JP2018173891A - Authentication device, authentication method, authentication program, and database - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent spoofing with photos and movies.SOLUTION: A database registers, for a plurality of users to be authenticated, a plurality of face reference images, a plurality of gesture types, and data representing a plurality of gesture operations, which are association with each other in a one-to-one relationship. An authentication device receives a face input image, collates the face input image with the face reference images registered in the database. If the face input image is registered, the authentication device displays a gesture selection screen for prompting to select one of a plurality of gestures that can be chosen. If the selected one gesture matches with a particular gesture type, the authentication device prompts a person to perform a gesture, recognizes the gesture operation performed by the person, and determines whether the recognized gesture operation matches with data of a specific gesture operation associated with a specific face authentication image.SELECTED DRAWING: Figure 3

Description

  The present invention relates to an authentication apparatus, an authentication method, an authentication program, and a database for authenticating an individual (person).

  Various methods are used as a method for authenticating an individual (person). As an authentication method, there are an authentication method using a several-digit password or a password and a pattern authentication method. Biometric authentication methods are also known as other authentication methods. This biometric authentication method is an authentication method that is robust against risks such as leakage, forgetting, and impersonation.

  The biometric authentication method is also called a biometric authentication method, and is a method of performing personal (personal) authentication using human biometric features. Even if it is said to be a biological feature, since there are various features in humans, the method is divided according to which feature is used.

  Biometric authentication methods are roughly classified into static authentication methods and dynamic feature authentication methods. Static features include human fingerprints, veins, faces, and irises. Dynamic features include voiceprints, gaits, signatures, gestures and the like. Hereinafter, an authentication method using static features is also abbreviated as “static authentication method”, and an authentication method using dynamic features is also abbreviated as “dynamic authentication method”.

  There are many practical examples of static authentication methods. However, there is a decision that fingerprints, veins, etc. are vulnerable to dirt, trauma, and dryness. Further, the face authentication method also has a problem that it can be collated by enlarging and displaying the registrant's face on the display (for example, using a photograph of the registrant). In addition, since the physical characteristics of a person cannot be changed, there is a risk that the characteristics cannot be used once stolen.

  On the other hand, in the dynamic authentication method, since personal authentication is performed using a personal bag, it is difficult to imitate even when a pattern is stolen.

  As described above, a face authentication system using images has been put into practical use, but a safe and secure system that further improves the security of personal authentication is expected.

  Various authentication systems of this type have been proposed.

  For example, Patent Document 1 discloses a “biometric collation apparatus” that performs biometric collation without directly touching the apparatus and prevents impersonation by taking a photograph, and is simpler and safer. The biometric matching device disclosed in Patent Literature 1 prescribes a specific action of closing and opening eyes for a predetermined time, and performs face image matching on the condition that this specific action is detected.

  Japanese Patent Laid-Open No. 2004-228688 later checks whether the person to be authenticated is a living body face or a pseudo face such as a photograph and prevents impersonation while confirming the person to be impersonated later. A “face recognition device” that can be used is disclosed. The face authentication apparatus disclosed in Patent Document 2 includes a holding unit, an imaging camera, an A / D conversion unit, an authentication unit, a biometric identification unit, a determination result output unit, and a voice generation unit. The holding unit registers and holds the face image of the authorized person as a template image in advance. The authentication unit performs template matching and determines whether or not the person of the extracted face image is the same as any registered person. The biometric identification means determines whether or not the face captured using the A / D converted face images of a plurality of frames is a living organism. The determination result output means outputs a determination result that permits entry of the person only when the authentication means authenticates that the person is a registered person and the biometric identification means identifies that it is a living body. The sound generation means requests the subject to be blinked or open / close the mouth with a sound registered in advance.

  Patent Document 2 describes a combination of eyes and mouth and a combination of mouth and eyebrows (lifting eyebrows or bringing eyebrows close) as a combination of two portions that can be easily moved on the face. Moreover, in patent document 2, the audio | voice which requests | requires blink or opening / closing of a mouth is uttered by the audio | voice generation means.

  Patent Document 3 discloses an “information processing apparatus” that performs efficient user authentication. Patent Document 3 describes three types of login processes 1 to 3. Among these, the login process 3 performs user authentication by detecting that the user's face has been moved to a predetermined area after the information processing apparatus has performed user face authentication using a captured image of the camera. Yes.

  Patent Document 4 discloses an “authentication apparatus and method” that enables individual users to perform authentication using the authentication procedure provided by the authentication environment as it is in a system shared by a plurality of users having different authentication environments. Disclosure. The authentication system disclosed in Patent Literature 4 includes an authentication device and an authentication device group. The authentication device includes an authentication device control unit, a rule storage unit, a rule execution unit, an authentication determination unit, an authentication result notification unit, and the like. The authentication device group includes, for example, authentication devices such as an IC card reader / writer, a magnetic card radar writer, a key input device, and a fingerprint reader.

  In Patent Document 4, the authentication device acquires personal identification information. The rule storage unit stores a rule defining a process to be performed after acquisition of personal identification information. The rule execution unit reads a rule from the rule storage unit, determines a rule to be used, and executes the rule. The authentication device control unit starts and stops the authentication device according to an instruction from the rule execution unit, and notifies the rule execution unit of the personal identification information and the authentication device that has acquired the personal identification information. The authentication determination unit confirms the validity of the user specified by the personal identification information. The authentication result notifying unit notifies the result of the rule execution by the rule executing unit (success or failure of authentication) to a service that requires authentication.

  In Patent Document 4, a means for acquiring authentication procedure information from an external device and storing it in a storage device may be further provided. For example, a template of procedure information (rule) is prepared, and the procedure information can be selectively adopted according to the installation environment. The external device may include a web server that displays and selects a list of authentication procedures.

JP 2002-251614 A JP 2006-330936 A Japanese Patent Laying-Open No. 2015-090661 JP 2006-072468 A

  However, Patent Documents 1 to 4 have the following problems.

  In Patent Document 1, face images are collated under the condition (trigger) of detection of a specific action. This specific operation is common to users. Therefore, once the specific operation is stolen, it becomes difficult to prevent impersonation.

  Japanese Patent Application Laid-Open No. 2004-228561 requests the subject to be blinked or to open / close the mouth with a previously registered voice. This request is common to users. Therefore, it becomes possible to impersonate a malicious person by using a moving image that is a person impersonating and performing the requested action.

  In Patent Literature 3, user authentication is performed using a captured image of a camera, and then user authentication is performed by detecting that the user's face has been moved to a predetermined area. Moving to a predetermined area is common to users. Therefore, in Patent Document 3, it is possible to impersonate by moving a registrant's face photograph to a predetermined area.

  Patent Document 4 merely discloses a technique that allows authentication means to be changed according to the installation environment by selectively adopting authentication procedure information (rules) according to the installation environment.

  The objective of this invention is providing the authentication apparatus, the authentication method, authentication program, and database which solve the subject mentioned above.

  An authentication apparatus according to the present invention is an authentication apparatus that performs personal authentication of a person, and corresponds to a plurality of users to be authenticated, respectively, a plurality of face reference images, a plurality of gesture types, and a plurality of gesture operation data. A one-to-one database for registration, an input unit that receives a face input image of the person's face photographed by a camera, the face input image, and the plurality of face reference images registered in the database And when the face authentication unit determines that a specific face authentication image matched with the face input image is registered in the database, the face authentication unit performs face authentication. Displaying a gesture selection screen describing a plurality of gesture options included in the gesture type, and selecting one of the plurality of gesture options for the person. A type selection unit that determines whether or not one gesture option selected by the person matches a specific gesture type associated with the specific face authentication image; When it is determined that one gesture option matches the specific gesture type, the gesture prompting unit that urges the person to make a gesture, and the gesture operation performed by the person are recognized and recognized. A gesture motion recognition / authentication unit that determines whether or not a gesture motion matches specific gesture motion data associated with the specific face authentication image, and the recognized gesture motion is the specific gesture motion. When matching with data, it is determined that the person is the person himself / herself.

  The authentication method of the present invention is an authentication method for performing personal authentication of a person, and a plurality of face reference images, a plurality of gesture types, and a plurality of gestures corresponding to a plurality of users to be authenticated in a database. Gesture motion data is pre-registered in a one-to-one correspondence, accepts a face input image of the person's face photographed by a camera, the face input image and the plurality of face reference images registered in the database Face recognition is performed, and when it is determined that a specific face authentication image matched with the face input image is registered in the database, a plurality of gesture options included in the plurality of gesture types are selected. The described gesture selection screen is displayed to prompt the person to select one of the plurality of gesture options, and one gesture selected by the person If a gesture option matches a specific gesture type associated with the specific face authentication image, and it is determined that the selected gesture option matches the specific gesture type , Prompting the person to make a gesture, recognizing the gesture action performed by the person, and whether the recognized gesture action matches the specific gesture action data associated with the specific face authentication image Judge whether or not.

  An authentication program according to the present invention is an authentication program that causes a computer to authenticate a person. The database includes a plurality of face reference images and a plurality of face reference images corresponding to a plurality of users to be authenticated. Gesture types and a plurality of gesture operation data are preliminarily registered in a one-to-one relationship, and the authentication program includes an input procedure for receiving a face input image of the person's face photographed by the camera on the computer. A face authentication procedure for performing face authentication by comparing the face input image with the plurality of face reference images registered in the database, and collating the face input image with the database in the face authentication procedure. Describe a plurality of gesture options included in the plurality of gesture types when it is determined that a specific face authentication image is registered The gesture selection screen is displayed to prompt the person to select one of the plurality of gesture options, and the one gesture option selected by the person is associated with the specific face authentication image. A type selection procedure for determining whether or not a particular gesture type matches, and if it is determined in the type selection procedure that the selected one gesture option matches the specific gesture type, A gesture prompting procedure for urging the user to make a gesture, and a gesture action performed by the person is recognized, and whether the recognized gesture action matches the specific gesture action data associated with the specific face authentication image. And a gesture motion recognition / authentication procedure for determining whether or not.

  The database of the present invention is formed by registering a plurality of face reference images, a plurality of gesture types, and a plurality of gesture operation data in a one-to-one correspondence with each other, corresponding to a plurality of users to be authenticated.

  According to the present invention, it is possible to prevent impersonation with a photograph or a moving image.

It is an external appearance perspective view which shows the hardware constitutions of the authentication system which concerns on one Embodiment of this invention. It is a block diagram which shows the structure of the authentication system shown by FIG. It is a block diagram which shows in detail the internal structure of the authentication apparatus used for the authentication system shown in FIG. It is a figure which shows the database stored in the memory | storage device used for the authentication apparatus shown in FIG. It is a flowchart for demonstrating operation | movement of the authentication system shown in FIG. It is a figure which shows an example of the gesture selection screen displayed on the display apparatus used for the authentication system shown by FIG.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  With reference to FIG. 1 thru | or FIG. 3, the authentication system which concerns on one Embodiment of this invention is demonstrated. FIG. 1 is an external perspective view showing the hardware configuration of the authentication system. FIG. 2 is a block diagram showing the configuration of the authentication system. FIG. 3 is a block diagram showing in detail the internal configuration of the authentication device 20 used in the authentication system shown in FIG.

  The authentication system includes a camera 10 for photographing a user, an authentication device 20 for authenticating a person, a display device 30 having a screen for guiding the user, and an input device for inputting information. 40.

  As the information input port of the camera 10, for example, a USB (Universal Serial Bus) can be used. In addition, the information input port may use another general camera interface such as MIPI-CSI (Mobile Industry Processor Interface-Camera Serial Interface).

  The display device 30 includes, for example, a liquid crystal display (LCD). The image information output port for the LCD can use, for example, an LVDS (Low Voltage Differential Signaling) interface. In addition, the image information output port may use another interface such as Display Port.

  The input device 40 can use a touch panel, for example. The input device 40 may be a keyboard or a mouse.

  In the illustrated example, the display device 30 and the input device 40 are shown as separate devices, but they may be integrated. For example, the display device 30 may be a display device with a touch panel.

  As shown in FIG. 2, the authentication device 20 includes a central processing unit (CPU) 50, a storage device 60, a program memory 70, and a work memory 80.

  The CPU 50 controls the camera 10, the input device 40, and the display device 40 to receive data, perform calculation using the data, store it, and output the result.

  The storage device 60 is composed of, for example, a hard disk drive (HDD). The storage device 60 may comprise a solid state drive (SSD). The storage device 60 stores a database to be described later.

  The work memory 80 is composed of, for example, a RAM (random access memory). The work memory 80 temporarily stores the processing result in the CPU 50.

  The program memory 70 is composed of, for example, a ROM (read only memory) or a RAM. The program memory 70 stores an authentication program described later.

  FIG. 4 is a diagram illustrating a database stored in the storage device 60.

  As is clear from FIG. 4, the database stored in the storage device 60 corresponds to a plurality of users to be authenticated (for each user ID (identification number)), and each includes a plurality of face reference images, A plurality of gesture types and a plurality of gesture motion data are registered in one-to-one correspondence.

  In the present embodiment, it is assumed that the number of gesture types registered in the database 60 is N (N is an integer of 3 or more). Specifically, N is equal to 20.

  The gesture type is represented by, for example, a name that distinguishes the type of gesture. Specifically, when N = 20 gesture types are defined, 20 gesture names are given accordingly. Specifically, gesture types (names) include “blink”, “close right eye”, “close left eye”, “raise right hand”, “raise left hand”, “raise neck”, etc. Get it. As the gesture type, when the user registers in the database, any one of the 20 gesture types is selected. Therefore, the selected gesture type is information that is normally only known by the user.

  The gesture motion data is data representing the characteristics of the gesture motion corresponding to each gesture type. This gesture motion data is data obtained by previously analyzing an image (gesture motion image) captured by a camera using a software image analysis system and recognizing the gesture motion. is there.

  Here, although the specific gesture operation itself is actually different for each individual, the gesture operation as a concept is common to users. Therefore, the “gesture operation” in the present embodiment means a gesture operation that can be commonly recognized by users, and does not mean individual gesture operations that are different for each individual. Therefore, in this embodiment, a plurality of gesture types and a plurality of gesture motion data correspond one-to-one.

  FIG. 3 is a block diagram showing functions of the CPU 50. The CPU 50 includes an input unit 10, a face authentication unit 52, a type selection unit 53, a gesture prompting unit 54, and a gesture motion recognition / authentication unit 55.

  The input unit 10 receives a face input image of a person's face photographed by the camera 10.

  The face authentication unit 52 performs face authentication by comparing the face input image with a plurality of face reference images registered in the database 60.

  It is assumed that the face authentication unit 52 determines that a specific face authentication image collated with the face input image is registered in the database 60. In this case, the type selection unit 53 displays on the display device 30 a gesture selection screen that describes M (for example, three) gesture options randomly selected from the N = 20 gesture types. . Accordingly, the type selection unit 53 prompts the person to select one of the displayed plurality of gesture options. Here, it should be noted that one of the M gesture options displayed on the display device 30 to be selected by the person is a specific gesture type associated with the specific face authentication image.

  Then, the type selection unit 53 determines whether or not one selected gesture option matches a specific gesture type associated with the specific face authentication image.

  The gesture prompting unit 54 urges a person to make a gesture when one selected gesture option matches a specific gesture type. For example, the gesture prompting unit 54 displays a “Please do gesture” message on the display device 30. Alternatively, the gesture prompting unit 54 may generate a voice “Please do gesture” from a speaker (not shown).

  The gesture motion recognition / authentication unit 55 recognizes a gesture motion performed by a person and determines whether or not the recognized gesture matches specific gesture motion data associated with the specific face authentication image. . When the recognized gesture action matches the specific gesture action data, the gesture action recognition / authentication unit 55 determines that the person is the person himself / herself.

  As described above, each unit of the authentication device 20 can be realized using a combination of hardware and software. In a form in which hardware and software are combined, an authentication program is stored in the program memory 70, and each unit is realized as various means by operating hardware such as the CPU 50 based on the authentication program. The authentication program may be recorded on a recording medium and distributed. The authentication program recorded on the recording medium is read into the memory via a wired, wireless, or recording medium itself, and operates a control unit or the like. Examples of the recording medium include an optical disk, a magnetic disk, a semiconductor memory device, and a hard disk.

  To describe the above-described embodiment in another expression, a computer (CPU 50) that operates as the authentication apparatus 20 is input based on an authentication program stored in a ROM, an input unit 51, a face authentication unit 52, a type selection unit 53, a gesture. This can be realized by operating as the prompting unit 54 and the gesture motion recognition / authentication unit 55.

  Next, the operation of the authentication system will be described with reference to FIG.

  First, a face image of a person is taken with the camera 10 (step S101). That is, the input unit 51 receives a human face input image.

  Next, the face authentication unit 52 determines whether a face input image is registered in the database 60 (step S102). That is, the face authentication unit 52 performs face authentication by comparing the face input image with a plurality of face reference images registered in the database 60.

  If it is not registered in the database 60 (No in step S102), the face authentication unit 52 determines that the authentication is NG (step S107) and does not perform the subsequent processing.

  On the other hand, when it is determined that a specific face authentication image that is collated with the face input image is registered in the database 60 (Yes in step S102), the type selection unit 53 displays a plurality of gesture options for gesture operation (step S102). S103). That is, the type selection unit 53 displays a gesture selection screen describing M gesture options randomly selected from N gesture types, and selects one of the M gesture options for the person. Encourage

  FIG. 6 shows an example of a gesture selection screen displayed on the display device 30. FIG. 6 shows an example where M is equal to 3.

  In the example of FIG. 6, the type selection unit 53 describes three gesture options randomly selected from the 20 gesture types: “blink”, “raise right hand”, and “raise the neck”. A gesture selection screen is displayed on the display device 30.

  As described above, one of the gesture options to be selected by the person displayed on the display device 30 is a specific gesture type associated with the specific face authentication image. In the following, the description will be continued assuming that the specific gesture type is “raise the right hand”.

  A person who is a user selects a gesture option (gesture type) registered in advance in the database 60 from the gesture options on the gesture selection screen displayed on the display device 30. Specifically, when the display device 30 is a display device with a touch panel, the person selects one of “blink”, “raise his right hand”, and “raise his neck” on the gesture selection screen.

  In the present embodiment, the type selection unit 53 displays the gesture selection screen by randomly switching the display locations of the selected M gesture options for each authentication (every time). The reason for this is as follows.

  Suppose there is a person (a third party) who tries to steal a gesture with malicious intent. In this case, it is difficult for a third party to grasp the gesture type registered by the person by displaying the gesture options at random each time, rather than displaying the gesture options fixedly each time.

  More specifically, when M gesture options are fixedly displayed each time, there is a possibility that the gesture type registered in the database by the third person may be determined by a third party in the following points. Concerned.

Example) A third person sees, "This person is moving his neck every time."
Example) A third party sees, "This person is acting as a gesture option displayed at the top every time."
Therefore, in the present embodiment, the M gesture options are displayed at different display positions at random times, and the third party can grasp the user's own gesture by allowing the user to select the gesture on the spot. Making it more difficult.

  Subsequently, the type selection unit 53 determines whether the selected gesture option matches the specific gesture type associated with the specific face authentication image (step S104). In other words, the type selection unit 53 determines whether or not the association between the specific face authentication image / specific gesture type matches in the database 60.

  When a specific gesture type registered in advance is not selected as a result of collation of the database 60 (specifically, when the person selects a gesture option of “blink” or “raise his neck”) (step The type selection unit 53 determines that the person is not the user and determines that the authentication is NG (step S107), and does not perform the subsequent processing.

  On the other hand, when a specific gesture type registered in advance is selected as a result of the collation of the database 60 (specifically, when the person selects a gesture option of “lifting the right hand”) (Yes in step S104). Subsequently, the gesture prompting unit 54 prompts the person to make a gesture. Specifically, for example, the gesture prompting unit 54 displays a “Please do gesture” message on the display device 30.

  The gesture motion recognition / authentication unit 55 determines whether or not the specific gesture motion is correctly performed (step S106). Specifically, the gesture motion recognition / authentication unit 55 analyzes an image of a gesture motion performed by the person captured by the camera 10 (hereinafter referred to as “gesture motion image”) using an image analysis system. Then, it recognizes what kind of gesture action the gesture action was. Then, the gesture motion recognition / authentication unit 55 determines whether the recognized gesture motion matches the specific gesture motion data associated with the specific face authentication image.

  In this example, the specific gesture type is “raise the right hand”. Accordingly, the gesture motion recognition / authentication unit 55 determines whether or not the gesture motion image recognized using the image analysis system matches the image of the gesture motion of “raising the right hand”.

  When the specific gesture operation is not performed correctly (that is, when the recognized gesture operation image does not match the image of the gesture operation of “lifting the right hand”) (No in step S105), the gesture operation recognition / authentication unit 55 determines that the person is not the person himself, and determines that the person is an authentication NG (step S107), and does not perform the subsequent processing.

  On the other hand, when the specific gesture motion is performed correctly (that is, when the recognized gesture motion image matches the image of the gesture motion of “lifting the right hand”) (Yes in step S105), the gesture motion recognition / authentication unit 55 Determines that the person is the person himself / herself, authentication is OK (step S106), and the continuation of the subsequent processing is permitted.

  As described above, in the authentication method of the present embodiment, personal authentication is performed in three stages: face authentication, selection of a pre-registered gesture type, and gesture operation.

  Next, the effect of this embodiment will be described.

  According to the authentication method of the present embodiment, it is possible to make it more difficult to impersonate a photo or a moving image. The reason is that not only face authentication but also two types of authentication, that is, selection of a pre-registered gesture type and actual gesture operation are added to perform personal authentication.

  The present invention has been described above with reference to the embodiments, but the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

  The present invention can also be used for personal authentication using a personal computer, a smartphone, or the like, or for personal authentication such as credit card payment / ATM (automatic teller's machine).

10 Camera 20 Authentication Device 30 Display Device 40 Input Device 50 CPU
DESCRIPTION OF SYMBOLS 51 Input part 52 Face authentication part 53 Type selection part 54 Gesture prompting part 55 Gesture movement recognition / identification part 60 Storage device (database)
70 Program memory 80 Work memory

Claims (7)

An authentication device for authenticating a person,
Corresponding to a plurality of users to be authenticated, a database for registering a plurality of face reference images, a plurality of gesture types, and a plurality of gesture operation data in a one-to-one correspondence,
An input unit for receiving a face input image of the face of the person photographed by a camera;
A face authentication unit that performs face authentication by comparing the face input image and the plurality of face reference images registered in the database;
In the face authentication unit, when it is determined that a specific face authentication image collated with the face input image is registered in the database, a gesture selection describing a plurality of gesture options included in the plurality of gesture types A screen is displayed to prompt the person to select one of the plurality of gesture options, and the one gesture option selected by the person is a specific gesture associated with the specific face authentication image. A type selection unit for determining whether or not the type matches,
In the type selection unit, when it is determined that the selected one gesture option matches the specific gesture type, a gesture prompting unit that prompts the person to make a gesture;
A gesture motion recognition / authentication unit for recognizing a gesture motion performed by the person and determining whether the recognized gesture motion matches specific gesture motion data associated with the specific face authentication image; ,
With
An authentication apparatus, wherein the person is determined to be the person when the recognized gesture action matches the specific gesture action data. When the number of the plurality of gesture types registered in the database is N,
The type selection unit selects, from the N gesture types, M gesture options that include the specific gesture type and fewer than N, and describes the selected M gesture options. The authentication device according to claim 1, wherein a selection screen is displayed.   The said type selection part selects the said M gesture types at random from the said N gesture types, and displays the gesture selection screen which described the selected M gesture options. Authentication device.   The authentication apparatus according to claim 2, wherein the type selection unit displays the gesture selection screen by randomly switching the display locations of the selected M gesture options for each authentication.   A database formed by associating a plurality of face reference images, a plurality of gesture types, and a plurality of gesture operation data in a one-to-one correspondence with a plurality of users to be authenticated. An authentication method for authenticating a person,
Corresponding to a plurality of users to be authenticated, a plurality of face reference images, a plurality of gesture types, and a plurality of gesture motion data are pre-registered in a one-to-one correspondence in the database,
Accept the face input image of the person's face taken with the camera,
Collating the face input image with the plurality of face reference images registered in the database, and performing face authentication,
When it is determined that a specific face authentication image collated with the face input image is registered in the database, a gesture selection screen describing a plurality of gesture options included in the plurality of gesture types is displayed. Prompt the person to select one of the plurality of gesture options, and whether one gesture option selected by the person matches a specific gesture type associated with the specific face authentication image Determine whether
Prompting the person to make a gesture when it is determined that the selected gesture option matches the specific gesture type;
Recognizing a gesture motion performed by the person, and determining whether the recognized gesture motion matches specific gesture motion data associated with the specific face authentication image;
Authentication method. An authentication program that causes a computer to authenticate a person,
Corresponding to a plurality of users to be authenticated, a plurality of face reference images, a plurality of gesture types, and a plurality of gesture operation data are respectively registered in the database in a one-to-one correspondence,
The authentication program is stored in the computer.
An input procedure for receiving a face input image of the person's face shot with a camera;
A face authentication procedure for performing face authentication by comparing the face input image with the plurality of face reference images registered in the database;
In the face authentication procedure, when it is determined that a specific face authentication image collated with the face input image is registered in the database, a gesture selection describing a plurality of gesture options included in the plurality of gesture types A screen is displayed to prompt the person to select one of the plurality of gesture options, and the one gesture option selected by the person is a specific gesture associated with the specific face authentication image. A type selection procedure for determining whether or not the type matches,
In the type selection procedure, when it is determined that the selected one gesture option matches the specific gesture type, a gesture prompting procedure for prompting the person to perform a gesture;
A gesture motion recognition / authentication procedure for recognizing a gesture motion performed by the person and determining whether or not the recognized gesture motion matches specific gesture motion data associated with the specific face authentication image; ,
Authentication program to execute

Images