit 13 (transmission unit, reception unit), a storage unit 14, a control unit 15, and a control signal transmission unit. 17 and.

The camera 11 constantly shoots a predetermined shooting area, and when a person enters the shooting area, the person is shot, and a shot image of a person subject to face recognition (including a registered user) can be acquired. In order to save power, a motion sensor (not shown) may be provided to detect the arrival of a person and activate the camera 11.

The display 12 displays the progress of face recognition and the face recognition result, and the person who receives the face recognition can confirm them. A speaker may be provided as an output unit for outputting the authentication result of the target person, and the face authentication result may be notified by voice. Further, the display 12 can be omitted depending on the installation location of the face recognition machine 1 (for example, when the face recognition machine 1 is installed at the security gate).

The communication unit 13 communicates with the face matching server 6 via the network. In the present embodiment, the face image data is transmitted to the face matching server 6. In addition, the user's authentication result is received from the face matching server 6. Further, the communication unit 13 communicates with the face management server 5 via the network. In the present embodiment, the collation group information and the like are received from the face management server 5.

The storage unit 14 stores data of a photographed image (face image) of a user, collation group information, a control program executed by a processor constituting the control unit 15, and the like.

The control unit 15 includes an activation processing unit 21, a face image extraction unit 22, a face matching request unit 23, an authentication result notification unit 24, an authentication result adjustment unit 25 (history information generation unit), and an operation status monitoring unit 26. And have. The control unit 15 is composed of a processor, and each unit of the control unit 15 is realized by executing a program stored in the storage unit 14 by the processor.

When the face recognition machine 1 is started, the activation processing unit 21 acquires collation group information from the face management server 5 and stores it in the storage unit 14. The acquisition of the collation group information is not limited to this, and the collation group information may be distributed from the face management server 5 to the target face recognition machine 1 along with the update of the collation group information.

The face image extraction unit 22 acquires a photographed image of a person subject to face authentication from the camera 11 (camera image capture), detects a person's face from the photographed image (face detection), and determines the size of the detected face. It is possible to determine whether it is appropriate (face size check), cut out a face area from the captured image (face cutout), and acquire data of the target person's face image (face image for authentication). The face image data may be only the image of the face area, but the data of the photographed image (the image of the predetermined image) and the position information (face frame information) of the face area on the data of the photographed image. It may be a combination of.

The face matching request unit 23 transmits a face matching request by the communication unit 13 to the face matching server 6 whose own device and the matching group match.

The authentication result notification unit 24 notifies the user of the face authentication result by displaying the face authentication result acquired from the face verification server 6 by the communication unit 13 on the display 12.

The authentication result adjustment unit 25 controls to cooperate with the external device 16 based on the user's face matching result acquired from the face matching server 6. Examples of the external device 16 include a gate device (see FIG. 2) that manages the passage of users in the facility, a door lock device that manages entry and exit of a conference room or a predetermined area in the facility, and a user in the facility. It is possible to use a payment terminal device for making payments for services received in the above or purchased items. Further, when the authentication result adjustment unit 25 cooperates with the lock device for the door of the conference room, it can generate information on the entry history associated with a plurality of users who have entered the conference room.

The face recognition machine 1 and the external device 16 are directly connected by a communication cable or the like, or are communicably connected via a known network. The authentication result adjustment unit 25 can notify the user's face matching result to the server of the external system including the external device 16 by being configured by the connection application.

The operation status monitoring unit 26 monitors the operation status of the own device and notifies the face management server 5 of the operation status of the own device.

The control signal transmission unit 17 transmits a control signal for controlling the operation to the external device 16.

Next, the management terminal 2 will be described. FIG. 7 is a block diagram showing a schematic configuration of the management terminal 2.

The management terminal 2 includes a communication unit 31, a display 32 (display unit), an input device 33 (operation unit), a storage unit 34, and a control unit 35.

The communication unit 31 communicates with the face management server 5 via the network. In the present embodiment, screen information and the like are received from the face management server 5, and the administrator's operation information and the like corresponding to the screen information and the like are transmitted to the face management server 5.

The display 32 displays various screens. The input device 33 is a mouse, a keyboard, or the like, and operates the screen displayed on the display 32.

The storage unit 34 stores a program (management application) or the like executed by the processor constituting the control unit 35.

The control unit 35 includes a GUI control unit 38. The control unit 35 is composed of a processor, and each unit of the control unit 35 is realized by executing a program (management application) stored in the storage unit 34 by the processor.

The GUI control unit 38 displays various operation screens distributed from the face management server 5 on the display 32. Further, according to the input operation of the administrator using the input device 33, the input information is acquired and the screen is controlled. In the present embodiment, the GUI control unit 38 controls the display and input of a screen related to login, specifically, a login screen. In addition, the GUI control unit 38 controls the display and input of screens related to user management, specifically, screens related to user information registration (individual registration, batch registration), reference, update, and deletion. conduct. In addition, the GUI control unit 38 performs display input control regarding screens related to collation group management, specifically, screens related to registration (individual registration, batch registration), reference, update, and deletion related to collation groups. In addition, the GUI control unit 38 controls display and input regarding screens related to authentication machine management, specifically, screens related to registration, reference, update, and deletion regarding the association between the authentication machine and the collation group. In addition, the GUI control unit 38 controls the display and input regarding the reference screen of the authentication log (history information of face authentication).

Next, the schematic configuration of the registration device 4 will be described. FIG. 8 is a block diagram showing a schematic configuration of the registration device 4.

The registration device 4 includes a camera 41A for face photography, a camera 41B (personal information acquisition unit) for information acquisition, a display 42 (display unit), a communication unit 43, a storage unit 44, and a control unit 45. I have.

The face photographing camera 41A can acquire a photographed image of the user, for example, by photographing the face of the user who has stopped by the reception desk of the facility.

The camera 41B for information acquisition can acquire a photographed image of a business card by photographing the business card presented by the user. However, the camera 41B for acquiring information is not limited to a business card, but includes the user's belongings (for example, the user's payment card and the user's accommodation information) on which information that can identify the user is displayed. It is also possible to take a picture of the written card, etc.) and acquire the taken image. The registration device 4 has a mounting table (mounting surface) 116 (FIG. 14 (FIG. 14)) on which a user can easily place a business card or the like in the shooting area of the camera 41B in order to reliably shoot the camera 41B for information acquisition. C)) can be provided.

In the registration device 4, one camera may be used to configure the cameras 41A and 41B for face photographing and information acquisition. Further, the shooting timing of the face shooting camera 41A and the information acquisition camera 41B can be determined by the user's operation (for example, pressing the shooting button) in the registration device 4. Alternatively, the receptionist at the facility may decide the shooting timing.

Further, the registration device 4 can be provided with an information reading device (for example, an RFID reader) that can read the user's specific information by wireless communication from an information recording medium (for example, RFID tag) possessed by the user. be.

The display 42 can display the authentication procedure (including the user's personal information) registration procedure (see FIGS. 14 (A)-(F)) to the user. Further, by attaching a speaker to the display 42, the registration procedure or the like may be guided to the user by voice. Further, by configuring the display 42 with a touch panel, it is possible to acquire the information input by the user by touch operation.

The communication unit 43 communicates with the face management server 5 via the network. In the present embodiment, the face image data and the specific information are transmitted to the face management server 5.

The storage unit 44 stores data of a photographed image (face image) of the user, specific information of the user, a control program executed by a processor constituting the control unit 45, and the like.

The control unit 45 includes a face image extraction unit 46, a specific information acquisition unit 47, a registration request unit 48, a GUI control unit 49, and an operation status monitoring unit 50. The control unit 45 is composed of a processor, and each unit of the control unit 45 is realized by executing a program stored in the storage unit 44 by the processor.

The face image extraction unit 46 acquires a photographed image of a person subject to face authentication from the camera 41A for face photography, detects a person's face from the photographed image, and determines whether or not the user's face image is appropriately acquired. Can be determined. If the acquired face image is not appropriate, the face image extraction unit 46 can display a message prompting the user to take a picture again on the display 42. The face image extraction unit 46 may have functions of camera image capture, face detection, face size check, and face cutout, as in the face image extraction unit 22 of the face authentication machine 1 described above.

The specific information acquisition unit 47 has an OCR (optical character recognition) function, and generates specific information (here, business card description information) from characters and symbols extracted from an image taken by the camera 41B for information acquisition.

The registration request unit 48 transmits a user registration request to the face management server 5 by the communication unit 43.

The GUI control unit 49 displays various guidance screens for the user distributed from the face management server 5 on the display 42. Further, the input information is acquired and the screen is controlled according to the input operation of the user using the input function of the display 42 (here, the touch panel).

The operation status monitoring unit 50 monitors the operation status of the own device and notifies the face management server 5 of the operation status of the own device.

Next, the schematic configuration of the face management server 5 will be described. FIG. 9 is a block diagram showing a schematic configuration of the face management server 5.

The face management server 5 includes a communication unit 51, a storage unit 52, and a control unit 53.

The communication unit 51 communicates with the management terminal 2 via the network. Further, the communication unit 51 communicates with the face recognition device 1 via the network. Further, the communication unit 51 communicates with the registration device 4. Further, the communication unit 51 communicates with the face matching server 6.

The storage unit 52 includes a face information database, a database related to association information between the face recognition machine 1 and the matching group, a database related to the association information between the face matching server 6 and the matching group, an administrator access log, and a control unit 53. Stores control programs executed by the constituent processors.

In the face information database, face images for registration are accumulated as information about each registered user. Further, in the face information database, specific information of each user is stored in association with the face image for registration. In addition, a collation group or the like is registered in the face information database. The user's face image and specific information may be stored in an encrypted state for privacy protection. Further, the face information database may store a user's face image and a part of specific information in a state of being replaced with anonymized information.

The control unit 53 includes an administrator access management unit 61, a user management unit 62, a collation group management unit 63, a device management unit 64, a face collation server management unit 65, a face information management unit 66, and a database management. It includes a unit 67, an operation status monitoring unit 68, an authentication log presentation unit 69, an encryption unit 70, an anonymization information generation unit 80, and an image generation unit 90. The control unit 53 is composed of a processor, and each unit of the control unit 53 is realized by executing a program stored in the storage unit 52 by the processor. Each unit of the control unit 53 is configured as a Web API (Web Application Programming Interface).

The administrator access management unit 61 permits or denies the administrator's access (login) according to the access authority of the administrator who accesses the own device from the management terminal 2. The administrator access management unit 61 monitors the status of access (login) from the management terminal 2 to the face management server 5 and the face matching server 6, and when the access from the management terminal 2 is detected, the information regarding the access (accessed). Record the administrator (administrator, date and time, etc.) as an administrator access log (history information). As a result, the face management server 5 provides the administrator access log to the management terminal 2 in response to the request from the management terminal 2 to refer to the administrator access log, and the administrator can view the administrator access log. can.

Further, the administrator access management unit 61 manages access to the own device from the management terminal 2, and when the administrator operates the management terminal 2 to access the own device, the information at that time is stored in the administrator access log (history). Information) is stored in the storage unit 52. Further, the administrator access log is presented to the management terminal 2 in response to the reference request from the management terminal 2.

The user management unit 62 manages information about the user such as a face image for registration and specific information, and performs necessary processing regarding the user in response to a request from the management terminal 2. In the present embodiment, the management terminal 2 makes a request for registration, reference, update, and deletion regarding the user, and the user management unit 62 performs necessary processing in response to the request. In addition, the user management unit 62 manages information on the meetings attended by the users and information on the personal connections of the users.

The collation group management unit 63 manages information about the collation group, and performs necessary processing regarding the collation group in response to a request from the management terminal 2. In the present embodiment, the management terminal 2 makes a request for registration, reference, update, and deletion regarding the collation group, and performs necessary processing in response to the request. Further, the matching group information, that is, the information necessary for the face recognition machine 1 to request face recognition from the face matching server 6 that matches the own matching group is generated for each face recognition machine 1, and the matching group is generated. Information is provided to the face recognition machine 1.

The device management unit 64 manages information about the face recognition machine 1 and the registration device 4, and performs necessary processing regarding the face recognition machine 1 and the registration device 4 in response to a request from the management terminal 2. In the present embodiment, the management terminal 2 requests registration, reference, update, and deletion regarding the association between the face recognition machine 1 and the collation group, and performs necessary processing in response to the request.

The face matching server management unit 65 manages information about the face matching server 6 and performs necessary processing related to the face matching server 6 in response to a request from the management terminal 2. In the present embodiment, the management terminal 2 requests registration, reference, and deletion regarding the association between the face matching server 6 and the matching group, and performs necessary processing in response to the request.

The face information management unit 66 includes the user's face information (face image, specific information, etc.) stored in the own device and the user's face information (user's face feature amount data) stored in the face matching server 6. Face information is synchronized so that and is maintained in a consistent state. Further, the face information management unit 66 copies the face information (user's face feature amount data).

The database management unit 67 manages the database provided in the own device, and backs up and restores the database.

The operation status monitoring unit 68 monitors the operation status of its own device, receives notification of the operation status from the face recognition machine 1, the registration device 4, and the face verification server 6, and operates the administrator on the management terminal 2. Accordingly, the operating status of the face recognition machine 1, the registration device 4, the own device (face management server 5), and the face matching server 6 is displayed on the screen of the management terminal 2.

The authentication log presentation unit 69 acquires the authentication log from the face matching server 6 in response to the reference request from the management terminal 2, and presents the authentication log to the management terminal 2.

The encryption unit 70 can perform encryption processing on at least a part of the user's face image and specific information. The encrypted information is stored in the storage unit 52. In the face management server 5, the original information (information before the encryption process) corresponding to the information encrypted by the encryption unit 70 may be deleted from the storage unit 52.

The anonymization information generation unit 80 can perform a process of anonymizing at least a part of the user's face image and specific information. The anonymization information generation unit 80 can replace, for example, a part of the user's face image and specific information with dummy information as an anonymization process. Further, in the anonymization process, dummy information may be used for information missing in the user's face image and specific information (for example, information refused to be provided by the user). The anonymized information is stored in the storage unit 52. The encryption unit 70 may encrypt only the original information (information before the anonymization process) corresponding to the information anonymized by the anonymization information generation unit 80.

The image generation unit 90 generates a management screen for the administrator to confirm or change the information management status by the user management unit 62 and the device management unit 64. The management screen is distributed to the management terminal 2 and the like.

Next, the schematic configuration of the face matching server 6 will be described. FIG. 10 is a block diagram showing a schematic configuration of the face matching server 6.

The face matching server 6 includes a communication unit 71, a storage unit 72, and a control unit 73.

The communication unit 71 communicates with the face recognition device 1 via the network. In the present embodiment, face image data and the like are received from the face recognition machine 1. In addition, the user's authentication result and the like are transmitted to the face recognition machine 1. Further, the communication unit 71 communicates with the face management server 5 via the network. In the present embodiment, requests for various processes and the like are received from the face management server 5, and responses and the like corresponding to the requests are transmitted to the face management server 5.

The storage unit 72 stores the face collation database, the association information between the user and the feature amount, the information about the collation group of the own device, the authentication log, the control program executed by the processor constituting the control unit 73, and the like.

In the face matching database, user's facial feature amount data and the like are registered as information about each registered user. Further, the face collation database is provided for each collation group, and the face feature amount data of the user is stored in each group. Here, even if the registered user's face feature amount data is backed up and saved in a non-volatile memory such as an HDD or SSD provided in the face matching server 6 in case of unexpected disappearance from the face matching database. good.

The control unit 73 includes a collation group management unit 81, an image quality check unit 82, a face image extraction unit 83, a face feature amount generation unit 84, a face feature amount management unit 85, and a face feature amount collation unit 86. It includes an authentication log management unit 87, a database management unit 88, and an operation status monitoring unit 89. The control unit 73 is composed of a processor, and each unit of the control unit 73 is realized by executing a program stored in the storage unit 72 by the processor. Each part of the control unit 73 is configured as WebAPI.

The collation group management unit 81 manages the collation group to which the own device belongs, and performs registration and deletion processing related to the collation group in response to a request from the face management server 5.

The image quality check unit 82 determines whether or not the image of the face region in the captured image satisfies a predetermined quality. Specifically, the presence / absence of wearing a mask and the presence / absence of sunglasses are detected from the target image, and the face recognition conformity (evaluation value based on the direction of the face and facial expression) is calculated.

The face image extraction unit 83 extracts a face image from the photographed image of the user acquired by the registration device 4 at the time of user registration. At this time, if necessary, the face image extraction unit 83 detects a person's face from the captured image (face detection), determines whether the detected face size is appropriate (face size check), and from the captured image. A face area is cut out (face cutout), and a face image of a person is acquired. The face image extraction unit 83 can also extract a face image from the captured image of the target person acquired by the face recognition machine 1 at the time of face recognition.

The face feature amount generation unit 84 detects face feature points from the data of the face image for registration and the face image for authentication at the time of user registration and face authentication, and generates face feature amount data, respectively.

The face feature amount management unit 85 registers the user's face feature amount data generated by the face feature amount generation unit 84 at the time of user registration in the face collation database corresponding to the user collation group. Further, at the time of updating or deleting, the face feature amount data registered in the face matching database is deleted in response to a request from the face management server 5. In addition, when the version of the program related to the face feature amount generation and the face matching algorithm is upgraded, the face feature amount data registered in the face matching database is newly added in response to the request from the face management server 5. Update to correspond to the program.

The face feature amount collation unit 86 has the face feature amount data of the target person generated from the face image data for authentication acquired from the face recognition machine 1 at the time of face recognition, and the registrant (registered) stored in the own device. It is determined whether or not the target person is a registrant by comparing with the facial feature amount data of the user). The face feature amount collation unit 86 corresponds to a collation group associated with the own device, and when a plurality of collation groups are associated with the own device, a plurality of face feature amount collation for each collation group is performed. It has a part 86. The face feature amount matching unit 86 (face matching process) constitutes a face matching unit in combination with the face matching database.

The face feature amount collation unit 86 calculates the degree of similarity (collation score) between the subject and the registrant. By comparing this similarity with a predetermined threshold value, the success or failure of face recognition can be determined. In addition to notifying the face recognition machine 1 of the success / failure determination result as a collation result, the face recognition machine 1 may be notified of the person ID and the similarity of the registrants having a high degree of similarity as the collation result.

When the face authentication machine 1 makes a face authentication request to its own device, the authentication log management unit 87 obtains information at that time, that is, the device ID of the face authentication machine 1 from which the face authentication is requested, the face verification result, and the like. The information is stored in the storage unit 72 as an authentication log. Further, the authentication log is provided to the face management server 5 in response to the request from the face management server 5.

The database management unit 88 manages the database provided in the own device, and backs up and restores the database.

The operation status monitoring unit 89 monitors the operation status of the own device and notifies the face management server 5 of the operation status of the own device.

In the present embodiment, the main functions of the face collation server 6 are face image extraction, face feature amount generation, and face feature amount collation, but these functions are processed separately from each other. It can be configured with a device. For example, the face image extraction function may be configured by another information processing device independent of other functions such as face feature amount generation and face feature amount matching.

Next, the schematic configuration of the external device 16 will be described. FIG. 11 is a block diagram showing a schematic configuration of the external device 16.

The external device 16 includes a control signal receiving unit 91, a driving unit 92, a power supply unit 93, and a control unit 94.

The control signal receiving unit 91 receives a control signal (operation command from the face authentication machine 1 linked with the external device 16) transmitted from the control signal transmitting unit 17 of the face authentication machine 1.

The drive unit 92 is controlled by the control unit 94 based on the control signal from the face recognition machine 1. For example, when the external device 16 is a gate device, the drive unit 92 supplies power for opening (or closing) the door portion of the gate device. When the external device 16 is a door lock device, the drive unit 92 supplies power for locking (or unlocking) the lock device. In the case of a device (for example, a payment terminal device) that does not have a portion for driving the external device 16, the drive unit 92 can be omitted.

The power supply unit 93 supplies electric power to each unit of the external device 16. Further, the power supply unit 93 is electrically connected to the power supply unit 97 of the face recognition machine 1, and can supply electric power to the power supply unit 97 as well. The power feeding unit 97 supplies electric power to each unit of the face recognition machine 1. Further , by connecting to the power feeding unit 97 of the face recognition machine 1 via a LAN cable, power can be supplied to the power feeding unit 97 by PoE power supply.

Further, the power supply unit 93 can receive power supply from an uninterruptible power supply (or emergency power supply) independent of the power supply line in the entire facility. As a result, for example, the gate device as the external device 16 can perform normal operation even when a power failure occurs in the facility, and has an advantage that the security of the facility can be maintained.

The control unit 94 controls the operation of each unit of the external device 16. Further, the control unit 94 can execute the necessary processing as the external device 16. The control unit 94 is configured by a processor, and is realized by executing a program stored in a storage unit (memory) (not shown) by the processor.

Next, the face image extraction process will be described. FIG. 12 is an explanatory diagram showing an outline of the face image extraction process.

As described above, the face image extraction unit 22 of the face authentication machine 1 generates face image data from the photographed image data of the user by the face image extraction process, that is, each process of face detection, face size check, and face cutout. The face image extraction process is also performed in the face image extraction unit 83 of the face matching server 6.

At the time of user registration, the data of the photographed image of the user is sent from the registration device 4 to the face matching server 6 via the face management server 5, so that the face image extraction process of the face matching server 6 performs the face image extraction process. conduct. In some cases, the data of the photographed image of the user may be sent from the management terminal 2 to the face matching server 6 via the face management server 5.

On the other hand, at the time of face recognition, it is preferable that the face image extraction process is performed only by the face recognition machine 1 and the face image extraction process is not performed by the face matching server 6. In this case, the face recognition machine 1 is provided with a highly accurate face detection function. Further, the face authentication machine 1 constantly photographs the shooting area with the camera 11 and sends face image information (captured image data and face frame information) to the face matching server 6 at the timing when the face is detected. As a result, the load of the face image extraction process is distributed to the plurality of face recognition machines 1, and the load of the face matching server 6 can be reduced. Moreover, since the amount of communication can be reduced, the load on the network can be reduced. Since the face recognition response can be performed at high speed, the face recognition of the target person appearing one after another can be efficiently performed.

That is, by sharing the user's face authentication process between the face authentication machine 1 and the face verification server 6, an expensive authentication machine that concentrates all the user's face authentication process on the face authentication machine 1 as in the conventional case. It is not necessary to provide a large number. Further, even if the face feature amount data is updated, the maintenance work of the large-scale authentication machine is not required, and the work on the face matching server 6 is sufficient. Therefore, according to the present embodiment, it is possible to construct a face recognition system having an inexpensive configuration and excellent workability.

However, the face recognition machine 1 may not be provided with the face image extraction function, that is, the face recognition machine 1 may not be provided with the face image extraction unit 22.

FIG. 13 is an explanatory diagram showing an outline of an inquiry for face recognition using collation group information.

The face management server 5 holds information about the matching group to which the user belongs, the association information between the face recognition machine 1 and the matching group, and the association information between the face matching server 6 and the matching group. Further, in the matching group management unit 63 of the face management server 5, the face recognition machine is based on the association information between the face recognition machine 1 and the matching group and the association information between the face recognition server 6 and the matching group. Generates collation group information for each 1.

This matching group information is information required for the face recognition machine 1 to request face recognition from the face matching server 6 that matches the matching group of its own. The collation group information includes the identification information (group number) of the collation group to which the face recognition machine 1 belongs and the face recognition server 6 that requests face recognition from the face recognition machine 1, that is, the collation group with the face recognition machine 1. The destination information of the face matching server 6 corresponding to the above is included. Here, the destination information is specifically the network address (for example, IP address) of the face matching server 6, and the face matching server 6 that is the request destination for face authentication is specified by the destination information, and the face authentication machine is specified. 1 is associated with the face matching server 6 to which the face authentication is requested. When the face recognition machine 1 belongs to a plurality of collation groups, the address (IP address) of each of the corresponding plurality of face collation servers 6 is included in the collation group information.

In the face recognition machine 1, the activation processing unit 21 acquires collation group information from the face management server 5 as operation setting information and stores it in its own device at the time of activation or the like. When the face recognition machine 1 detects the face of a person, it sends a face matching request to the face matching server 6 that matches the matching group to which the own device belongs. This face matching request includes information on the matching group to which the own device belongs.

In the face recognition machine 1, the timing of acquiring the setting information related to the collation group information and the like from the face management server 5 may be at a predetermined timing or at a predetermined interval in addition to the start time, or may be performed periodically. , It is also possible to adopt a configuration in which the collation group information is distributed from the face management server 5.

The face matching server 6 holds the association information between the face matching process of the own device and the matching group. When the face recognition server 6 receives the face recognition request from the face recognition machine 1, the face recognition is performed based on the association information between the face recognition group and the face recognition process and the face recognition group obtained from the face recognition machine 1. The face recognition process corresponding to the machine 1 is specified, and the face recognition process is made to perform face matching. As a result, the face matching process is performed in the face recognition process corresponding to the designated matching group.

By the way, in the present embodiment, in the face matching server 6, a face matching database is provided for each matching group, and face feature amount data is stored for each matching group. Therefore, at the time of face collation, face collation may be performed by targeting the face feature amount data registered in the face collation database corresponding to the collation group of the face recognition machine 1. On the other hand, in the face collation server 6, the collation may be performed regardless of the collation group of the face recognition machine 1, and then the collation result may be filtered. That is, after performing face matching on the face feature amount data of all users stored in the face matching server 6, only the matching result with the user belonging to the matching group of the face recognition machine 1 is extracted. You may do so. In this case, the face matching database may not be provided separately for each matching group.

Next, the registration operation by the user in the registration device 4 will be described. FIG. 14 is an explanatory diagram showing an example of a registration screen displayed on the registration device 4.

The user can perform a registration operation for authentication by the registration device 4 when starting to use the facility (or when the facility is not revisited within a predetermined period after registration). In the present embodiment, the user can perform the registration operation according to the guidance screen displayed on the display 42 (here, the touch panel) of the registration device 4. In addition, a part of the registration operation for authentication (such as pressing a button) described below may be performed by a receptionist who has obtained the consent of the user.

The display 42 of the registration device 4 displays an initial screen on which the registration start button 101 is displayed (see FIG. 14A). Then, when the user presses (touches) the registration start button 101, an explanation regarding the handling of personal information in the face recognition system and a screen requesting the user for the consent are displayed (see FIG. 14 (B)).

By pressing the consent button 102 on the screen of FIG. 14B, the user can consent to the handling of personal information in the face recognition system. As a result, the registration device 4 starts acquiring the authentication information of the user. Therefore, the display 42 displays a screen (first screen) showing a method of acquiring specific information used for identifying the user (see FIG. 14C). The user can cancel the registration operation by pressing the cancel button 103. The screen for asking the user for consent may not require consent depending on the user and can be omitted. Further, the consent may be replaced by the operation of the shooting button 104, the shooting button 107, and the registration button 111, which will be described later. Further, consent may be obtained by a predetermined procedure before using the registration device 4.

In the example shown in FIG. 14C, a screen (first) showing a method of photographing a user's business card (here, a guide for placing the user's business card in the photographing area of the information acquisition camera 41B). Screen) is displayed. Therefore, when the user places the business card on the mounting table 116 of the registration device 4 and presses the shooting button 104, the business card is shot by the registration device 4. The user can cancel the shooting of the business card by pressing the stop button 105.

After the shooting of the business card is completed, the registration device 4 is subsequently subjected to a method of shooting a face image for registration of the user (guidance for positioning the user's face in the shooting area of the face shooting camera 41A). Is displayed (see FIG. 14 (D)). At this time, the registration device 4 can prompt the user to align the face by blinking the shooting frame 106 corresponding to the shooting area. A live image (real-time moving image) taken by the face shooting camera 41A is displayed in the shooting frame 106. Therefore, when the user positions his / her face in the shooting frame and presses the shooting button 107, the registration device 4 takes a picture of the user's face. At this time, the registration device 4 can generate a shutter sound when the photographing button 107 is pressed. The user can stop the face shooting by pressing the stop button 108.

When the face shooting is completed, the registration device 4 subsequently displays a screen (first screen) for the user to confirm the shot face image (see FIG. 14E). A captured face image (still image) is displayed on this screen, and the user presses the registration button 111 when it is determined that there is no problem with the captured face image. On the other hand, the user can press the re-shooting button 112 when it is determined that there is a problem with the captured face image. As a result, the screen shown in FIG. 14D is displayed on the registration device 4, and the face can be re-photographed. The user can stop the face shooting by pressing the stop button 113.

When the face photographing is completed, the acquired face image and specific information (here, including the business card image and the business card description information extracted from the business card image by OCR) are transmitted to the face management server 5. In this case, for a user who does not want to provide both the face image and the business card image, at least one of the face image and the business card image may be transmitted to the face management server 5. As a result, the face management server 5 executes the registration process (see FIG. 15) regarding the user. When the registration process of the face management server 5 is normally completed, the registration device 4 displays a screen (second screen) indicating that the registration is completed (“Processing” in FIGS. 14 (F) and 15 (FIG. 15). Display results on screen "). Therefore, when the user presses the end button 115, the registration operation in the registration device 4 ends. The order of acquiring the face image and the business card image may be reversed.

After the registration operation is completed, the registration device 4 can delete the face image and the specific information stored in the storage unit 44. As a result, it is possible to prevent the personal information from being stored in the registration device 4 used by an unspecified person for a long period of time, and it becomes possible to manage the personal information of the user more safely.

Instead of registering the user by the registration device 4 described above, the administrator prepares a file of a face image for user registration in advance, and then registers the user using the management terminal 2. Can be done. Further, the administrator can also modify the authentication information registered by the registration device 4 and add new information from the management terminal 2 in order to supplement the registration by the registration device 4.

Next, the registration process for the user executed in response to the above-mentioned user registration operation will be described. FIG. 15 is a sequence diagram showing a procedure of registration processing for a user.

When the registration operation of the user is performed in the registration device 4 as described above, the face management server 5 starts the registration process for the user as shown in FIG. At this time, the face management server 5 receives the photographed image of the face and specific information from the registration device 4 together with the operation information of the user. In this registration process, first, the face management server 5 sends a request for face image extraction to the face matching server 6 whose matching group matches the target user. This request includes a photographed image of the user's face acquired from the registration device 4. At this time, if there are a plurality of face matching servers 6 having the same matching group as the target user, one face matching server 6 is selected and a request for face image extraction is made to the one face matching server 6. To send.

When the face matching server 6 receives a request for face image extraction from the face management server 5, the face image extraction process is performed. In this face image extraction process, processing such as face detection and face cutting is performed on the photographed image of the user's face acquired from the face management server 5, and the user's face image is extracted. Then, the response of the face image extraction is sent to the face management server 5. This response includes a user's facial image.

When the face management server 5 receives the response of the face image extraction from the face matching server 6, the face management server 5 sends a face registration request to the face matching server 6 in which the target user and the matching group match. This request can include a user's face image acquired from the specific face matching server 6. At this time, if there are a plurality of face matching servers 6 having the same matching group as the target user, all of them including the user's face image acquired from the specific face matching server 6 (2 in FIG. 15). Send a face registration request to the face matching server 6 of (1).

The face management server 5 omits the request for face image extraction to the face matching server 6 described above, and requests the face matching server 6 to register the photographed image of the user's face acquired together with the operation information from the registration device 4. May be added to.

When the face matching server 6 receives the face registration request from the face management server 5, the face registration process is performed. In this face registration process, a face feature amount is generated from a user's face image, and the face feature amount is registered in a database. At this time, a face registration ID is assigned in association with the face feature amount of the user. Then, the response of the face registration is sent to the face management server 5. This response includes a result indicating whether or not the face registration is normally completed and a face registration ID. With the completion of registration of the face feature amount in the database, the user's face image on the face matching server 6 is deleted. Further, the face feature amount data may be backed up and stored in a non-volatile memory such as an HDD or SSD managed separately from the face matching DB of the face matching server 6.

The face management server 5 receives the face registration response from the face matching server 6, and if the process is completed normally, the face management server 5 performs the registration process for the user. In this registration process, the user's face image acquired from the registration device 4 is registered in the face information database. Further, the face management server 5 associates the user's specific information acquired from the face matching server 6 with the face image and registers it in the face information database. Further, the face management server 5 registers the face registration ID issued by the face matching server 6 in the face information database as user information.

Next, the process of face recognition of the user in the face recognition machine 1 will be described. FIG. 16 shows a first example of face recognition processing for a user. Here, an example is shown in which a gate device (security gate) that manages the passage of the user is used as the external device 16 that cooperates with the face recognition machine 1.

When the face recognition machine 1 detects a person's face from the captured image of the camera 11 and acquires the face image, the face recognition request is transmitted to the face matching server 6. In this face matching request, the device ID of the requesting face recognition machine 1, the matching group of the face recognition machine 1, the data of the photographed image (face image for authentication) of the person to be face-recognized, and the face frame. Information and matching conditions are included.

When the face matching server 6 receives the face matching request, the face feature amount generation unit 84 first generates the face feature amount of the target person from the face image of the target person acquired from the face recognition machine 1. Next, in the face feature amount collation unit 86 of the face recognition unit corresponding to the collation group of the face recognition machine 1, the face feature amount of the target person is collated with the face feature amount of the user registered in the face recognition database. .. At this time, the face matching process is performed in the face recognition process in which the face recognition machine 1 of the requesting source and the matching group match. When the face matching process is completed, the face matching server 6 transmits the face matching response to the requesting face recognition machine 1. The response of this face collation includes a collation result (success, failure), a collation score, and a user code.

The face feature amount collation unit 86 calculates a collation score indicating the degree of similarity between the face recognition target person and the registered user, and when the collation score is equal to or higher than a predetermined reference value, the face recognition is performed. The target person is regarded as the registered user, and a matching result indicating that the face matching is successful is generated. On the other hand, if the collation score does not exceed the standard value for all users, it is determined that the target person for face authentication is not a registered user, and a collation result indicating that the face verification has failed is generated. ..

The face matching processing conditions (face matching parameters) performed by the face matching server 6 may be added to the matching request. As a result, the face recognition machine 1 can instruct the processing content of the face matching performed by the face matching server 6. For example, as a processing condition for face matching, a threshold value related to the matching score is specified, and a matching result having a matching score equal to or higher than a predetermined threshold value is included in the response. In addition, as a processing condition for face collation, the number of collation results is specified, and a predetermined number of collation results are included in the response from the one with the highest collation score.

Further, in the face matching server 6, when the face matching is completed, the authentication log management unit 87 stores information such as the matching result acquired by the face matching in the database as an authentication log (history information of face authentication). At this time, in addition to saving only the result of face recognition (success, failure) as an authentication log, the collation score may be included in the authentication log. Further, the face image of the target person acquired from the face recognition machine 1 may be included in the authentication log. In this case, it is advisable to encrypt and save the face image of the target person.

When a large number of valid collation results whose collation score exceeds the threshold value are obtained, the collation score may be narrowed down to a predetermined number from the one with the highest collation score and saved as an authentication log. In addition, when face matching is not completed normally or when a valid matching result whose matching score exceeds the threshold value cannot be obtained, only the information included in the request from the face recognition machine 1 is used as the authentication log. You may try to save it.

Further, when the face recognition machine 1 receives the face matching response indicating the success of face recognition from the face matching server 6, the face recognition machine 1 transmits an opening / closing control signal of the door portion to the linked gate device. That is, the face recognition machine 1 functions as a control device (gate opening / closing control device) that converts the face matching response into an opening / closing control signal of the door portion, thereby controlling the operation of the door portion of the linked gate device. As a result, the gate device operates to open the door portion (that is, allow the user to pass) by the drive portion 92. The face recognition machine 1 deletes the user's face image stored in the storage unit 14 when the transmission of the control signal (open / close control signal) is completed. Here, the response of face matching is converted into the opening / closing control signal of the door portion, but when the opening / closing control of the door portion of the gate device is performed by the entrance / exit management system constructed as an external system, the opening / closing control of the door portion is performed. A control signal is transmitted to the entry / exit management system so that the face recognition result is reflected in the opening / closing control of the gate device.

FIG. 17 shows a second example of face recognition processing for a user. FIG. 18 is an explanatory diagram showing an example of an authentication screen (screen at the time of authentication) displayed on the face recognition machine 1. Here, an example is shown in which a lock device provided on the door of the conference room is used as the external device 16 that cooperates with the face recognition machine 1. Regarding the second example of the face recognition process, detailed description of the same matters as the first example described above will be omitted.

The face recognition machine 1 is installed, for example, on the wall of a corridor facing the conference room. When the face recognition machine 1 detects the approach of a person (that is, detects a user) by a motion sensor or the like, the face recognition machine 1 displays a standby screen (initial screen) of the authentication screen on the display 12 (see FIG. 18A). Here, the authentication screen (notification screen) displayed on the display 12 includes an image display area 121 that displays a live image (real-time moving image) of a predetermined shooting area taken by the camera 11, and the progress of face recognition processing. Includes an information display area 122 that displays information indicating the above.

After that, the face recognition machine 1 detects the face of a person from the captured image of the camera 11 and acquires the face image, and then sends a face matching request to the face matching server 6. At this time, the character 123 and the figure 124 indicating that the user's authentication is in progress are displayed in the information display area 122 (see FIG. 18B). In the information display area 122, the display of either the character 123 or the figure 124 may be omitted. Further, in the case of the face authentication machine 1 that does not have a display, for example, an LED lamp is used, the LED is blinked during authentication, and the LED is turned on when the authentication is successful, so that the progress of face authentication is progressed. May be notified. Further, the progress of face recognition may be notified by the display color of the LED. Further, the display color of the LED may be changed depending on the success or failure of the user's face authentication.

Upon receiving the face matching request, the face matching server 6 generates a face feature amount and executes the face matching process as in the first example described above. When the face matching process is completed, the face matching server 6 transmits the face matching response to the requesting face recognition machine 1.

Further, when the face recognition machine 1 receives the face matching response indicating the success of the face recognition from the face matching server 6, the character 127 and the graphic indicating that the user authentication is completed are displayed in the information display area 122 of the authentication screen. 128 is displayed (see FIG. 18C). At the same time, the face recognition machine 1 transmits a control signal (command to unlock the door) to the linked door lock device. That is, the face recognition machine 1 functions as a control device that controls the operation of the linked lock device. As a result, the lock device performs an operation of releasing the lock (that is, permitting the user to enter the room). The face recognition machine 1 deletes the user's face image stored in the storage unit 14 when the transmission of the control signal is completed.

The face recognition machine 1 can also be linked with an external device that performs only information processing without mechanical operation like the above-mentioned gate device and door lock device. FIG. 19 shows a third example of face recognition processing for a user.

Here, an example of using a payment terminal device as an external device 16 linked with the face recognition machine 1 is shown. Regarding the third example of the face recognition process, detailed description of the same matters as the first or second example described above will be omitted.

In the third example of the face recognition process, when the face recognition machine 1 receives the face matching response indicating the success of the face recognition from the face matching server 6, the face recognition terminal device (or the payment system) to cooperate with the face recognition terminal device (or the payment system) receives a face recognition response. Send a control signal. As a result, the payment terminal device executes a process of associating the usage fee data for the service used by the corresponding user (for example, the provision of meals at the restaurant) with the authenticated user, and as a result, the relevant user. The usage fee is added to the billing data at the facility for the user.

Next, the management work by the administrator of the face recognition system will be described. FIG. 20 is an explanatory diagram showing a login screen displayed on the management terminal 2.

On the management terminal 2, when the administrator starts the management application and accesses the face management server 5, the login screen is displayed. On this login screen, the administrator can enter his / her user ID and password. When the administrator inputs the user ID and password on this login screen and operates the login button, the face management server 5 performs user authentication. If the login is successful, various management screens can be displayed. If login fails, an error will be displayed.

Next, registration, reference, update, and deletion regarding the association between the face recognition machine 1 and the matching group in this system will be described. 21 and 22 are explanatory views showing a management screen displayed on the management terminal 2 at the time of registration, reference, update, and deletion regarding the association between the face recognition machine 1 and the collation group.

When the management terminal 2 logs in to the face management server 5, the reference screen shown in FIG. 21 (A), FIG. 21 (when registering, referencing, updating, and deleting the association between the face recognition machine 1 and the matching group) The registration screen shown in B) and the deletion screen shown in FIG. 22 can be displayed.

The reference screen shown in FIG. 21A displays a list of association settings between the registered face recognition machine 1 and the collation group. By viewing this reference screen, the administrator can confirm the registered contents of the collation group.

The reference screen is provided with a list display unit 211 and a face recognition machine designation unit 212. In the list display unit 211, the device ID of each authentication machine, the collation group (number) associated with each authentication machine, and the explanation about the collation group are displayed. In the face recognition machine designation unit 212, the target face recognition machine 1 (device ID) can be selected from the pull-down menu. As a result, the display content of the list display unit 211 can be narrowed down to the specific face recognition machine 1, and only the collation group associated with the specific face recognition machine 1 is displayed on the list display unit 211. If the face recognition machine 1 is not specified by the face recognition machine designation unit 212, information about all the face recognition machines 1 is displayed on the list display unit 211.

Further, on this reference screen, items (device ID, face recognition machine 1, description) can be specified and sorting can be executed. The number of items displayed by the list display unit 211 (the number of face recognition machines 1) and the display range of the face matching server 6 and the matching group may be specified. Further, the search condition may be specified for each item (device ID, face recognition machine 1, description) so that the search can be performed.

In the example shown in FIG. 21A, the face recognition machine 1 is grouped according to the place where the face recognition machine 1 is installed (for example, the location of the business establishment).

The registration screen shown in FIG. 21B is for setting the association between the face recognition machine 1 and the collation group. Information that associates a collation group with the authentication machine can be registered.

The registration screen is provided with an authentication machine designation unit 213, a collation group designation unit 214, and a registration button 215. In the authentication machine designation unit 213, the target face recognition machine 1 (device ID) can be selected from the pull-down menu. In the collation group designation unit 214, the collation group of the target face recognition machine 1 can be selected from the pull-down menu. This pull-down menu displays the collation groups that have already been registered. When the registration button 215 is operated after inputting the face recognition machine 1 and the collation group on this registration screen, the face management server 5 performs a process of registering the collation group information with the input contents.

In some cases, a plurality of collation groups are associated with one face recognition machine 1. In this case, the registration operation of associating one collation group with one face recognition machine 1 may be repeated on the registration screen. Further, a plurality of collation group designation units 214 may be provided on the registration screen.

The deletion screen shown in FIG. 22 deletes the association setting between the face recognition machine 1 and the collation group.

The deletion screen is provided with a list display unit 216, a face recognition machine designation unit 217, and a deletion button 218. In the list display unit 216, the device ID of each authentication machine, the collation group (number) associated with each face recognition machine 1, and the explanation about the collation group are displayed. Further, the list display unit 216 is provided with a check box for each association setting. You can select the association setting by the check box. When the registration item to be deleted is selected by the check box and the delete button 218 is operated, the process of deleting the selected registration item is performed on the face management server 5. In the face recognition machine designation unit 217, the face recognition machine 1 can be selected from the pull-down menu. As a result, the display content of the list display unit 216 is updated in a state of being narrowed down to the specific face recognition machine 1.

Here, by selecting a reference on the menu screen (not shown), the transition to the reference screen shown in FIG. 21 (A) is performed. Further, by selecting registration on the menu screen, the screen transitions to the registration screen shown in FIG. 21 (B). Further, by selecting deletion on the menu screen, the screen transitions to the deletion screen shown in FIG. Also, by selecting update on the menu screen, the screen transitions to the update screen (not shown). Further, when the administrator selects the association between the face recognition machine 1 and the collation group on the reference screen, the screen transitions to the update screen (not shown).

In the example shown in FIG. 22, the registered associations are displayed in a list on the deletion screen, but the deletion may be possible on the individual editing screen. The update screen (not shown) is the same as the registration screen (FIG. 21 (B)).

FIG. 23 is an explanatory diagram showing an example of a user's management screen displayed on the management terminal 2.

On the management terminal 2, when managing information about the user, the user management screen shown in FIG. 23 is displayed.

The user management screen shown in FIG. 23 is for the administrator to perform operations of registration, reference, update, deletion, and search of a face image and specific information (here, business card information) relating to the user. .. The information (including images) input to the user's management screen includes the information acquired by the registration device 4.

The user management screen includes an operation selection area 131, a user information area 132, and a business card information area 133.

In the operation selection area 131, the administrator can select items to be executed (information registration, reference, update, deletion, search, etc.). In FIG. 23, business card registration is selected as an item to be executed. As a result, the administrator can newly input the business card description information displayed on the user's management screen.

The user information area 132 includes a user's face image, first and last name, and information used for face authentication. Here, in the user information area 132, the face image display unit 141, the first and last name input unit 142, the user ID input unit 143, the identification ID input unit 144, the collation group input unit 145, the access permission group input unit 146, and additional information. An input unit 147, an activation date input unit 148, and an invalidation date input unit 149 are included.

The face image of the user is displayed on the face image display unit 141. The administrator can delete the face image of the face image display unit 141 by pressing the delete button 151. When the face image cannot be acquired by the registration device 4, the administrator can select the prepared face image (image file) as a display target by pressing the image selection button 152. Alternatively, the administrator can display the face image selected by the image selection button 152 on the face image display unit 141 instead of the face image acquired by the registration device 4. The face image prepared by the administrator is not limited to the actual face image of the user, but may be a dummy image (for example, an avatar image of the user).

The user's first and last name is input to the first and last name input unit 142. A user ID (for example, an employee number) is input to the user ID input unit 143. In the identification ID input unit 144, an ID that can identify the user (for example, a series and unique number assigned in the order of registration of the user) is input. A collation group is input to the collation group input unit 145. In the access permission group input unit 146, an access permission group indicating an administrator group who can access the user's information is input. The additional information input unit 147 can input additional information. In the activation date input unit 148, a date on which face collation of the person can be enabled is input. In the invalidation date input unit 149, a date on which face collation can be invalidated is input. By setting the activation date and the invalidation date, it is possible to start and end the service for a plurality of users at the same time.

The business card information area 133 includes a photographed image of a business card registered by the registration device 4 and information extracted from the photographed image of the business card. Here, the business card information area 133 includes a business card image display unit 161, an affiliation information input unit 162, an activation date input unit 163, and an invalidation date input unit 164.

The business card image of the user is displayed on the business card image display unit 161. By pressing the delete button 171 and the image selection button 172, the administrator can delete the business card image and select the prepared business card image (image file) as in the case of the above-mentioned face image.

In the affiliation information input unit 162, the company name to which the user belongs, the department name of the user, the job title, the telephone number, the fax number, the e-mail, the zip code, the address, and the like are input. In the activation date input unit 163, a date on which information regarding the business card of the person can be activated is input. In the invalidation date input unit 164, a date on which the information regarding the business card of the person can be invalidated is input.

When the administrator presses the registration button 167 after completing the input to the desired input unit, the input (added or changed) information becomes valid.

In addition, on the user management screen, along with the business card description information or instead of the business card description information, other specific information (for example, information displayed on the payment card or the user's staying place in the facility) is displayed. It is also possible to manage the information shown).

In addition, the administrator can search for necessary information by selecting a search tab (for example, collation group search) in the operation selection area 131 of the user's management screen. In the search screen for such information, each input unit on the above-mentioned user management screen is displayed as a blank. For example, the administrator can search for users belonging to the entered collation group by inputting the collation group to be searched into the blank collation group input unit 145.

When the administrator enters search conditions in the required items on the search screen and presses the search button (not shown), a list display screen that reflects the search results (for example, a list of extracted users is shown. Screen).

Next, the use of information about the user will be described. FIG. 24 is an explanatory diagram showing an example of a conference room search screen (meeting search screen) displayed as a management screen on the management terminal 2 or the like. FIG. 25 is an explanatory diagram showing an example of an attendee display screen displayed as a management screen on the management terminal 2.

The conference room search screen shown in FIG. 24 is for the administrator to use (reference) the information about the conference acquired in the face recognition process. By accessing the face management server 5 from the management terminal 2, the administrator can appropriately add information that cannot be acquired by the face recognition machine 1 regarding the conference held at the facility.

Information such as the date, time, and room number (meeting room No.) is displayed on the meeting room search screen regarding the meetings held in the past in the meeting room of the facility. Information on the entry history acquired by the face recognition machine 1 linked with the lock device for the door of the conference room can be used to generate the conference room search screen. When the administrator selects a desired meeting (click the check box of the meeting shown at the top in FIG. 24) and presses the selection button 173, the screen transitions to the search result display screen in which the search result is reflected.

As the search result display screen, for example, as shown in FIG. 25, an attendee display screen including an attendee list is displayed on the management terminal 2.

The attendee list includes attendee information area 175 and respondent information area 176. In the attendee information area 175, facial images of the attendees of the conference and specific information (here, the name of the company to which the user belongs and the name of the user) are displayed. Further, in the responder information area 176, information (here, the department to which the employee belongs and the name) regarding the person who responded to the attendee (for example, a worker of a company occupying the facility) is displayed.

FIG. 26 is an explanatory diagram showing an example of a personal network search screen displayed as a management screen on the management terminal 2.

In the personal connection search screen shown in FIG. 26, the administrator uses the information about the personal connections of the user 10 acquired in the face recognition process (here, the information about the connections of people estimated from the information of the attendees of the conference) ( (See). By accessing the face management server 5 from the management terminal 2, the administrator can appropriately add necessary information regarding the user's personal connections.

The personal network search screen includes a correlation diagram display area 181 and a list display area 182. In the correlation diagram display area 181, a diagram that visualizes the personal connections of the person selected by the administrator is displayed. Further, in the list display area 182, a personal network list 185 relating to the person (user 10) selected by the administrator is displayed. In this personal connection list 185, a face image of a person, affiliation information (company name, department, etc.), a numerical value indicating the strength of the personal connection (personal network strength), and the like are displayed.

Such a personal network search screen may be displayed, for example, by the administrator selecting (clicking) an area of a desired attendee on the attendee display screen of FIG. 25.

The use of the conference room search screen and the personal connection search screen described above does not have to be limited to those via the management terminal 2. For example, by granting access rights as an administrator to a facility user (for example, a worker of a company occupying the facility), a conference room search screen can be displayed from an information processing terminal (PC, tablet, etc.) owned by the user. Or you may be able to refer to the network search screen.

As described above, embodiments have been described as an example of the techniques disclosed in this application. However, the technique in the present disclosure is not limited to this, and can be applied to embodiments in which changes, replacements, additions, omissions, etc. have been made. It is also possible to combine the components described in the above embodiment to form a new embodiment.

In recent years, in the world of IoT (Internet of Things), CPS (Cyber-Physical Systems), which is a new concept of creating new added value by linking information between physical space and cyber space, has been attracting attention. Under these circumstances, the CPS concept can be adopted in this embodiment as well. That is, as a basic configuration of CPS, for example, an edge device arranged in a physical space and a cloud server arranged in a cyber space are connected via a network, and the edge device and the processor mounted on the cloud server make a face. It is possible to distribute the authentication process. In this face recognition process, the user's face image data taken by the camera of the authentication machine is acquired as an edge device. Next, the cloud server executes a process of generating feature amount data from the face image data received from the edge device via the network, collates it with the feature amount data of the user registered in advance, and authenticates the user. The process is executed, the authentication result related to the authentication process is taken over by the face authentication application software, and displayed on the display etc. in the output format defined on the application. Here, each data generated in the edge device or the cloud server is preferably generated by a Web application or the like mounted on a standardized platform, and by using such a standardized platform, various various sensor groups are used. It is possible to improve the efficiency when constructing a system including IoT application software.

The room entry control device and the room entry control method according to the present disclosure can easily perform face recognition when a user enters a predetermined area of the facility while safely managing the face image of the user who uses the facility. Based on the result of face recognition of the subject, it is useful as an entry control device and an entry control method for controlling the unlocking of the lock device of the door into which the subject enters.

1: Face recognition machine 2: Management terminal 3: Face recognition server 4: Registration device 5: Face management server 6: Face matching server 10: User 11: Camera (face image acquisition unit)
12: Display (display unit)
14: Storage unit 15: Control unit (processor)
16: External device 25: Authentication result adjustment unit (history information generation unit)
41A: Face shooting camera (personal information acquisition department)
41B: Information acquisition camera (personal information acquisition department)
42: Display (display unit)
45: Control unit (processor)
62: User management unit 64: Equipment management unit 90: Image generation unit 97: Power supply unit

Claims (4)

For the face recognition server that executes face recognition of the user of the facility, the face recognition is performed when the user enters the predetermined area by unlocking the lock device such as a door configured as an external system of the facility. This is an entry control device that controls the unlocking of the lock device based on the authentication result.
A face image acquisition unit that photographs the user before entering the predetermined area and acquires a face image for authentication of the user.
A transmission unit that transmits a processing request for face authentication to the face recognition server based on the face image for authentication, and a transmission unit.
A receiver that receives the authentication result returned from the face recognition server, and
A control signal transmission unit that transmits the authentication result as a control signal to the external system is provided.
A room entry control device, characterized in that a processor of the external system receives the control signal and reflects it in a lock release signal of the lock device. The entry control device according to claim 1, further comprising a power feeding unit that receives power from the lock device. A storage unit for storing the face image for authentication is further provided.
The entry control device according to claim 1 or 2, wherein the face image for authentication stored in the storage unit is deleted after the face authentication process is completed. For the face authentication server that executes face authentication of the user of the facility, the face authentication is performed when the user enters the predetermined area by unlocking the lock device such as a door configured as an external system of the facility. This is a room entry control method that controls the unlocking of the lock device based on the authentication result of the processing request.
The user before entering the predetermined area is photographed, and the face image for authentication of the user is acquired.
Based on the face image for authentication, the face authentication processing request is transmitted to the face authentication server, and the face authentication processing request is transmitted.
Upon receiving the authentication result returned from the face recognition server,
The authentication result is transmitted as a control signal to the external system.
A room entry control method, wherein a processor of the external system receives the control signal and reflects it in a lock release signal of the lock device.

Images