JP7363455B2 - Authentication system, authentication device and authentication method - Google Patents

Description

The present invention relates to an authentication system, an authentication device, and an authentication method using an information code.

BACKGROUND ART Conventionally, as an authentication system that requires a high level of security, a system that uses facial recognition is sometimes adopted. Systems using such face recognition require large-scale system configurations, so they have been used in places where a high level of security is required. On the other hand, there are also needs for relatively high security with a simple configuration as an authentication system, such as for qualification examinations held at universities and concerts and events held at hall venues. As a system for achieving relatively high security with such a simple configuration, for example, an authentication system disclosed in Patent Document 1 listed below is known.

In this authentication system, a server generates an authentication code that records information regarding the user's facial feature points, and this authentication code is stored in the user's mobile terminal. In the authentication device, when the authentication code displayed on the mobile terminal is imaged by the information code reading unit and the user's face is imaged by the imaging unit, facial features are read from the imaged authentication code. The points are compared with facial feature points extracted from the captured image of the user's face, and the user is authenticated based on the comparison results. Thereby, the user can perform authentication using facial feature points simply by having the authentication device take an image of the authentication code obtained in advance and his or her own face.

Japanese Patent Application Publication No. 2014-222445

By the way, in Patent Document 1, the authentication code is imaged by an information code reading section, and the user's face is imaged by the imaging section, so that authentication information is acquired. Since two imaging means are thus required, the installation location of each imaging means may be limited. Furthermore, since there are two imaging means, there is a problem that depending on the installation situation of the authentication device including both imaging means, it may be difficult to take an image of the user.

The present invention has been made to solve the above-mentioned problems, and its purpose is to capture an image of the user's face without the user being consciously aware of it, and to extract facial feature points from the facial image. The purpose of this invention is to provide a configuration that allows authentication using .

In order to achieve the above object, the invention described in claim 1 of the claims is as follows:
an information code recording medium (30) carried by the user;
an authentication device (20) that authenticates the user;
An authentication system (10) comprising:
The information code recording medium is
comprising an information code (C) in which facial feature points extracted from the user's facial image are recorded;
The authentication device includes:
An imaging unit (23);
a determination unit (21) that determines whether or not a face recognition possible state is in which a person's face is imaged in a recognizable manner by the imaging unit;
an extraction unit (21) that extracts facial feature points from a captured image of the person's face captured by the imaging unit after the face recognition is determined by the determination unit;
a reading unit (21) that reads information regarding the facial feature points recorded in the information code from an image captured by the imaging unit of the information code displayed on the information code recording medium;
a matching unit (21) that matches the facial feature points extracted by the extraction unit and the facial feature points read by the reading unit;
an authentication unit (21) that authenticates the user based on the verification result of the verification unit;
It is characterized by having the following.

The invention according to claim 10 is
An authentication device (20) that authenticates a user who holds up an information code (C),
Facial feature points extracted from the facial image of the user are recorded in the information code,
An imaging unit (23);
a determination unit (21) that determines whether or not a face recognition possible state is in which a person's face is imaged in a recognizable manner by the imaging unit;
an extraction unit (21) that extracts facial feature points from a captured image of the person's face captured by the imaging unit after the face recognition is determined by the determination unit;
a reading unit (21) that reads information regarding the facial feature points recorded in the information code from a captured image captured by the imaging unit;
a matching unit (21) that matches the facial feature points extracted by the extraction unit and the facial feature points read by the reading unit;
an authentication unit (21) that authenticates the user based on the verification result of the verification unit;
It is characterized by having the following.

The invention according to claim 11 is
An authentication method for authenticating a user who holds up an information code (C),
Facial feature points extracted from the facial image of the user are recorded in the information code,
an imaging step of generating a captured image by an imaging unit (23);
a determination step of determining whether or not the human face is in a face-recognizable state where the human face is recognizably imaged in the imaging step;
an extraction step of extracting facial feature points from a captured image of the person's face captured by the imaging unit after the face recognition is possible in the determination step;
a reading step of reading information regarding the facial feature points recorded in the information code from a captured image captured by the imaging unit;
a matching step of matching the facial feature points extracted in the extraction step and the facial feature points read in the reading step;
an authentication step of authenticating the user based on the verification result of the verification step;
It is characterized by having the following.
Note that the reference numerals in parentheses above indicate correspondence with specific means described in the embodiments described later.

In the first aspect of the invention, the information code recording medium includes an information code in which facial feature points extracted from a user's facial image are recorded. Then, in the authentication device, the extracting unit extracts facial feature points from an image captured by the imaging unit after the determination unit determines that the face can be recognized, and the facial feature points are displayed on the information code recording medium. A reading unit reads information regarding facial feature points recorded in the information code from an image captured by the imaging unit. Then, the facial feature points extracted by the extraction section and the facial feature points read by the reading section are compared by the matching section, and the user is authenticated by the authentication section based on the matching result.

Thereby, the user can perform authentication using facial feature points simply by having the imaging unit of the authentication device take an image of his or her own face and the information code displayed on the information code recording medium. In particular, since one imaging section serves as both the imaging means for imaging the user's face and the imaging means for imaging the information code displayed on the information code recording medium, the user cannot see the imaging section. At the same time, the user must hold the information code over the imaging unit. In such a case, before capturing an image of the information code, the image capturing section can capture a recognizable image of the face of the user who is about to hold up the information code without the user being conscious of it. Furthermore, there is no need for the user to stand still relative to the imaging unit in order to capture an image of the user's face. Therefore, before reading the information code, it is possible to take an image of the user's face and start processing to extract its facial feature points, so there is no need to wait until the information code is held up to the imaging unit and the matching result is obtained by the matching unit. The authentication time required by the user can be shortened.

In the invention of claim 2, a display section is provided that shows a state in which the information code can be read in a predetermined display state, and the imaging section is arranged at a position within the field of view of a user looking at the display section. Ru. As a result, when the user tries to hold the information code over the imaging unit, the user can visually confirm that the display unit is in the predetermined display state, so the user can use the imaging unit that is within the user's field of vision. The user's face can be imaged in a state where it is easier to recognize, and authentication accuracy can be improved.

In the third aspect of the invention, the facial feature points extracted by the extraction section are temporarily stored in the storage section and deleted after a certain period of time has elapsed. As a result, even if the information code cannot be read because, for example, the information code recording medium is not held up in an appropriate manner, the extracted facial feature points will be retained for a certain period of time, and the information code will be properly read within this certain period of time. This eliminates the need to take a recognizable image of the user's face. On the other hand, the extracted facial feature points are no longer stored after a certain period of time has elapsed, so even if the face of another person who is different from the authentication target is unintentionally captured, it can be appropriately deleted.

In the invention of claim 4, the illumination unit controls the illumination state so that, under predetermined conditions, the irradiation state when the imaging unit images a person's face is different from the irradiation state when the imaging unit images the information code. controlled by the department. As a result, for example, under predetermined conditions such as ambient brightness, even when using the same imaging unit, an imaging state suitable for imaging a person's face and an imaging state suitable for imaging an information code can be determined. can be switched, the extraction accuracy of facial feature points and the reading accuracy of information codes can be improved, and as a result, the authentication accuracy can be improved.

In the fifth aspect of the present invention, the determination unit determines that the face recognition is possible when the face of the person captured in a recognizable manner gradually increases in size in the continuous images taken by the imaging unit. When a third party uses the imaging unit to capture a face photo displayed on a mobile device, even if it is recognized as a person's face, it is assumed that face recognition is possible unless the person's face gradually increases in size. There will be no judgment. Therefore, even if a face photograph is taken for fraudulent purposes, it is less likely to be erroneously determined that the face is in a state where face recognition is possible, thereby making it possible to prevent spoofing.

In the invention according to claim 6, the determination unit performs face recognition when the inclination of the center line of the person's face captured in a recognizable manner with respect to the reference line in the image captured by the imaging unit is within a predetermined angular range. It is determined that the state is possible.

When a third party uses the imaging unit to capture a face photo displayed on the screen of a mobile device, if the face photo is tilted, the center line of the face in the captured image may be a vertical line (a line along the direction of gravity). ). On the other hand, when capturing an image of a person walking toward the imaging unit, the inclination of the center line of the face relative to the vertical line in the captured image is sufficiently small. In other words, if the inclination of the center line of the person's face that is recognizably imaged with respect to the reference line such as the vertical line in the image captured by the imaging unit does not fall within a predetermined angular range, the photograph of the face is not captured. There is a possibility that there are. For this reason, it is determined that the face can be recognized when the inclination of the center line of a person's face that is imaged in a recognizable manner falls within a predetermined angular range, and when the center line does not fall within the predetermined angular range. By determining that the face is not in a state where face recognition is possible, even if a face photo is taken for fraudulent purposes, if the face picture is tilted significantly, it will be difficult to misjudge that the face is in a state where face recognition is possible, thereby preventing spoofing. be able to.

In the seventh aspect of the present invention, the determination unit determines that the face recognition is possible when the face of the person imaged in a recognizable manner can be recognized as blinking in the continuous images taken by the imaging unit. When a third party causes the image capturing unit to capture a face photo or the like displayed on the screen of a mobile terminal, even if it is recognized as a human face, it is not determined that the face is in a face-recognizable state because the person cannot blink. Therefore, even if a face photograph is taken for illicit purposes, it will not be erroneously determined that the face is in a state where face recognition is possible, and therefore, spoofing can be prevented.

In the eighth aspect of the present invention, it is determined that the face recognition is possible when the range of the shadow of the person's face, which is photographed in a recognizable manner, changes in the continuous images taken by the imaging unit. When a third party uses the imaging unit to capture a photo of a face displayed on a mobile device, the area of the shadow cast on the face does not change even if it is recognized as a human face, so the face can be recognized. There will be no judgment. Therefore, even if a face photograph is taken for illicit purposes, it will not be erroneously determined that the face is in a state where face recognition is possible, and therefore, spoofing can be prevented.

In the ninth aspect of the present invention, the determination unit determines that the face recognition is possible when the depth of the human face that is recognizably captured by the imaging unit is recognized. When a third party causes the imaging unit to capture a face photo displayed on the screen of a mobile device, even if it is recognized as a human face, the depth of that face is not recognized, so it is determined that face recognition is possible. There's nothing wrong with that. Therefore, even if a face photograph is taken for illicit purposes, it will not be erroneously determined that the face is in a state where face recognition is possible, and therefore, spoofing can be prevented.

According to the tenth aspect of the invention, an authentication device having the same effects as the first aspect can be realized.
According to the eleventh aspect of the invention, an authentication method having the same effects as the first aspect can be realized.

FIG. 1 is an explanatory diagram schematically showing the configuration of an authentication system according to a first embodiment. FIG. 2 is an explanatory diagram illustrating a part of the location where facial feature points are extracted from captured facial image data. FIG. 2 is a block diagram schematically showing the electrical configuration of an authentication device. FIG. 2 is a block diagram schematically showing the electrical configuration of a mobile terminal. 3 is a flowchart illustrating the flow of authentication processing by an authentication device. FIG. 2 is an explanatory diagram illustrating a state in which a face photo displayed on a screen of a mobile terminal is captured by an imaging unit of an authentication device. FIG. 3 is an explanatory diagram illustrating a state in which a facial photograph displayed on a screen of a mobile terminal is tilted and captured by an imaging unit of an authentication device. FIG. 6 is an explanatory diagram illustrating a state in which a face photograph in which a part of the face is shaded is displayed on the screen.

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an authentication system, an authentication device, and an authentication method according to the present invention will be described below with reference to the drawings.
The authentication system 10 shown in FIG. 1 mainly includes an authentication device 20, and an information code (hereinafter also referred to as authentication code C) in which facial feature points and personal information extracted from a user's face image are recorded. ) is configured as a system that performs personal authentication of the user.

In this embodiment, the authentication system 10 is employed, for example, in an access control system in which a user whose face authentication is successful is permitted to enter a specific room. The user carries a mobile terminal 30 as an information code recording medium including an authentication code C, and the authentication displayed on the screen of the mobile terminal 30 is sent to the authentication device 20 installed at the entrance to the specific room. By performing an operation to image the business code C, entry to the specific room is permitted upon successful authentication. Specifically, authentication success information is transmitted from the authentication device 20 to a locking/unlocking control device that controls locking/unlocking of an electric lock provided on an entrance/exit door, so that the electric lock is placed in an unlocked state by the locking/unlocking control device. This will allow you to enter the specific room mentioned above.

The authentication code C is generated by a predetermined code generation device that has acquired facial image data from a user applying for personal authentication. This code generation device extracts facial feature points from facial image data acquired from a user, converts information regarding the facial feature points, personal information, etc. into an information code, and generates an authentication code C. In this embodiment, items such as the size, shape, outline, and arrangement position of facial organs such as eyes, eyebrows, nose, ears, and mouth are adopted as facial feature points. Then, in accordance with a known facial feature point extraction process for extracting facial feature points performed by the code generation device, for example, the features of each item for the acquired facial image data D as illustrated in FIG. For example, data such as "D1: long eyes" and "D2: large mouth" are calculated and extracted as facial feature points.

Then, according to the publicly known information code generation process performed by the code generation device, it can be decrypted using a decryption key (encryption key) based on the information related to the facial feature points extracted as described above, personal information, etc. The authentication code C is generated by encrypting the code. In this embodiment, as the authentication code C, for example, a QR code (registered trademark) whose security is enhanced by encryption is employed.

Next, the authentication device 20 will be explained using FIGS. 1 and 3.
The authentication device 20 is configured as a device that authenticates the user using the authentication code C displayed on the screen of the mobile terminal 30, and is configured as a device that authenticates the user by using the authentication code C displayed on the screen of the mobile terminal 30. It is installed on the wall near the door at the user's eye level. As shown in FIG. 3, the authentication device 20 includes a control section 21, a storage section 22, an imaging section 23, a display section 24, a lighting section 25, an operation section 26, a communication section 27, and the like.

The control unit 21 is mainly configured with a microcomputer, and performs overall control of the authentication device 20 and various calculations, and functions, for example, to execute authentication processing to be described later. The storage unit 22 is made up of a known storage medium such as ROM, RAM, HDD, and non-volatile memory, and has application programs, predetermined databases, etc. for executing authentication processing made available to the control unit 21 in advance. Stored. Furthermore, the storage unit 22 stores in advance a decryption key for decrypting the authentication code C.

The imaging unit 23 is a camera equipped with a light receiving sensor (for example, a C-MOS area sensor, a CCD area sensor, etc.), and image data captured through a reading window (imaging window) is output to the control unit 21. It is configured as follows. Based on the facial image data captured by the imaging unit 23, the control unit 21 functioning as an extraction unit performs facial feature point extraction processing in the same manner as the code generation device described above, thereby extracting facial feature points. . Further, based on the authentication code C imaged by the imaging unit 23, the control unit 21 functioning as a reading unit decrypts the encrypted authentication code C using the decryption key stored in the storage unit 22. By performing the information code reading process (decoding process) as described above, the facial feature points recorded in the authentication code C are read.

The display unit 24 is, for example, an LED, and is arranged in a substantially rectangular ring shape surrounding the reading window of the imaging unit 23 as shown in FIG. 1, and is controlled by the control unit 21 to extract facial feature points. The lighting/blinking state is configured to change depending on the success or failure of the authentication process, the result of the authentication process, etc. Due to this arrangement, the imaging section 23 is arranged at a position within the field of view of the user looking at the display section 24.

The illumination unit 25 includes an illumination light source, an illumination lens, and the like, and is configured to be controlled by the control unit 21 to irradiate the imaging range of the imaging unit 23 with illumination light. The operation unit 26 is configured to output an operation signal according to an input operation to the control unit 21, and the control unit 21 receives this operation signal and performs processing according to the input operation. There is. The communication unit 27 is configured as a communication interface for performing data communication with external devices such as the above-mentioned locking/unlocking control device and an entrance/exit management server, and cooperates with the control unit 21 to perform communication processing. is configured to do so.

Next, the mobile terminal 30 will be explained using FIGS. 1 and 4.
The mobile terminal 30 shown in FIGS. 1 and 4 is, for example, a smartphone, and includes a control unit 31 including a CPU, a memory 32, an imaging unit 33 configured as a camera, and a touch panel whose display contents are controlled by the control unit 31. 34, an operation section 35 that outputs a signal to the control section 31 in response to a touch operation on the touch panel 34 or a key operation, a communication section 36 for communicating with an external device, etc.

The authentication code C generated by the code generation device described above is stored in advance in the memory 32 of the mobile terminal 30 configured as described above. Then, in response to a predetermined operation on the operation unit 35 by the user, the authentication code C read out from the memory 32 is displayed on the touch panel 34 (see FIG. 1).

Next, the authentication process performed by the authentication device 20 as an authentication method will be described in detail with reference to the flowchart shown in FIG.
When the authentication process is started by the control unit 21 of the authentication device 20, the imaging process shown in step S101 in FIG. 5 is performed. In this process, the imaging unit 23 captures an image while the illumination unit 25 illuminates the image with illumination light through the reading window. Subsequently, in the determination process of step S103, it is determined whether or not the image capturing unit 23 is in a state in which a human face is captured in a recognizable manner (hereinafter also referred to as a face recognition possible state), and whether or not the human face is recognized is determined. If it is determined that no image has been taken and the face recognition is not possible (No in S103), the imaging unit 23 continues to take an image. Note that the imaging process described above may correspond to an example of an "imaging step" in which the imaging unit 23 generates a captured image. Further, the determination process in step S103 above may correspond to an example of a "determination step", and the control unit 21 that executes this determination process may correspond to an example of a "determination unit".

A user entering the specific room attempts to hold the authentication code C displayed on the screen of the mobile terminal 30 over the reading window of the authentication device 20. At this time, when the face of the user looking at the reading window is imaged by the imaging unit 23 and the face becomes ready for face recognition, a determination of Yes is made in step S103, and the face image data generation process shown in step S105 is performed. Ru. In this process, an image for extracting facial feature points is newly captured by the imaging unit 23, and facial image data is generated. Note that in the face image data generation process, a plurality of new images for extracting facial feature points may be taken, and one of the image data from which the facial feature points can be easily extracted may be used as the facial image data.

Subsequently, facial feature point extraction processing shown in step S107 is performed, and facial feature points are extracted from the facial image data generated as described above in the same manner as the code generation device described above, and are temporarily stored in the storage unit 22. is memorized. Note that the facial feature point extraction process described above may correspond to an example of an "extraction step."

When the facial feature points are extracted in this way, the display unit 24 is turned on in the first lighting state (for example, (blue lighting state) (S109). Subsequently, an information code reading process shown in step S111 is performed. In this process, a decoding process is performed using the decryption key stored in the storage unit 22 to read the information code imaged by the imaging unit 23. Note that the above information code reading process may correspond to an example of a "reading step".

Next, in the determination process of step S113, it is determined whether or not the authentication code C has been successfully read. Here, if the authentication code C is not imaged or an information code different from the authentication code C is read, it is determined as No in step S113, and a certain period of time has elapsed since the first lighting state. If the time has not passed (No in S115), the information code reading process of step S111 is performed, and the state in which the information code can be read is maintained.

Then, when a certain period of time has not elapsed since the first lighting state was entered (No in S115), the user whose face image was captured as described above receives the authentication code displayed on the screen of the mobile terminal 30. The authentication code C is imaged by holding C over the reading window of the authentication device 20, and if the information regarding facial feature points, personal information, etc. is successfully read (Yes in S113), a verification process shown in step S117 is performed. In this process, based on the facial feature points extracted from the facial image data and the facial feature points read from the authentication code C, a check is made to see if all items in the facial feature points match. Note that the above-mentioned verification process corresponds to an example of a "verification step", and the control unit 21 that executes this verification process may correspond to an example of a "verification unit".

If all the items in the facial feature points match, the determination process in step S119 is determined as Yes, indicating that the authentication has been successful, so that the display unit 24 remains in the second lighting state for a certain period of time (e.g. , green lighting state) (S123). Subsequently, a transmission process shown in step S125 is performed, and authentication success information indicating successful authentication is transmitted to the lock/unlock control device via the communication unit 27, and the authentication success information and the read personal information are transmitted to the communication unit 27. The information is sent to a server for access management via the host computer. Then, the process from step S101 described above is performed, and the face of the next user who wishes to enter the room can be imaged. Note that in the determination process in step S119, the determination is not limited to Yes when all the items in the facial feature points match, but also the determination is Yes when a predetermined number of items in the facial feature points match. Good too. Further, the determination process in step S119 above may correspond to an example of an "authentication step", and the control unit 21 that executes this determination process may correspond to an example of an "authentication unit".

As described above, the authentication device 20 transmits authentication success information, and the lock/unlock control device that receives this authentication success information unlocks the electric lock, allowing the user to enter the specific room. becomes possible.

On the other hand, if the third party is holding up the authentication code C displayed on the screen of the other person's mobile terminal 30 over the reading window of the authentication device 20, each item in the facial feature points does not match, so in step S119 Since the determination is No and indicates that the authentication has failed, the display unit 24 is in the third lighting state (for example, red lighting state) for a certain period of time (S127). In this case, since authentication success information is not transmitted from the authentication device 20, the electric lock will not be unlocked by the lock/unlock control device. Note that in such a case, the authentication failure information, the read personal information, etc. may be transmitted to a server for entry/exit management, etc. via the communication unit 27.

Further, when a certain period of time has elapsed after the authentication code C enters the first lighting state without being imaged (Yes in S115), the facial feature points extracted and stored as described above are stored in the storage unit 22. , and the processing from step S101 described above is performed again.

As explained above, in the authentication system 10 according to the present embodiment, the authentication code C in which the facial feature points extracted from the user's face image are recorded is stored in the memory 32 of the mobile terminal 30 so as to be displayable on the screen. remembered. Then, in the authentication device 20 , facial feature points are extracted from the captured image of the human face captured by the imaging unit 23 after it is determined that the face recognition is possible, and an authentication code is displayed on the screen of the mobile terminal 30 . Information regarding the facial feature points recorded in the authentication code C is read from the captured image captured by the imaging unit 23. Then, the extracted facial feature points and the read facial feature points are compared, and the user is authenticated based on the comparison result.

As a result, the user can perform authentication using facial feature points by simply having the imaging unit 23 of the authentication device 20 capture an image of his or her own face and the authentication code C displayed on the screen of the mobile terminal 30. . In particular, since one imaging section 23 serves as both the imaging means for imaging the user's face and the imaging means for imaging the authentication code C displayed on the screen of the mobile terminal 30, the user can While looking at the reading window of the section 23, the user holds the authentication code C over the reading window. In such a case, before capturing an image of the authentication code C, the image capturing unit 23 should capture an image of the face of the user who is about to hold the authentication code C so that the user can recognize it without the user being conscious of it. Can be done. Furthermore, there is no need for the user to stand still relative to the imaging unit 23 in order to capture an image of the user's face. Therefore, before reading the authentication code C, it is possible to start the process of capturing an image of the user's face and extracting its facial feature points. The time it takes to obtain a verification result is shortened, and the authentication time required by the user can be shortened.

In particular, the facial feature points extracted by the facial feature point extraction process are temporarily stored in the storage unit 22 until a certain period of time has elapsed since the first lighting state (until it is determined Yes in S115). It is stored and deleted after a certain period of time. As a result, even if the authentication code C displayed on the screen of the mobile terminal 30 is not read because the authentication code C is not held up in an appropriate manner, the extracted facial feature points are retained for a certain period of time. By appropriately reading the authentication code C within this certain period of time, it is possible to eliminate the trouble of taking a recognizable image of the user's face. On the other hand, the extracted facial feature points will be deleted and no longer stored after a certain period of time has passed (Yes in S115), so even if the face of another person who is different from the authentication target is unintentionally captured, can be erased.

As a first modification of the present embodiment, the display unit 24 is controlled by the control unit 21 to perform authentication during the process in which the determination of No in step S103 is repeated, that is, before the face recognition becomes possible. In order to indicate that the code C can be read, a predetermined display state (for example, a blue flashing state) different from the first to third lighting states may be used. For this reason, a display section 24 is provided which shows the readable state of the authentication code C in the predetermined display state, and the imaging section 23 is positioned within the field of view of the user looking at the display section 24. It will be placed.

As a result, when the user attempts to hold the authentication code C over the imaging unit 23, the user can visually confirm that the display unit 24 is in the above-mentioned predetermined display state, so that it is within the user's field of vision. The user's face can be imaged in a state where the user's face is more easily recognized by the image capturing unit 23, and authentication accuracy can be improved. Note that the first to third lighting states and the above-mentioned predetermined display states are just examples, and changing colors, blinking at predetermined intervals, changing lighting/blinking locations, etc. may also be used. Further, during the first to third lighting states or the above-mentioned predetermined display states, a notification may be made using audio or the like.

As a second modification of the present embodiment, under predetermined conditions, the irradiation state when the imaging unit 23 images a person's face and the irradiation state when the imaging unit 23 images the authentication code C are different. The illumination unit 25 may be controlled by the control unit 21 differently. For example, assuming that the illumination unit 25 is configured to be able to emit infrared light in addition to visible light, under certain conditions such as nighttime when the surroundings are dark, infrared light may be emitted when capturing an image of a person's face. However, when the authentication code C is imaged, visible light whose brightness is adjusted to be suppressed may be emitted. If bright visible light is irradiated in a dark environment, the user may feel dazzled, and when capturing an image of the authentication code C using this infrared light, the information code recording medium on which the authentication code C is displayed may be This is because depending on the type, etc., it may not be possible to capture an image in a readable manner.

In this way, even when using the same imaging unit 23 under predetermined conditions such as ambient brightness, the imaging state (irradiation state) suitable for imaging a human face and the authentication code C can be captured. Since the imaging state (irradiation state) suitable for the image can be switched, the extraction accuracy of facial feature points and the reading accuracy of the authentication code C can be improved, and as a result, the authentication accuracy can be improved.

Note that in order to further enhance the confidentiality of information regarding facial feature points, a partially non-disclosed code may be employed as the authentication code C. This partially private code is configured as an information code having a public area and a private area in which information that is decryptably encrypted using a decryption key is recorded. Therefore, by recording information about facial feature points in a private area, a code reader that does not have a decryption key can only read the information recorded in the public area, and it is possible to prevent other information from being recorded. Since even facial feature points cannot be recognized, the confidentiality of information regarding facial feature points can be increased. On the other hand, when the present authentication system 10 is used in an environment with relatively low security requirements, the authentication code C is used without encryption based on the information regarding the facial feature points extracted as described above. may be generated.

[Second embodiment]
Next, an authentication system, an authentication device, and an authentication method according to the second embodiment will be described with reference to the drawings.
The second embodiment differs from the first embodiment mainly in that the conditions for determining whether the face is in a recognition possible state are made more strict. Therefore, components that are substantially the same as those in the first embodiment are designated by the same reference numerals, and their descriptions will be omitted.

In the determination process in step S103 in the authentication process described above (the process performed by the control unit 21 functioning as a determination unit), the human face is imaged in a recognizable manner by the imaging unit 23 (face recognition possible state). It is determined whether or not there is. For this reason, if a third party causes the imaging unit 23 to capture a face photo or the like displayed on the screen of the mobile terminal for fraudulent purposes, it may be erroneously determined that the face recognition is possible. In particular, in a system that performs authentication by reading an information code such as an authentication code C displayed on a screen of a mobile terminal, as in the present invention, it is possible to fraudulently use the mobile terminal by displaying a face photo, etc. on the screen. This is easily assumed.

Therefore, in the present embodiment, by making the criteria for determining whether the face is in a recognition possible state more strict, even if the imaging unit 23 captures a face photograph for fraudulent purposes, it will be erroneously determined that the face is in a recognition possible state. To prevent impersonation by pretending to be a legitimate authority (authorized user) and performing acts permitted only to legitimate authority.

Specifically, in the determination process in step S103, if the face of a person captured in a recognizable manner gradually increases in size in the continuous images taken by the imaging unit 23, it is determined that the face recognition is possible.

For this reason, when a legitimate authority person approaches the imaging unit 23 when entering the specific room, the face of the legitimate authority person is recognized, and then the image is captured so that the face gradually becomes larger. It is determined that the state is possible. On the other hand, as illustrated in FIG. 6, when a third party causes the imaging unit 23 to capture a face photo P displayed on the screen of a mobile terminal, even if it is recognized as a person's face, the person's face gradually changes. It will not be determined that the face recognition is possible unless the face becomes large.

As described above, in the present embodiment, in the continuous images taken by the imaging unit 23, when the face of a person photographed in a recognizable manner gradually increases in size, it is determined that the face is in a state where the face can be recognized. Even when an image is captured, it is less likely to be mistakenly determined to be in a state where face recognition is possible, so spoofing can be prevented.

Note that, as a first modification of the present embodiment, in the determination process of step S103, the inclination of the center line Lf of the human face that is imaged in a recognizable manner with respect to the reference line Lo in the image captured by the imaging unit 23 is determined. If the angle falls within a predetermined angle range, it may be determined that the face recognition is possible.

In this first modification, the reference line Lo is set to be a vertical line (a line along the direction of gravity), and a third party can capture a face photo displayed on the screen of a mobile terminal on the imaging unit 23. For example, if the facial photograph is tilted when the image is captured, as illustrated in FIG. 7, the center line Lf of the face in the captured image will be tilted with respect to the vertical line serving as the reference line Lo. On the other hand, when capturing an image of a person walking toward the imaging unit 23, the inclination of the center line Lf of the face relative to the reference line Lo in the captured image becomes sufficiently small. That is, in the image captured by the imaging unit 23, if the inclination θ of the center line Lf with respect to the reference line Lo is not within the predetermined angular range θth (for example, 30°), there is a possibility that a facial photograph is being captured. .

Therefore, in the first modified example, it is determined that the face is in a recognition possible state when the inclination θ of the center line Lf with respect to the reference line Lo with respect to the human face imaged in a recognizable manner is within the predetermined angle range θth. , if the inclination θ of the center line Lf with respect to the reference line Lo is not within the predetermined angle range θth, it is determined that the face recognition is not possible. As a result, even if a face photograph is taken for illicit purposes, if the face photograph is tilted significantly, it will be less likely to be mistakenly determined to be in a face-recognizable state, thereby making it possible to prevent spoofing.

Further, as a second modification of the present embodiment, in the determination process of step S103, if the face of the person photographed in a recognizable manner in the continuous images taken by the imaging unit 23 can be recognized as blinking, It may be determined that the state is in a recognizable state. When a third party causes the image capturing unit to capture a face photo or the like displayed on the screen of a mobile terminal, even if it is recognized as a human face, it is not determined that the face is in a face-recognizable state because the person cannot blink. Therefore, even if a face photograph is taken for illicit purposes, it will not be erroneously determined that the face is in a state where face recognition is possible, and therefore, spoofing can be prevented.

Further, as a third modification of the present embodiment, in the determination process of step S103, if the range of the shadow cast on the face of a person photographed in a recognizable manner changes in the continuous images taken by the imaging unit 23, face recognition is performed. It may be determined that the state is possible.

For example, as illustrated in FIG. 8, in a photograph P of a person's face, a portion of the face is cast in shadow (see hatched area S in FIG. 8) depending on the brightness of the surroundings at the time of photographing. When the person to be imaged moves, the extent to which the shadow is cast changes because the way light hits the face changes. On the other hand, in the face photo P, even if the face is shaded, its range does not change. That is, if the range of the shadow on a person's face does not change in consecutive images, there is a high possibility that a face photograph is being taken.

For this reason, in the continuous images taken by the imaging unit 23, when the range of the shadow cast on the face of a person photographed in a recognizable manner changes, it is determined that the face is in a state where face recognition is possible, and facial photographs, etc. are transferred to the imaging unit 23. In the case where the image is captured as a human face, the range of the shadow cast on the face does not change even if it is recognized as a human face, so it is not determined that the face is in a face-recognizable state. Therefore, even if a face photograph is taken for illicit purposes, it will not be erroneously determined that the face is in a state where face recognition is possible, and therefore, spoofing can be prevented.

In addition, as a fourth modification of the present embodiment, in the determination process of step S103, if the depth is recognized regarding the face of a person photographed in a recognizable manner by the imaging unit 23, it is determined that the face recognition is possible. You may judge. When a third party causes the imaging unit to capture a face photo displayed on the screen of a mobile device, even if it is recognized as a human face, the depth of that face is not recognized, so it is determined that face recognition is possible. There's nothing wrong with that. Therefore, even if a face photograph is taken for illicit purposes, it will not be erroneously determined that the face is in a state where face recognition is possible, and therefore, spoofing can be prevented.

Note that the present invention is not limited to the above-described embodiments and modifications, and may be embodied as follows, for example.
(1) The authentication code C is not limited to being stored in the memory 32 so that it can be displayed on the touch panel 34 of the mobile terminal 30, but can also be printed on one side of an information code recording medium such as a paper medium so that it can be imaged. You may also be prepared by being asked to do so.

(2) The authentication system 10 according to the present invention is not limited to being applied to an entrance/exit control system that allows entry into a specific room, but can be adopted, for example, to a management system that manages the entrance gate of a concert venue. Alternatively, it may be adopted in a system used at a retail store or the like when providing a service that requires authentication.

(3) The authentication process performed by the control unit 21 of the authentication device 20 is not limited to starting the process to read the authentication code C after the facial feature points are extracted, but also before the face recognition becomes possible. The authentication code C may be read at the same time. In other words, if the authentication code C is read before the face recognition becomes possible, facial feature points are extracted from the facial image data captured after the face recognition becomes possible, and the extracted facial features are Verification processing can be performed based on the points and the facial feature points previously read from the authentication code C.

(4) In the authentication process performed by the control unit 21 of the authentication device 20, matching is performed based on the facial feature points extracted from the captured facial image data and the facial feature points read from the authentication code C as described above. In addition to processing, by sending captured facial image data or extracted facial feature points, captured image data of authentication code C, or read facial feature points to a server, etc., it can be verified on the server, etc. It is also possible to perform the processing and obtain the verification results.

10...Authentication system 20...Authentication device 21...Control unit (determination unit, extraction unit, reading unit, verification unit, authentication unit)
22...Storage unit 23...Imaging unit 24...Display unit 25...Lighting unit 30...Mobile terminal (information code recording medium)
C...Authentication code (information code)

Claims (11)

an information code recording medium carried by the user;
an authentication device that authenticates the user;
An authentication system comprising:
The information code recording medium is
comprising an information code in which facial feature points extracted from the user's facial image are recorded;
The authentication device includes:
an imaging unit;
a determination unit that determines whether or not the human face is in a face-recognizable state where the human face is recognizably imaged by the imaging unit;
an extraction unit that extracts facial feature points from a captured image of the person's face captured by the imaging unit after the determination unit determines that the face recognition is possible;
a reading unit that reads information regarding the facial feature points recorded in the information code from a captured image captured by the imaging unit of the information code displayed on the information code recording medium;
a matching unit that matches the facial feature points extracted by the extraction unit and the facial feature points read by the reading unit;
an authentication unit that authenticates the user based on the verification result of the verification unit;
An authentication system comprising: comprising a display section that indicates a readable state of the information code in a predetermined display state;
2. The authentication system according to claim 1, wherein the imaging unit is arranged at a position within a field of view of the user viewing the display unit. The authentication system according to claim 1 or 2, further comprising a storage unit in which the facial feature points extracted by the extraction unit are temporarily stored and deleted after a certain period of time has elapsed. an illumination unit that irradiates illumination light toward an imaging range by the imaging unit;
a control unit that controls the irradiation state of the illumination light in the illumination unit;
Equipped with
The control unit controls the illumination so that, under predetermined conditions, an irradiation state when the imaging unit images the person's face is different from an irradiation state when the imaging unit images the information code. The authentication system according to any one of claims 1 to 3, characterized in that the authentication system controls a portion of the authentication system. Claims 1 to 4, wherein the determination unit determines that the face recognition is possible when the face of a person photographed in a recognizable manner gradually becomes larger in the continuous images taken by the imaging unit. The authentication system described in any one of the following. The determination unit determines that the face is in the face recognition possible state when the inclination of the center line of the person's face that is recognizably imaged is within a predetermined angle range with respect to a reference line in the image captured by the imaging unit. The authentication system according to any one of claims 1 to 4, wherein the authentication system determines that. The determining unit determines that the face is in the face-recognizable state when the face of the person imaged in a recognizable manner can be recognized as blinking in the continuous images taken by the imaging unit. The authentication system according to any one of Items 1 to 4. The determining unit determines that the face is in the face recognition possible state when a range of a shadow cast on a person's face photographed in a recognizable manner changes in the continuous images taken by the imaging unit. The authentication system according to any one of 1 to 4. 5. The determination unit determines that the face recognition is possible when the depth of the person's face photographed in a recognizable manner by the imaging unit is determined to be in the face recognition possible state. The authentication system described in paragraph (1) above. An authentication device that authenticates a user who holds up an information code,
Facial feature points extracted from the facial image of the user are recorded in the information code,
an imaging unit;
a determination unit that determines whether or not the human face is in a face-recognizable state where the human face is recognizably imaged by the imaging unit;
an extraction unit that extracts facial feature points from a captured image of the person's face captured by the imaging unit after the determination unit determines that the face recognition is possible;
a reading unit that reads information regarding the facial feature points recorded in the information code from a captured image captured by the imaging unit;
a matching unit that matches the facial feature points extracted by the extraction unit and the facial feature points read by the reading unit;
an authentication unit that authenticates the user based on the verification result of the verification unit;
An authentication device comprising: An authentication method for authenticating a user who holds up an information code,
Facial feature points extracted from the facial image of the user are recorded in the information code,
an imaging step of generating a captured image by an imaging unit;
a determination step of determining whether or not the human face is in a face-recognizable state where the human face is recognizably imaged in the imaging step;
an extraction step of extracting facial feature points from a captured image of the person's face captured by the imaging unit after the face recognition is possible in the determination step;
a reading step of reading information regarding the facial feature points recorded in the information code from a captured image captured by the imaging unit;
a matching step of matching the facial feature points extracted in the extraction step and the facial feature points read in the reading step;
an authentication step of authenticating the user based on the verification result of the verification step;
An authentication method comprising:

Images