JP2022068019A - Login system, electronic device and login control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a login system, an electronic device and a login control method for selecting an appropriate authentication method according to the number of people detected from a captured image.
SOLUTION: A cash register, a surveillance camera that captures an area including at least the area around the cash register, a transaction management server that manages transactions of the cash register, and a camera image management server that manages images captured by the surveillance camera. In a monetary processing system consisting of a face recognition server that executes face recognition and a network that interconnects them, the control unit of the face recognition server acquires a captured image that includes the requesting device in the imaging range. S2, S4 for determining whether or not the number of people included in the captured image is one. When the number of people is one, S5 sets both 1: N authentication and 1: 1 authentication to be feasible. If the number of people is not one, 1: N authentication is set to be non-executable, while 1: 1 authentication is set to be executable S10.
[Selection diagram] FIG. 6

Description

The present invention relates to a login system, an electronic device and a login control method.

Some devices require login prior to use. Today, there are various authentication technologies for login, one of which is face recognition. In the login using face authentication, the user's face image and the registered face image are collated, and if the collation is successful, the login is permitted.
By the way, it is possible that facial images of a plurality of people are captured at the same time. Therefore, Patent Document 1 describes a technique for specifying a user who has executed a predetermined hand gesture as a login target person and performing face recognition for the specified user.

Japanese Unexamined Patent Publication No. 2019-159573

In the case of Patent Document 1, execution of a hand gesture is required even when there is only one login target person. However, when there is only one face image to be captured, it is possible to identify the person to be logged in without performing a hand gesture, and the request for the gesture forces the user to perform unnecessary work. Become.

It is an object of the present invention to be able to select an appropriate authentication method according to the number of people detected from a captured image.

The invention according to claim 1 is a login system that executes face authentication based on a face image acquired from a captured image, and executes 1: N authentication based on the number of people information detected from the captured image. It is a login system equipped with a control unit that controls the availability of the above.
The invention according to claim 2 is an acquisition unit that acquires face information from a captured image, a storage unit that stores face information of a plurality of people, and the acquisition unit by 1: 1 authentication or 1: N authentication. The first aspect of claim 1, further comprising a collation unit that collates the acquired face information with the face information stored in the storage unit, and a permission unit that permits login based on the collation result by the collation unit. It is a login system.
The invention according to claim 3 is the login system according to claim 1 or 2, wherein the control unit uses the number of faces acquired from the captured image as the number of people information.
The invention according to claim 4 is the login system according to claim 1 or 2, wherein the control unit uses the number of skeletons acquired from the captured image as the number of people information.
The invention according to claim 5 further includes a display unit, and the control unit switches the display mode on the display unit depending on whether the number of people information is a plurality of persons or one person. It is a login system described in.
In the invention according to claim 6, when the number of people information is a plurality of people, the control unit displays a screen for 1: 1 authentication on the display unit in the first size, and the number of people information is one person. The login system according to claim 5, wherein the screen for 1: 1 authentication is displayed on the display unit by the second size, which is smaller than the first size.
In the invention according to claim 7, the control unit displays a screen for 1: 1 authentication on the display unit when the number of people information is a plurality of people, and 1: 1 when the number of people information is one person. The login system according to claim 5, wherein the screen for authentication is not displayed on the display unit.
The login system according to claim 7, wherein the control unit displays a button for calling the screen for 1: 1 authentication on the display unit when the number of people information is one person. Is.
The invention according to claim 9 is the login system according to claim 1 or 2, wherein the control unit instructs the execution of 1: N authentication when the number of people information is one.
The invention according to claim 10 further includes a display unit, and when the control unit additionally registers a face image to be used for authentication, the control unit displays a candidate for the face image to be added on the display unit. Or the login system according to 2.
The invention according to claim 11 is the login system according to claim 10, wherein the control unit determines a candidate selected by the user from the candidates displayed on the display unit as a registration target.
In the invention according to claim 12, the control unit displays a candidate for a face image to be added on the display unit when the number of people information is a plurality of people, and the face image to be added when the number of people information is one person. The login system according to claim 10, wherein the candidate is not displayed on the display unit.
According to the thirteenth aspect of the present invention, when the control unit additionally registers a face image used for authentication, the added face image matches the face image newly acquired from the user during the login operation. The login system according to claim 10, wherein a rate is calculated and an alarm is output when the calculated match rate is lower than a predetermined standard.
In the invention according to claim 14, when the control unit additionally registers a face image used for authentication, the face image to be added and the face already registered for the user associated with the face image are registered. The login system according to claim 10, which calculates a matching rate with an image and outputs an alarm when the calculated matching rate is lower than a predetermined standard.
The invention according to claim 15 is the login system according to claim 1 or 2, wherein the control unit records as a history whether the user has logged in with 1: N authentication or logged in with 1: 1 authentication.
The invention according to claim 16 is the login system according to claim 1 or 2, wherein the control unit always permits execution of 1: 1 authentication.
The invention according to claim 17 is the login system according to claim 1 or 2, wherein the control unit cannot execute 1: 1 authentication when the number of people information is one person.
The invention according to claim 18 is the login system according to any one of claims 1 to 17, further comprising an imaging unit that captures an image peripheral area of an electronic device to be logged in as the captured image.
The invention according to claim 19, wherein the imaging unit is a surveillance camera that monitors the inside of a facility where the electronic device is provided, a first server that stores processing data of the electronic device, and processing of the monitoring camera. The login system according to claim 18, further comprising a second server different from the first server for storing data.
18. The invention according to claim 20, wherein the image pickup unit is a camera provided or attached to the electronic device, and further includes a server for storing processing data of the electronic device and processing data of the camera. It is a login system described in.
The invention according to claim 21 is an electronic device that requests face authentication using a face image acquired from a captured image at login, and 1: N authentication based on the number of people information detected from the captured image. It is an electronic device provided with a control unit that controls whether or not the execution of the above is possible.
The invention according to claim 22 is a login control method in a computer, which is detected from the captured image when face authentication using the face image acquired from the captured image is required at the time of logging in to an electronic device. It is a login control method having a process of controlling whether or not 1: N authentication can be executed based on the number of people information.

According to the present invention, an appropriate authentication method can be selected according to the number of people detected from the captured image.

It is a figure explaining the configuration example of the money processing system used in Embodiment 1. It is a figure explaining the appearance example of the balance machine used in Embodiment 1. FIG. It is a block diagram which shows the functional structure example of the balance machine used in Embodiment 1. FIG. It is a block diagram which shows the functional structure example of the face authentication server used in Embodiment 1. FIG. It is a figure explaining two kinds of methods used for the detection of the number of people in the number of people detection unit in an image. (A) is a figure explaining a method of counting the number of faces, and (B) is a figure explaining the method of counting the number of skeletons. It is a flowchart explaining an example of the processing operation executed by the face authentication server used in Embodiment 1. FIG. It is a figure which shows an example of the login screen which is displayed in a balance machine when the number of people included in a captured image is one. It is a figure explaining the example of the image taken by the surveillance camera. It is a figure which shows an example of the login screen which is displayed in a balance machine when the number of people included in a captured image is a plurality of people. It is a figure explaining another example of the image taken by the surveillance camera. It is a figure which shows the other example of the login screen displayed on the balance machine when the number of people included in a captured image is one. It is a figure which shows the other example of the login screen displayed on the balance machine when the number of people included in a captured image is one. It is a block diagram which shows the functional structure example of the face authentication server used in Embodiment 2. It is a figure explaining the transition example of the login screen in Embodiment 2. (A) is an example of a screen displayed during execution of face authentication, (B) is an example of a screen displayed when face authentication fails, and (C) is selected to log in by entering a number. (D) is an example of a screen displayed when login is successful, and (E) is an example of a menu screen. It is a figure explaining another transition example of the login screen in Embodiment 2. FIG. (A) is an example of a screen displayed during execution of face authentication, (B) is an example of a screen displayed when face authentication fails, and (C) is selected to log in by entering a number. (D) is an example of a screen displayed when login is successful, and (E) is an example of a menu screen. It is a figure explaining another transition example of the login screen in Embodiment 2. FIG. (A) is an example of a screen displayed during execution of face authentication, (B) is an example of a screen displayed when face authentication fails, and (C) is selected to log in by entering a number. (D) and (E) are examples of screens displayed when login is successful, and (F) is an example of menu screens. It is a flowchart explaining an example of the processing operation which the control part of a face authentication server executes when the additional registration of a face image is instructed by the person in charge of login. It is a flowchart explaining another example of the processing operation which the control part of a face authentication server executes when the additional registration of a face image is instructed by the person in charge of login. It is a flowchart explaining an example of the processing operation executed by the face authentication server used in Embodiment 3. FIG. It is a figure explaining the configuration example of the money processing system used in Embodiment 4. It is a figure explaining the configuration example of the money processing system used in Embodiment 5. It is a block diagram which shows the functional structure example of the balance machine used in Embodiment 5. It is a figure explaining the configuration example of the money processing system used in Embodiment 6. It is a figure explaining the example of the withdrawal opening provided in the balance machine. It is a flowchart explaining an example of the processing operation executed in the mistake monitoring server. It is a figure explaining the scene which two workers use one balance machine at the same time. It is a figure explaining the withdrawal opening related to Mr. A's transaction and the withdrawal opening related to Mr. B's transaction. (A) represents the position of the withdrawal port related to Mr. A's transaction, and (B) represents the position of the withdrawal port related to Mr. B's transaction. It is a figure explaining the output example of an alarm when a mistake occurs. It is a figure explaining the configuration example of the money processing system used in Embodiment 7. It is a figure explaining the functional configuration example of the medical expense payment machine used in Embodiment 7.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings.
<Embodiment 1>
<System configuration>
FIG. 1 is a diagram illustrating a configuration example of the money processing system 1 used in the first embodiment.
The money processing system 1 in the present embodiment is installed in a store (facility) of a financial institution such as a bank. In the following, when banknotes and coins are collectively referred to as "money".

The money processing system 1 shown in FIG. 1 includes a balance machine 10, a surveillance camera 20 that captures an image of at least an area including a peripheral area 20A of the balance machine 10, a transaction management server 30 that manages transactions of the balance machine 10, and a surveillance camera. It is composed of a camera image management server 40 that manages the images captured by the 20th, a face recognition server 50 that executes face recognition, and a network 60 that connects them to each other.
A plurality of surveillance cameras 20 may be installed in the store. However, in the present embodiment, it is assumed that at least one surveillance camera 20 is attached at a position where the face image of the worker who logs in to the balance machine 10 can be captured. Although the surveillance camera 20 is assumed in FIG. 1, a dedicated camera installed for face recognition may be used.

FIG. 1 illustrates the vicinity of the front of the balancer 10 as an example of the peripheral area 20A imaged by the surveillance camera 20. However, the mounting position of the surveillance camera 20 is not limited to the position where the face of the worker who logs in to the balance machine 10 can be imaged from the front, and the worker can image the face by changing the direction of the body or the direction of the face. It may be a position. For example, as shown in FIG. 1, it may be mounted on the wall or ceiling on the left side of the balancer 10.

A wide-angle lens is used in the surveillance camera 20 so that a wide range can be imaged at once. The surveillance camera 20 may be provided with a zoom function. For example, in the case of face recognition, the zoom function may be used to capture a close-up image of the face in the captured image.
The balance machine 10 here is an example of an electronic device. The money processing system 1 shown in FIG. 1 is an example of a login system. Further, the surveillance camera 20 is an example of an imaging unit.

In the case of FIG. 1, the transaction management server 30 and the camera image management server 40 are installed in the store as independent servers.
The transaction management server 30 constitutes an information system together with the accounting machine 10. The transaction management server 30 is an example of a first server that stores the processing data of the balance machine 10.
The camera image management server 40 constitutes a surveillance system together with the surveillance camera 20. The camera image management server 40 is an example of a second server that stores the processing data of the surveillance camera 20.
However, the transaction management server 30 and the camera image management server 40 may be integrated into one server.

The face recognition server 50 performs face recognition on a worker who logs in to the cashier 10 by collating the face image captured by the surveillance camera 20 with the face image registered in advance. Specifically, the face authentication server 50 detects, for example, 100 features from the entire face, and determines the difference by using the detected features of the entire face. However, the number of features to be detected is an example. In the case of the present embodiment, the features of the entire face are detected evenly from the entire face. The facial features may be detected not from the entire face but from specific parts such as eyes and nose and their surroundings.
The face recognition server 50 may acquire a face image used for face recognition via the camera image management server 40.

The face recognition server 50 in this embodiment supports both 1: 1 authentication and 1: N authentication.
1: 1 authentication consists of a face image of the person in charge of logging in (hereinafter also referred to as "login person in charge") and a face image associated with identification information such as an ID (= Identification) entered by the person in charge of login. Refers to the method of collating.
1: N authentication refers to a method of collating a face image of a person in charge of login with a plurality of face images registered in a database.

The network 60 is composed of a communication cable such as a LAN (= Local Area Network) cable, and is also composed of wireless communication.
The network 60 shown in FIG. 1 includes other devices used in transactions related to the business of a financial institution, such as a money deposit / withdrawal machine (not shown), a money deposit / withdrawal machine (not shown), and a window terminal (not shown). ), A money changer (not shown) is also connected.

<Configuration of each device>
<Structure of the balance machine 10>
FIG. 2 is a diagram illustrating an external example of the balance machine 10 used in the first embodiment. The cashier 10 shown in FIG. 2 includes a coin processing unit 11 that processes coin deposits and withdrawals, a bill processing unit 12 that processes bill deposits and withdrawals, and banded bills (hereinafter also referred to as “bundle bills”) and packaged coins (hereinafter also referred to as “bundle bills”). It has a currency storage unit 13 for storing (hereinafter also referred to as "bar"), an operation display unit 14 for them, and a printing unit 15.

The coin processing unit 11 executes deposit / withdrawal processing of loose coins, preparation processing of packaged coins, withdrawal processing of packaged coins, scrutiny processing, and the like. Packaging coins are made by, for example, packaging 50 loose coins.
The banknote processing unit 12 executes deposit / withdrawal processing of loose banknotes, creation processing of banded banknotes, withdrawal processing of banded banknotes, scrutiny processing, and the like. The banded banknote is created by, for example, banding 100 loose banknotes.

The money storage unit 13 is a cabinet-type processing unit called a so-called cash bus. The money storage unit 13 stores coins such as loose banknotes, strapped banknotes, and packaged coins, as well as securities and the like. The money storage unit 13 is also provided with a function of scrutinizing the money being stored.
The operation display unit 14 is composed of, for example, a touch panel or the like. The operation display unit 14 is used to input an operation to the balancer 10. The operations here include operations associated with login. Further, the operation display unit 14 displays information such as the processing status of banknotes and coins in the suiting machine 10, and the amount of banknotes and coins stored in the suiting machine 10. The operation display unit 14 is an example of a display unit.
The printing unit 15 is composed of, for example, a printer. As a result of the processing executed by the balance machine 10, the printing unit 15 prints information such as the amount of money stored in the balance machine 10 on a receipt or the like.

FIG. 3 is a block diagram showing a functional configuration example of the balance machine 10 used in the first embodiment.
As shown in FIG. 3, the balance unit 10 is provided with a control unit 100 that controls the movement of the entire device.
The control unit 100 gives a command signal to the coin processing unit 11, the bill processing unit 12, and the money storage unit 13 to control their operations. Further, the operation display unit 14, the printing unit 15, the storage unit 16, and the communication interface unit 17 are connected to the control unit 100, respectively. In addition, a motion sensor that detects the approach of the operator to a position where the operation display unit 14 can be operated may be provided.

The storage unit 16 is composed of a hard disk device or the like. The storage unit 16 stores data on transactions executed by the balance machine 10 (that is, transaction data), the amount of banknotes and coins stored in the balance machine 10, and the like.
The communication interface unit 17 is composed of a communication module that enables communication via the network 60 (see FIG. 1).

The control unit 100 includes a CPU (= Central Processing Unit), a ROM (= Read Only Memory), and a RAM (= Random Access Memory). That is, the control unit 100 constitutes a computer. The ROM stores BIOS (= Basic Input Output System) and various settings. The CPU uses the RAM as a work area to execute the program.
The control unit 100 in the present embodiment functions as a login permission unit 101 and a transaction processing unit 102 through the execution of the program by the CPU.

The login permission unit 101 is a processing unit that determines whether to allow or disallow login in cooperation with the face authentication server 50 (see FIG. 1). The login permission unit 101 permits the operation of the own machine when the face authentication on the face authentication server 50 is successful. On the other hand, the login permission unit 101 does not allow the operation of its own machine when the face authentication on the face authentication server 50 fails.

The transaction processing unit 102 is a processing unit that executes processing related to a transaction received by an instruction from a logged-in user. The transaction processing unit 102 gives a command signal to the coin processing unit 11, the bill processing unit 12, and the money storage unit 13 according to the contents of the transaction, and controls the operation and progress of each unit. Further, the transaction processing unit 102 transmits information identifying the logged-in user, transaction details, transaction data including the date and time of the transaction, and the like to the transaction management server 30 (see FIG. 1).

<Configuration of face recognition server 50>
The face recognition server 50 has a hardware configuration as a so-called computer.
FIG. 4 is a block diagram showing a functional configuration example of the face authentication server 50 used in the first embodiment.
As shown in FIG. 4, the face recognition server 50 is provided with a control unit 500 that controls the movement of the entire device.

The control unit 500 includes a face image database (hereinafter referred to as “face image DB”) 51 in which a face image registered in advance is stored, and a face recognition history database in which a history of face recognition executed at login is stored. 52 (hereinafter referred to as "face recognition history DB") and communication interface unit 53 are connected to each other.

The face image DB 51 is composed of a hard disk device or the like. In the case of this embodiment, the face image DB 51 is provided in the face authentication server 50. However, the face image DB 51 may be externally attached to the face authentication server 50, or may be connected via the network 60.
The face image DB 51 may exist in a place different from the store where the cashier 10 is provided. In the face image DB 51, a face image for authentication is registered for a store in which the cashier 10 is installed or an employee working at the same financial institution.

The face authentication history DB 52 is composed of a hard disk device or the like. In the case of this embodiment, the face authentication history DB 52 is provided separately from the face image DB 51. The face authentication history DB 52 stores the authentication method used at the time of login together with information for identifying the authenticated user, the date and time of authentication, information for identifying the device to be logged in, and the like. The authentication methods here are 1: 1 authentication and 1: N authentication.
The communication interface unit 53 is composed of a communication module that enables communication via the network 60 (see FIG. 1).

The control unit 500 includes a CPU, a ROM, and a RAM. That is, the control unit 500 constitutes a computer. The ROM stores BIOS, various settings, and the like. The CPU uses the RAM as a work area to execute the program.
The control unit 500 in the present embodiment has an in-image number detection unit 501, an execution enablement control unit 502, a face image acquisition unit 503, a collation unit 504, a collation result notification unit 505, and an authentication history recording unit through the execution of a program by the CPU. Functions as 506.

The number of people detection unit 501 in the image is a processing unit that detects the number of people included in the captured image captured in the peripheral area 20A (see FIG. 1) of the balancer 10 (see FIG. 1) and outputs it as the number of people information.
In the case of this embodiment, the captured image is read from the camera image management server 40. However, the in-image number detection unit 501 may directly acquire the captured image from the surveillance camera 20 (see FIG. 1).

As a method of detecting the number of people, there are a method of counting the number of faces included in the captured image and a method of counting the number of skeletons included in the captured image.
FIG. 5 is a diagram illustrating two types of methods used for detecting the number of people in the image number detection unit 501. (A) is a figure explaining a method of counting the number of faces, and (B) is a figure explaining the method of counting the number of skeletons.
In FIG. 5A, one face image 512 is detected from the captured image 511. In FIG. 5B, the human skeleton 513 detected from the captured image 511 is schematically shown. The skeleton 513 is not represented by the human skeleton itself, but by a skeleton model prepared in advance.

The method of counting the number of faces can also be used as the function of the face image acquisition unit 503. However, the detection of the face image 512 by the in-image number detection unit 501 may be the detection of the head, unlike the case of face recognition.
The method of counting the number of skeletons is to accurately determine the number of people reflected in the captured image 511 even when multiple heads overlap in the image and it is difficult to separate one head from the other. It is possible to count.
The number of people detection unit 501 in the image gives the detected number of people to the executionability control unit 502 as the number of people information.

The executionability control unit 502 is a processing unit that controls the executionability of each of 1: 1 authentication and 1: N authentication.
The execution enablement control unit 502 in the present embodiment permits the execution of 1: N authentication and the execution of 1: 1 authentication when the number of people included in the captured image is one. However, when the number of people included in the captured image is one, the execution of 1: N authentication has priority over the execution of 1: 1 authentication. In other words, 1: N authentication is executed unless the person in charge of login instructs face authentication by 1: 1 authentication.

When the number of people included in the captured image is a plurality of people, the execution enablement control unit 502 sets 1: N authentication to be unexecutable while 1: 1 authentication is set to be executable.
In the case of this embodiment, the 1: 1 authentication is permitted regardless of the number of people included in the captured image.
On the other hand, 1: N authentication can be executed or not depending on the number of people included in the captured image.
The executionability control unit 502 here is an example of the function of the control unit.

The face image acquisition unit 503 is a processing unit that detects a face region included in the captured image and acquires it as a face image when face authentication is requested from the balance machine 10. The request for face recognition is, for example, when a motion sensor (not shown) detects the approach of the operator to a position where the operation display unit 14 (see FIG. 3) can be operated, or when the operator detects the approach of the operator to the operation display unit 14's standby screen. It is output when you touch. The standby screen is a screen that is output during the period when no work is being performed.

The collation unit 504 is a processing unit that collates the face image detected from the captured image with the face image stored in the face image DB 51.
1: In the case of N authentication, the collation unit 504 collates one face image detected from the captured image with a plurality of face images stored in the face image DB 51. Specifically, a face image matching the face image of the person in charge of login detected from the captured image is searched from the face image DB 51.

On the other hand, in the case of 1: 1 authentication, the collation unit 504 accepts the input of the person in charge number or the personal identification number that identifies the person in charge of login. After that, the collation unit 504 reads out the face image registered for the received person in charge number and the like from the face image DB 51, and collates the read face image with the face image detected from the captured image. In the case of the present embodiment, the captured image may include one face image or a plurality of face images.

The collation result notification unit 505 is a processing unit that notifies the balancer 10 requesting face recognition of the collation result by the collation unit 504. If the verification is successful, the balancer 10 is also notified of information such as information for identifying the authenticated login person and the date and time of authentication. On the other hand, when the collation fails, the accountant 10 is notified of the fact that the authentication has failed and information such as the date and time.
The authentication history recording unit 506 is a processing unit that records the history of executed authentication in the face recognition history DB 52. The stored authentication history also includes information on the method used for authentication. That is, it contains information indicating whether 1: N authentication has been executed or 1: 1 authentication has been executed.

<Processing operation>
Hereinafter, the processing operation in the present embodiment will be described with reference to FIGS. 6 to 12.
FIG. 6 is a flowchart illustrating an example of a processing operation executed by the face authentication server 50 (see FIG. 1) used in the first embodiment. The symbol S shown in the figure means a step. The processing operation shown in FIG. 6 is an example of a login control method.

First, the control unit 500 (see FIG. 4) determines whether or not the request for face recognition has been received (step 1). While the negative result is obtained in step 1, the control unit 500 repeats the determination in step 1.
If an affirmative result is obtained in step 1, the control unit 500 acquires an captured image including the requesting device in the imaging range (step 2). In the case of this embodiment, the requesting device is the balance machine 10 (see FIG. 1).

Next, the control unit 500 detects the number of people included in the acquired captured image (step 3). One of the above-mentioned two methods is used to detect the number of people.
When the number of people is detected, the control unit 500 determines whether or not the number of people is one (step 4).
When the number of people included in the captured image is one, the control unit 500 obtains a positive result in step 4. On the other hand, when the number of people included in the captured image is a plurality of people, the control unit 500 obtains a negative result in step 4.

If an affirmative result is obtained in step 4, the control unit 500 sets both 1: N authentication and 1: 1 authentication to be feasible (step 5).
Next, the control unit 500 instructs the requesting device to display the login screen for 1: N authentication (step 6). In the present embodiment, when both 1: N authentication and 1: 1 authentication can be executed, the execution of 1: N authentication has priority. As a result, the login screen for 1: N authentication is displayed on the operation display unit 14 (see FIG. 3) of the balance machine 10 (see FIG. 1).

FIG. 7 is a diagram showing an example of a login screen 140 displayed on the balancer 10 when the number of people included in the captured image is one. The login screen 140 is for 1: N authentication. On the login screen 140, the words "login waiting screen" and the words "please turn your face toward the surveillance camera" are displayed.

FIG. 8 is a diagram illustrating an example of an image (that is, an captured image) 511 captured by the surveillance camera 20. In the case of FIG. 8, the captured image 511 includes only one worker.
1: In the case of N authentication, the person in charge of login only needs to turn his face toward the surveillance camera 20. That is, the login person does not need to additionally perform a hand gesture or enter the person in charge number or the personal identification number.

Returning to the description of FIG.
Following step 6, the control unit 500 acquires the face image of the person in charge of login from the captured image (step 7).
When the face image of the login person is acquired, the control unit 500 collates the captured face image of the login person with the face image stored in the face image DB (step 8).
After that, the control unit 500 notifies the requesting device of the collation result (step 9). Specifically, the balancer 10 is notified of the collation result.

On the other hand, when a negative result is obtained in step 4, the control unit 500 sets 1: N authentication to be non-executable, while 1: 1 authentication is set to be executable (step 10).
Next, the control unit 500 instructs the requesting device to display the login screen for 1: 1 authentication (step 11). In the present embodiment, the login screen for 1: 1 authentication is displayed on the operation display unit 14 (see FIG. 3) of the balance machine 10 (see FIG. 1).

FIG. 9 is a diagram showing an example of a login screen 141 displayed on the balancer 10 when the number of people included in the captured image is a plurality of people. The login screen 141 is for 1: 1 authentication. On the login screen 141, the words "login waiting screen" and the words "please enter a number and turn your face toward the surveillance camera" are displayed. Further, on the login screen 141, a person in charge number or a password input field 142 and a numeric keypad button 143 are displayed. The person in charge number and the personal identification number are examples of login codes.

However, initially, only the input field 142 is displayed on the login screen 141, and when the input field 142 is tapped, the numeric keypad button 143 may be displayed in a pop-up.
In the case of 1: 1 authentication, even if multiple people are reflected in the captured image of the area 20A (see Fig. 1) around the balance machine 10, one person in charge of login can be entered by entering the person in charge number and PIN. Can be specified in.
In FIG. 9, the person in charge of login is specified by inputting the person in charge and the password. Instead of inputting these numbers, the person in charge of login reads the login card possessed by the person in charge of login with a card reader (not shown). A method of reading by a unit may be adopted.
FIG. 10 is a diagram illustrating another example of the image (that is, the captured image) 511 captured by the surveillance camera 20. In the case of FIG. 10, the captured image 511 includes four workers.

Returning to the description of FIG.
Following step 11, the control unit 500 acquires a login code from the requesting device (step 12). In the present embodiment, the person in charge number and the personal identification number are acquired as the login code.
When the code is acquired, the control unit 500 acquires the face image associated with the login code from the face image DB 51 (see FIG. 4) (step 13). Here, the face image of the worker who entered the login code (that is, the person in charge of login) is acquired.

Next, the control unit 500 collates the face image of the person in charge of login acquired from the face image DB 51 with the face image included in the captured image (step 14).
After that, the control unit 500 notifies the requesting device of the collation result (step 15). Specifically, the balancer 10 is notified of the collation result.

As described above, in the case of the present embodiment, if there is only one worker reflected in the captured image, face recognition can be started only by turning the face toward the surveillance camera 20 (see FIG. 1). That is, the person in charge of login does not need to perform any special operation such as executing a gesture or inputting a login code. As a result, the number of required operations can be reduced as compared with the method described in Patent Document 1.
On the other hand, in the case of this embodiment, if there are a plurality of workers reflected in the captured image, the login screen for 1: 1 authentication is automatically displayed, and the login person who has entered the login code Face recognition can be started.

<Examples of other login screens>
<Example 1>
FIG. 11 is a diagram showing another example of the login screen 140 displayed on the balancer 10 when the number of people included in the captured image is one. FIG. 11 is shown with reference numerals corresponding to the portions corresponding to those in FIG. 7.

The login screen 140 shown in FIG. 11 assumes 1: N authentication as described above. However, the wording 144 and the button 145 for switching to the login by 1: 1 authentication are added to the login screen 140.
This is because, in the case of the present embodiment, the execution of 1: 1 authentication is permitted regardless of whether the number of workers reflected in the captured image is one or more.

In the case of FIG. 11, the wording 144 indicates "* When logging in by entering a number, press the button on the right to switch the screen." Login using "number input" means using 1: 1 authentication for face authentication.
The position where the button 145 for calling the login screen for 1: 1 authentication is arranged is arbitrary, but in FIG. 11, it is arranged on the right side of the wording 144. Therefore, the wording 144 expresses "right button".
It is also possible to use the term "1: 1 authentication" instead of the phrase "number input".

<Example 2>
FIG. 12 is a diagram showing another example of the login screen 140 displayed on the balancer 10 when the number of people included in the captured image is one. In FIG. 12, a reference numeral corresponding to a portion corresponding to that in FIG. 11 is added.
The login screen 140 for 1: N authentication shown in FIG. 11 described above displayed a button 145 for calling the login screen for 1: 1 authentication, but the login screen 140 shown in FIG. 12 displayed 1: 1. The login screen for authentication is displayed from the beginning.

In the case of FIG. 12, the wording 144A indicates "* When logging in by entering a number, perform the following operations and then turn your face toward the surveillance camera." Further, on the lower side of the wording 144A, an input field 146 for a person in charge number or a personal identification number and a numeric keypad 147 are displayed. The input field 146 and the numeric keypad button 147 here are the same as the input field 142 and the numeric keypad button 143 in FIG.
Since the input field 146 and the numeric keypad button 147 are displayed from the beginning on the login screen 140 shown in FIG. 12, the person in charge of login can immediately input the number for 1: 1 authentication.

However, the login screen 140 is only a screen for 1: N authentication, and 1: 1 authentication is an auxiliary position. Therefore, the input field 146 for 1: 1 authentication and the numeric keypad 147 displayed on the login screen 140 for 1: N authentication are the input field 142 and the numeric keypad on the login screen 140 for 1: 1 authentication shown in FIG. It is displayed smaller than the size of the button 143.

The size of the input field 142 and the numeric keypad button 143 shown in FIG. 9 is an example of the first size, and the size of the input field 146 and the numeric keypad button 147 shown in FIG. 12 is an example of the second size.
If the size of the numeric keypad 147 is too small, it cannot be operated correctly, and if the size of the input field 146 is too small, it is difficult to confirm the input number. Therefore, the sizes of the input field 146 and the numeric keypad 147 are determined from a practical point of view.
Further, the numeric keypad button 147 may be displayed in a pop-up when the input field 146 is tapped.

<Embodiment 2>
<System configuration>
In actual face recognition, face recognition may fail even though the worker is a genuine worker whose face image is registered. For example, depending on the orientation of the newly imaged face, the amount of illumination light, the amount of sunlight, etc., the matching of the face image cut out from the captured image with the registered face image may end in failure.
In the present embodiment, when the personal authentication by the number input executed after the failure is successful, the face image for which the face authentication has failed can be registered as an additional face image for the person in charge of login. Will be explained.

Also in the case of the second embodiment, the money processing system 1 (see FIG. 1) described in the first embodiment is assumed.
FIG. 13 is a block diagram showing a functional configuration example of the face authentication server 50 used in the second embodiment. FIG. 13 is shown with reference numerals corresponding to the portions corresponding to those in FIG.
The face authentication server 50 shown in FIG. 13 differs from the face authentication server 50 shown in FIG. 4 in that it has a face image addition control unit 507.
The face image addition control unit 507 is a processing unit for additionally registering the face image used for face authentication by associating it with the person in charge of login when the login by entering the number is successful immediately after the face authentication fails. Is. The face image additional control unit 507 is an example of the function of the control unit.

<Login screen example>
<Example 1>
FIG. 14 is a diagram illustrating a transition example of the login screen according to the second embodiment. (A) is an example of a screen displayed during execution of face authentication, (B) is an example of a screen displayed when face authentication fails, and (C) is selected to log in by entering a number. (D) is an example of a screen displayed when login is successful, and (E) is an example of a menu screen.

Each screen shown in FIG. 14 is displayed on the operation display unit 14 of the balance machine 10 (see FIG. 1) to be logged in. The display of the login screen shown in FIG. 14 is executed regardless of whether the face authentication method is 1: 1 authentication or 1: N authentication.
The screen 150 shown in FIG. 14A is displayed during the execution of face authentication by the face authentication server 50 (see FIG. 1), that is, when the cashier 10 is waiting for the result of face authentication. The words "Communicating" and "Please wait" are displayed on the screen 150.

The screen 151 shown in FIG. 14B is displayed when face recognition fails. On the screen 151, two options are displayed following the words "Face recognition failed" and "Do you want to perform face recognition again?". In the case of screen 151, "Yes: Re-execute face recognition" and "No: Log in by entering the number" are displayed.
Incidentally, login by number input does not mean 1: 1 authentication in the above-mentioned face authentication, but means an authentication method for collating the input number with the registered number.

In the case of FIG. 14B, the cursor 151A indicating the selected state is displayed at the position of "No: Log in by inputting a number". For example, when the login person taps any option, the cursor 151A moves to the side of the tapped option. However, the person in charge of login may drag the cursor 151A to the position of the desired option.
To confirm the selection, tap the confirmation button 151B located at the bottom of the screen 151.
In the case of FIG. 14B, the selection is confirmed by logging in by entering the number.

The screen 152 shown in FIG. 14C is displayed when login by inputting a number is selected. On the screen 152, a wording prompting the input of a number, a number input field, and a numeric keypad are arranged. However, in the case of the screen 152, only the message "Please enter the number" is displayed, and it is not required to turn the face toward the surveillance camera. This is the difference from 1: 1 authentication, which is premised on face recognition.

The screen 153 shown in FIG. 14D is displayed when the login is successful. On the screen 153, the personal information of the worker identified by the face authentication server 50 as the person in charge of login is displayed together with the wording indicating the success of the login. Further, on the screen 153, a wording for confirming whether to additionally register the face image for which the face authentication has failed as the face image of the person in charge of login is displayed. Specifically, the phrase "Do you want to additionally register the previous face image?" Is displayed.
Further, on the screen 153, a button 153A operated when additional registration is desired and a button 153B operated when additional registration is not desired are also arranged.

When the button 153A or the button 153B is operated, the screen 153 transitions to the screen 154 shown in FIG. 14 (E). The screen 154 is a menu screen. On the screen 154, a button associated with a transaction or the like that can be executed by the balance machine 10 is displayed. In the case of the screen 154, 12 buttons are displayed. When the button 153A is operated, the face image when the face authentication fails is registered in association with the person in charge of login (Taro Yamada in the example of FIG. 14).

In the case of FIG. 14, the registration of the additional facial image is performed incidentally immediately after the login is successful. Therefore, the work load of the person in charge is reduced as compared with the case where the dedicated function is activated only for the registration of the face image.
In addition, since the face image captured in an environment where face recognition fails can be registered, it becomes easy to succeed in face recognition from the next time onward in the same environment. As a result, as illustrated in FIG. 14, the number of times to select login by inputting a number can be reduced.

<Example 2>
FIG. 15 is a diagram illustrating another transition example of the login screen in the second embodiment. (A) is an example of a screen displayed during execution of face authentication, (B) is an example of a screen displayed when face authentication fails, and (C) is selected to log in by entering a number. (D) is an example of a screen displayed when login is successful, and (E) is an example of a menu screen.
FIG. 15 is shown with reference numerals corresponding to the portions corresponding to those in FIG.

The difference from FIG. 14 is that the candidate face image 153C to be added is displayed for confirmation on the screen 153 shown in FIG. 15 (D). In the case of FIG. 15D, the person in charge of login can determine whether or not to register after confirming the state of the face image registered as his / her own face image. The face image candidate 153C shown in FIG. 15 (D) is imaged with a higher illuminance than when the face image registered in the face authentication server 50 is imaged. Therefore, the edge component is slightly small.

In the case of FIG. 15, since the candidate 153C is displayed, the person in charge of login can specifically determine whether or not to additionally register the same face image as his / her own face image. As a result, it is possible to avoid registration of a face image having poor image quality.
The cause of the failure of face recognition is not limited to the case of high illuminance, but may be the case of low illuminance. It is also possible that the person in charge of login did not correctly face the surveillance camera 20.

FIG. 16 is a diagram illustrating another transition example of the login screen in the second embodiment. (A) is an example of a screen displayed during execution of face authentication, (B) is an example of a screen displayed when face authentication fails, and (C) is selected to log in by entering a number. (D) and (E) are examples of screens displayed when login is successful, and (F) is an example of menu screens.
In FIG. 16, a reference numeral corresponding to a portion corresponding to that in FIG. 15 is added.

FIG. 16 (D) differs from FIG. 15 (D) in that the candidate face image 153C to be added displayed on the screen 153 displayed when the login is successful is different from the person in charge of login. .. Since the surveillance camera 20 captures a wide range, there is a possibility that a third party other than the login person is reflected in the captured image, and this third party may be mistakenly detected as the face image of the login person. Because there is.
Also in the case of this embodiment, when the faces of a plurality of people are reflected in the captured image, the candidates for registration are not uniquely determined, and the face images of the plurality of people are the candidates for registration 153C.
Therefore, the person in charge of login selects the "No" button 153B in FIG. 16 (D), and selects the "Yes" button 153A in FIG. 16 (E). As a result, it is possible to prevent a situation in which a face image of another person is mistakenly additionally registered.

<Processing operation to eliminate inappropriate additional registration>
Subsequently, the processing operation for eliminating the additional registration of the inappropriate face image will be described.
<Processing operation 1>
FIG. 17 is a flowchart illustrating an example of a processing operation executed by the control unit 500 (see FIG. 4) of the face authentication server 50 (see FIG. 1) when an additional registration of the face image is instructed by the login person. ..

First, the control unit 500 acquires a face image to be additionally registered (step 21). By the way, step 21 is executed when the button 153A is operated in FIG. 14 (D), the button 153A is operated in FIG. 15 (D), and the button 153A is operated in FIG. 16 (E). The button.
Next, the control unit 500 acquires the registered face image associated with the person in charge of login from the face image DB 51 (see FIG. 4) (step 22).

Next, the control unit 500 calculates the matching rate between the face image selected as the target of additional registration and the registered face image (step 23). In this processing operation, the face image when the face authentication fails and the registered face image are collated again.
The match rate here is a value indicating the degree of match as a face image and the degree of possibility of being the same person, and is calculated in the process of collation.
When the match rate is calculated, the control unit 500 determines whether or not the match rate is larger than the reference value (step 24). The reference value may be the same as the reference value used for face recognition, but may be different. For example, a reference value lower than that of face recognition may be used on the premise that another person can be clearly excluded.

If an affirmative result is obtained in step 24, the control unit 500 additionally registers the face image in the face image DB 51 (step 25). Of course, the face image additionally registered is associated with the person in charge of login.
On the other hand, if a negative result is obtained in step 24, the control unit 500 outputs an alarm (step 26). The negative result is output when the face image of another person is the target of additional registration as in the example of FIG. 16D. In this case, the additional registration is not executed and an alarm is output. As a result, it is possible to avoid an erroneous registration of the person in charge of login or an erroneous registration caused by intention.

<Processing operation 2>
FIG. 18 is a flowchart illustrating another example of the processing operation executed by the control unit 500 (see FIG. 4) of the face authentication server 50 (see FIG. 1) when the login person is instructed to additionally register the face image. Is. FIG. 18 is shown with reference numerals corresponding to the portions corresponding to those in FIG.

Also in the case of FIG. 18, the control unit 500 first acquires the face image to be additionally registered (step 21).
However, in FIG. 18, the process executed after step 21 is different from that in FIG. In the case of FIG. 18, the control unit 500 reacquires the face image of the person in charge of login (step 22A). In this processing operation, the face image when the face authentication fails and the re-acquired face image are collated.

Next, the control unit 500 calculates the matching rate between the face image to be additionally registered and the re-acquired face image (step 23A).
Subsequent operations are the same as in FIG.
That is, when the matching rate is calculated, the control unit 500 determines whether or not the matching rate is larger than the reference value (step 24).
If an affirmative result is obtained in step 24, the control unit 500 additionally registers the face image in the face image DB 51 (step 25). Of course, the face image additionally registered is associated with the person in charge of login.

On the other hand, if a negative result is obtained in step 24, the control unit 500 outputs an alarm (step 26). As a negative result, it may occur when the face image of another person is the target of additional registration as in the example of FIG. 16 (D). In this case, the additional registration is not executed and an alarm is output. In this processing operation, it is possible to avoid erroneous registration due to the error of the person in charge of login.

<Embodiment 3>
In this embodiment, other processing operations executed by the face recognition server 50 (see FIG. 1) constituting the money processing system 1 (see FIG. 1) will be described.
FIG. 19 is a flowchart illustrating an example of a processing operation executed by the face authentication server 50 used in the third embodiment. FIG. 19 is shown with reference numerals corresponding to the portions corresponding to those in FIG.

In the first embodiment described above, the execution of 1: 1 authentication is permitted regardless of the number of people included in the captured image, but in the present embodiment, the execution of 1: 1 authentication is performed according to the number of people included in the captured image. Switch whether or not to enable.
Therefore, when a positive result is obtained in step 4, the control unit 500 in the present embodiment sets 1: N authentication to be executable, while 1: 1 authentication is set to non-executable (step 5A). ..
Incidentally, the operation of the control unit 500 when a negative result is obtained in step 4 is the same as that of the first embodiment. That is, the control unit 500 sets 1: N authentication to be non-executable, while 1: 1 authentication is set to be executable (step 10).

In the case of the present embodiment, the operation display unit 14 (see FIG. 3) of the balancer 10 (see FIG. 1) has a login screen 140 (see FIG. 7) and a login screen 141 (see FIG. 7) according to the number of people included in the captured image. Either one of (see FIG. 9) is displayed.
In the case of this embodiment, the login screen 140 shown in FIG. 11 and the login screen 140 shown in FIG. 12 are not displayed on the operation display unit 14. This is because in the case of this embodiment, 1: N authentication and 1: 1 authentication are permitted only exclusively.

<Embodiment 4>
In this embodiment, another system configuration of the money processing system 1 (see FIG. 1) will be described.
FIG. 20 is a diagram illustrating a configuration example of the money processing system 1A used in the fourth embodiment. In FIG. 20, a reference numeral corresponding to a portion corresponding to that in FIG. 1 is added.

The money processing system 1A shown in FIG. 20 uses the camera 20B attached to the balancer 10A instead of the surveillance camera 20 (see FIG. 1). In the case of FIG. 20, the camera 20B is provided near the top of the support column 20C. However, the camera 20B may be attached to a peripheral portion or the like of the operation display unit 14. Further, the camera 20B may be integrated with the main body of the balancer 10A.

The camera 20B in the present embodiment may be attached at least at a position where the face image of the person in charge of login can be captured.
However, the camera 20B may be capable of taking an image at the hand of the operator who operates the balancer 10A. The camera 20B is also an example of the image pickup unit.

In the case of FIG. 20, only one camera 20B is attached behind the operation display unit 14 on the bill processing unit 12 side, but only one camera 20B is attached behind the operation display unit 14 on the coin processing unit 11 side. It may be mounted in both positions.
Further, one camera 20B may be arranged at an intermediate position between the operation display unit 14 on the coin processing unit 11 side and the operation display unit 14 on the bill processing unit 12.

In the case of the money processing system 1A shown in FIG. 20, the captured image 70A of the camera 20B and the transaction data 70B of the cashier 10A are stored in a common data management server 70.
Incidentally, the captured image captured by the camera 20B may be transmitted to the data management server 70 via the cashier 10A and the network 60, or may be directly transmitted to the data management server 70 via the network 60.

<Embodiment 5>
Subsequently, another system configuration of the money processing system 1 (see FIG. 1) will be described.
FIG. 21 is a diagram illustrating a configuration example of the money processing system 1B used in the fifth embodiment. 21 is shown with a reference numeral corresponding to the portion corresponding to FIG. 20.
In the money processing system 1B shown in FIG. 21, the function of the face authentication server 50 (see FIG. 20) is provided in the balance machine 10B.

FIG. 22 is a block diagram showing a functional configuration example of the balance machine 10B used in the fifth embodiment. 22 is shown with reference numerals corresponding to the corresponding portions of FIGS. 3 and 4.
In the case of this embodiment, the face image DB 51 and the face authentication history DB 52 are stored in the storage unit 16.
Further, the control unit 100B realizes the function of the control unit 100 (see FIG. 3) and the function of the control unit 500 (see FIG. 4) in the first embodiment through the execution of the program. In the case of FIG. 22, the collation result by the collation unit 504 is directly output to the login permission unit 101.
The balance machine 10B in the present embodiment is an example of a login system and an example of an electronic device.

<Embodiment 6>
<System configuration>
Hereinafter, other system configurations of the money processing system 1 (see FIG. 1) will be described.
FIG. 23 is a diagram illustrating a configuration example of the money processing system 1C used in the sixth embodiment. In FIG. 23, reference numerals corresponding to the portions corresponding to those in FIG. 1 are added.
A mistaking monitoring server 80 is added to the money processing system 1C shown in FIG. 23. The mistaking monitoring server 80 monitors the behavior of the worker who operates the cashier 10 and provides a function of immediately notifying the worker when a mistaking of money occurs.

Specifically, the mistake monitoring server 80 acquires transaction data from the transaction management server 30, specifies the withdrawal port to which each worker should take out money through the transaction data, and originally from the specified withdrawal port. The operation of monitoring whether or not the worker has taken out the money is executed through the captured image read from the camera image management server 40 (see FIG. 1).
Then, when money is taken out from the wrong withdrawal port, the mistake monitoring server 80 notifies the target person or the like of the occurrence of the mistake through an alarm sound or a warning display.

The mistaking monitoring server 80 also has a computer as a basic configuration, like other servers. In addition, the function of notifying the mistaking of money is realized through the execution of a program by a computer.
The surveillance camera 20 in the present embodiment is attached at a position where the operator can take an image of from which withdrawal port of the cashier 10 the money is to be taken out. Therefore, a surveillance camera (not shown) for surveillance may be installed separately from the surveillance camera 20.

FIG. 24 is a diagram illustrating an example of a withdrawal slot provided in the balance machine 10. In FIG. 24, reference numerals corresponding to the portions corresponding to those in FIG. 2 are added.
In the case of the accounting machine 10 shown in FIG. 24, the coin processing unit 11 is provided with six withdrawals and the bill processing unit 12 is provided with four withdrawals, for a total of ten withdrawals. Money is discharged to these withdrawals according to the transaction being executed.
In addition, the balance machine 10 is also provided with a deposit slot used for depositing money.

<Processing operation>
FIG. 25 is a flowchart illustrating an example of a processing operation executed by the mistaking monitoring server 80. The symbol S shown in the figure means a step.
The mistaking monitoring server 80 determines whether or not the user has logged in to the checkerboard 10 (step 31). Whether or not there is a login is determined by cooperation with the face authentication server 50.

While the negative result is obtained in step 31, the mistake monitoring server 80 repeats the determination in step 31. The determination in step 31 may be executed in a predetermined cycle (for example, 10 seconds).
It should be noted that the mistaking monitoring server 80 may be configured to receive a notification of the occurrence of login from the accounting machine 10 or the face authentication server 50, etc., instead of inquiring the occurrence of login to the accounting machine 10 or the face authentication server 50 or the like.

If an affirmative result is obtained in step 31, the mistake monitoring server 80 starts monitoring and tracking of the worker by the captured image (step 32). Tracking here refers to tracking the worker. The worker who instructed the operation display unit 14 to make a transaction moves to the position of the withdrawal port corresponding to the transaction in order to take out the withdrawn money. The movement to this withdrawal port and the movement of the worker's hand are tracked. The workers may be distinguished by the difference in the facial image or the difference in the skeleton.

Next, the mistake monitoring server 80 identifies a withdrawal port from which money related to the worker's transaction is withdrawn (step 33).
Subsequently, the mistake monitoring server 80 determines whether or not the worker has withdrawn money from the wrong withdrawal port (step 34).
If an affirmative result is obtained in step 34, the mistaking monitoring server 80 instructs the balancer 10 to output an alarm (step 35). The output of the alarm here is executed by any or a combination of sound, voice, and image.

On the other hand, if a negative result is obtained in step 34, the mistaking monitoring server 80 ends the process as it is.
In the flowchart shown in FIG. 25, it is assumed that there is one withdrawal port for withdrawing money for one transaction, but money is withdrawn from a plurality of withdrawal ports for one transaction. If so, the determination in step 34 is executed for all withdrawal ports.

<Specific example>
Hereinafter, an example of notification of a mistake by the mistake monitoring server 80 will be described with reference to FIGS. 26 to 28.
FIG. 26 is a diagram illustrating a scene in which two workers simultaneously use one balance machine 10. In the case of FIG. 26, two people, Mr. A and Mr. B, are using the balance machine 10 at the same time.

FIG. 27 is a diagram illustrating a withdrawal port related to Mr. A's transaction and a withdrawal port related to Mr. B's transaction. (A) represents the position of the withdrawal port related to Mr. A's transaction, and (B) represents the position of the withdrawal port related to Mr. B's transaction.
In the case of FIG. 27, the withdrawal ports related to Mr. A's transaction are the five withdrawal ports provided in the coin processing unit 11 and the one withdrawal port provided in the banknote processing unit 12. There are a total of six. On the other hand, there are a total of four withdrawal ports related to Mr. B's transaction, one withdrawal port provided in the coin processing unit 11 and three withdrawal ports provided in the banknote processing unit 12. Is.

FIG. 28 is a diagram illustrating an example of an alarm output when an error occurs. In FIG. 28, the position of the withdrawal port from which Mr. A should take out money is surrounded by a dotted line. In the case of FIG. 28, the withdrawal port surrounded by the dotted line exists in the right middle stage of the coin processing unit 11.
However, it is possible that Mr. A takes out money from the withdrawal port at the lower right of the coin processing unit 11. In this case, the mistaking monitoring server 80 (not shown) displays on the operation display unit 14 that Mr. A has taken out the money from the wrong withdrawal port.

In the example of FIG. 28, a misunderstanding alert is displayed on the operation display unit 14 on the coin processing unit 11 side operated by Mr. A. For example, it is possible to notify the difference between the removal of a new bill and the removal of an old bill.
In the example of FIG. 28, the operation requested by Mr. A is also displayed, saying, "It is taken out from the wrong withdrawal port. Please give it to Mr. B." Although not shown, the alarm screen may be turned off when Mr. B inputs the identification information and then inputs the confirmation of receipt of money from Mr. A.

In the present embodiment, the cashier 10 is illustrated as an example of a money processing device provided with a plurality of withdrawal ports, but it can also be applied to other deposit / withdrawal machines.
Further, in the present embodiment, an error in the withdrawal slot is notified by an alarm, but the present invention is not limited to this. For example, an alarm screen may be displayed even when money taken out from the machine due to a jam error or the like is mistakenly returned to the deposit slot even though the error cash return process has not been executed.

<Embodiment 7>
In this embodiment, a medical expense payment machine, which is an example of a money processing device, will be described.
FIG. 29 is a diagram illustrating a configuration example of the money processing system 1D used in the seventh embodiment. The money processing system 1D shown in FIG. 29 includes a medical expense payment machine 90, a medical accounting system 1000, and a network 1010.
The money processing system 1D shown in FIG. 29 is different from the other systems described above in that it is installed in hospitals, clinics, dispensing pharmacies, and the like.

The medical expense payment machine 90 used in the present embodiment is a self-operated machine operated by a patient. The medical expense payment machine 90 is provided with information on medical expenses and drug charges from the medical accounting system 1000.
When the time elapsed after receiving the operation exceeds a certain period of time, assuming that the operator disappears in the middle of the operation or the operator who is unfamiliar with the operation is confused by the medical expense payment machine 90. It has an operation timer function to decide whether to cancel the transaction being processed or to continue it while it is still valid.

Currently, the optimum value is used for a certain period of time managed by the operation timer function without knowing the optimum value, but in this embodiment, machine learning is performed by linking information such as the patient's age, gender, and department to be examined with the working time. By doing so, the above-mentioned function of automatically adjusting the fixed time for each patient is added. The patient's age, gender, department, medical expenses, drug charges, etc. are included in the medical data.
FIG. 30 is a diagram illustrating a functional configuration example of the medical expense payment machine 90 used in the seventh embodiment.

As shown in FIG. 30, the medical expense payment machine 90 is provided with a control unit 900 that controls the movement of the entire device.
The control unit 900 gives a command signal to the coin processing unit 91 and the bill processing unit 92, and controls their operations. Here, the coin processing unit 91 and the bill processing unit 92 correspond to the coin processing unit 11 (see FIG. 3) and the bill processing unit 12 (see FIG. 3), respectively.

Further, the operation display unit 93, the printing unit 94, the storage unit 95, the card reader 96, the motion sensor 97, and the communication interface unit 98 are connected to the control unit 900, respectively.
The operation display unit 93 is composed of a touch panel or the like. The printing unit 94 is composed of, for example, a printer. The storage unit 95 is composed of a hard disk device or the like.
The storage unit 95 contains medical data 95A, a database 95B that records the time required for transactions linked to medical data, and a learning model that learns the relationship between the patient's age, gender, medical examination department, and the time required for transactions. 95C is stored.

The card reader reads data from a card-like medium such as a medical examination ticket, a credit card, or a cash card.
The motion sensor 97 is arranged in front of the medical expense payment machine 90 and detects the presence of a patient during operation.
The communication interface unit 98 is a communication module that enables communication via the network 1010 (see FIG. 29).

The control unit 900 includes a CPU, a ROM, and a RAM. That is, the control unit 900 constitutes a computer. The ROM stores BIOS, various settings, and the like. The CPU uses the RAM as a work area to execute the program.
The control unit 900 functions as a medical data acquisition unit 901, an operation time measurement unit 902, an operation time learning unit 903, and a timer time adjustment unit 904 through the execution of a program by the CPU.

The medical data acquisition unit 901 is a processing unit that acquires medical data corresponding to the medical examination ticket read by the card reader 96 from the medical accounting system 1000.
The operation time measurement unit 902 is a processing unit that measures the time required from the time when the medical examination ticket is read to the end of the settlement.
The operation time learning unit 903 is a processing unit that machine-learns the relationship between the combination of age, gender, and the department to be examined and the measured operation time. By machine learning by the operation time learning unit 903, a learning model 95C that outputs the operation time when a combination of age, gender, and consultation department is input is generated.

When the medical examination ticket is read by the card reader 96, the timer time adjustment unit 904 reads out the age, gender, and medical examination department of the medical data acquired for the read patient and gives it to the learning model 95C. It is a processing unit that obtains the expected operation time and adjusts the timer time according to the obtained operation time.
By adjusting the timer time by the timer time adjusting unit 904, the optimum timer time is set for each patient. For example, the timer time of a young person is set shorter than the timer time of an elderly person.

If the timer time is machine-learned for each medical expense payer 90, it takes a long time to improve the accuracy of the generated learning model. Therefore, the learning effect of the learning model 95C is enhanced by aggregating the information of the plurality of medical expense payment machines 90 installed in hospitals and the like, that is, the information of the operation time linked to the medical data in the cloud server. May be good.
Further, the data aggregation may be executed not only in the case of a hospital or the like as a unit, but also in the case of an affiliated hospital or the like as a unit, or may be executed regardless of the affiliated hospital or the like.

<Other embodiments>
Although the embodiments of the present invention have been described above, the technical scope of the present invention is not limited to the scope described in the above-described embodiments. It is clear from the description of the claims that the above-mentioned embodiments with various modifications or improvements are also included in the technical scope of the present invention.

In the case of the above-described embodiment, login to the money processing device is assumed as the login system, but it can also be used to manage login to the device that manages keys and important items. A device that manages keys and important items is an example of an electronic device.

1, 1A, 1B, 1C, 1D ... Money processing system, 10, 10A, 10B ... Treasurer, 11 ... Coin processing unit, 12 ... Bill processing unit, 13 ... Money storage unit, 14 ... Operation display unit, 20 ... Monitoring Camera, 20B ... Camera, 30 ... Transaction management server, 40 ... Camera image management server, 50 ... Face recognition server, 60, 1010 ... Network, 70 ... Data management server, 80 ... Mistake monitoring server, 90 ... Medical expense payment machine , 1000 ... Medical accounting system

Claims (22)

A login system that performs face recognition based on the face image obtained from the captured image.
A login system including a control unit that controls whether or not 1: N authentication can be executed based on the number of people information detected from the captured image. An acquisition unit that acquires face information from captured images,
A storage unit that stores facial information of multiple people,
A collation unit that collates the face information acquired by the acquisition unit with the face information stored in the storage unit by 1: 1 authentication or 1: N authentication.
The permission unit that permits login based on the verification result by the verification unit, and the permission unit.
The login system according to claim 1. The control unit uses the number of faces acquired from the captured image as the number of people information.
The login system according to claim 1 or 2. The control unit uses the number of skeletons acquired from the captured image as the number of people information.
The login system according to claim 1 or 2. With more display
The control unit switches the display mode on the display unit depending on whether the number of people information is a plurality of people or one person.
The login system according to claim 1 or 2. The control unit
When the number of people information is a plurality of people, the screen for 1: 1 authentication is displayed on the display unit in the first size.
When the number of people information is one person, the screen for 1: 1 authentication is displayed on the display unit by the second size, which is smaller than the first size.
The login system according to claim 5. The control unit
When the number of people information is a plurality of people, a screen for 1: 1 authentication is displayed on the display unit.
When the number of people information is one person, the screen for 1: 1 authentication is not displayed on the display unit.
The login system according to claim 5. When the number of people information is one, the control unit displays a button for calling the screen for 1: 1 authentication on the display unit.
The login system according to claim 7. When the number of people information is one, the control unit instructs 1: N authentication to be executed.
The login system according to claim 1 or 2. With more display
When the control unit additionally registers a face image to be used for authentication, the control unit displays candidates for the face image to be added on the display unit.
The login system according to claim 1 or 2. The control unit determines a candidate selected by the user from the candidates displayed on the display unit as a registration target.
The login system according to claim 10. The control unit
When the number of people information is a plurality of people, the candidates for the face image to be added are displayed on the display unit.
When the number of people information is one person, the candidate of the face image to be added is not displayed on the display unit.
The login system according to claim 10. When the face image used for authentication is additionally registered, the control unit calculates the matching rate between the added face image and the face image newly acquired from the user who is logging in, and the calculated matching is performed. Output an alarm when the rate is lower than the predetermined standard,
The login system according to claim 10. When the face image used for authentication is additionally registered, the control unit calculates the matching rate between the added face image and the face image already registered for the user associated with the face image. When the calculated match rate is lower than the predetermined standard, an alarm is output.
The login system according to claim 10. The control unit records as a history whether the user has logged in with 1: N authentication or logged in with 1: 1 authentication.
The login system according to claim 1 or 2. The control unit always allows the execution of 1: 1 authentication.
The login system according to claim 1 or 2. When the number of people information is one, the control unit cannot execute 1: 1 authentication.
The login system according to claim 1 or 2. Further provided with an imaging unit that captures the peripheral area of the electronic device to be logged in as the captured image.
The login system according to any one of claims 1 to 17. The image pickup unit is a surveillance camera that monitors the inside of a facility where the electronic device is provided.
A first server that stores the processing data of the electronic device,
A second server different from the first server, which stores the processing data of the surveillance camera, is further provided.
The login system according to claim 18. The image pickup unit is a camera provided or attached to the electronic device.
Further provided with a server for storing the processing data of the electronic device and the processing data of the camera.
The login system according to claim 18. An electronic device that requires face recognition using a face image obtained from a captured image at login.
An electronic device including a control unit that controls whether or not 1: N authentication can be executed based on the number of people information detected from the captured image. It is a login control method in a computer.
When face authentication using the face image acquired from the captured image is required when logging in to the electronic device.
A login control method having a process of controlling whether or not 1: N authentication can be executed based on the number of people information detected from the captured image.

Images