JP2010237756A - Method and device for authenticating face - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a face authentication method and a face authentication device for improving the precision of face recognition. <P>SOLUTION: A featured value data table in which face featured value data in image data are classified by time in photographing is stored in an RAM 12 in advance. When a power supply is turned on, an authentication mode is set, and when a user photographs his or her own face, a clock section 14 obtains time data in photographing. An authentication data selection part 15 retrieves time data matched with time data obtained by the clock section 14 in photographing new image data from the featured value data table, and when the matched time data includes the classified face featured value data, the authentication data selection part 15 selects the face featured value data as face featured value data for authentication. A face matching determination part 16 determined whether or not the face featured values calculated on the basis of the new image data are matched with the face featured values of the face featured value data for authentication selected by the authentication data selection part 15 as face authentication. When the matching is determined, the authentication mode is switched to a portable telephone mode. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a face authentication method and a face authentication device.

  In recent years, as a security measure, a technology that performs face authentication to determine whether or not a pre-stored authentication face image and a newly photographed new face image match, and releases the lock only when they match. Has become commonplace. For example, it is conceivable to install such a face authentication function in a camera-equipped mobile phone. In this case, the camera cannot be used normally as a mobile phone, and only the face authentication can be taken, and it is used as a mobile phone only when it is determined that the face authentication has been taken. It becomes possible. Accordingly, if only the face image of the owner is stored as an authentication image, it is not determined that the face is authenticated even if a face other than the owner performs face authentication.

  Depending on the photographing conditions, the authentication face image and the newly photographed new face image may differ greatly. In such a case, even in the case of the same person, it may be determined that the face authentication does not match, so Patent Document 1 stores the date and time at the time of shooting, the age of the subject, and the face image for authentication. The facial image feature amount is corrected according to the elapsed time and the face authentication is performed using the corrected facial image feature amount, thereby supporting the change of the face of the same person due to aging. . Further, in Patent Document 2, when there is no subject in the shooting range, the surrounding lighting environment is analyzed every predetermined time and the environmental conditions are controlled to prevent erroneous authentication due to the lighting environment / illuminance.

JP 2007-102517 A Japanese Patent Laying-Open No. 2005-092700

  Depending on the situation such as the shooting time and the shooting position, the face when shooting is different. For example, the face that has just been woken up in the morning and the face when several hours have passed since waking up are different. Furthermore, in the case of women, since they apply makeup, the face when they are at home in the early morning before applying makeup is different from the face when they are at work around noon after applying makeup. Also in Patent Documents 1 and 2, since different faces cannot be authenticated as the same person by such a situation, the accuracy of face authentication is low.

  SUMMARY An advantage of some aspects of the invention is that it provides a face authentication method and a face authentication apparatus capable of improving the accuracy of face authentication.

  In order to achieve the above object, the face authentication method of the present invention acquires situation information when image data including a face image of a subject is captured, and is classified in advance for each situation information including a face image of the subject. The authentication data classified into one that matches the situation information obtained when the image data was captured is selected from the plurality of authentication data thus selected, and the subject's face in the selected authentication data And the face of the subject in the image data is detected, and it is determined whether or not the detected face of the subject in the authentication data matches the detected face of the subject in the image data It is characterized by doing.

  The situation information is preferably at least one of time information, time information, and position information when the image data or the authentication data is captured.

  Further, in the plurality of authentication data, when there is no authentication data classified into the situation information that matches the situation information when the image data is captured, the situation information when the image data is captured is When the authentication data classified into the closest situation information is selected, or when it is determined that they do not match in the determination of whether or not they match, the situation information at the time of shooting the image data is displayed. It is preferable to select again the authentication data classified into the closest situation information and determine again whether or not they match.

  Further, the face authentication device of the present invention includes a photographing unit that captures a face image of a subject to acquire image data, a situation information acquisition unit that acquires situation information when the image data is captured by the photographing unit, A situation in which a plurality of pieces of authentication data including a face image of a subject are pre-classified and stored for each situation information at the time of shooting, and a situation when the image data is taken from the plurality of pieces of authentication data A selection unit that selects the authentication data classified into one that matches information, a subject face in the authentication data selected by the selection unit, and a subject face in the image data are detected. A face detection unit that detects a face of the subject in the authentication data detected by the face detection unit, and a subject in the image data detected by the face detection unit. The face and determination means for determining whether match, characterized in that it comprises a.

  According to the present invention, the authentication data classified into one that matches the situation information obtained when the image data is photographed out of a plurality of authentication data classified in advance for each situation information including the subject image. Since data is selected and it is determined whether or not the face of the subject in the selected authentication data matches the face of the subject in the image data, even if the face differs depending on the situation at the time of shooting, It is possible to suppress the determination of mismatch when face authentication is performed on the same person. Thereby, the accuracy of face authentication can be improved.

It is a front view which shows the mobile phone with a camera. It is a block diagram which shows the electric constitution of the mobile phone with a camera. It is explanatory drawing of the feature-value data table which classified the face feature-value data for every hour. It is a flowchart which shows the flow which performs face authentication. It is explanatory drawing of the feature-value data table which classified the face feature-value data every 3 hours. It is explanatory drawing of the feature-value data table which classified the face feature-value data for every season. It is a front view which shows the mobile phone with a camera of 2nd Embodiment. It is a flowchart which shows the flow which performs face authentication of 2nd Embodiment. It is a block diagram which shows the electrical structure of the mobile telephone with a camera of 3rd Embodiment. It is explanatory drawing of the feature-value data table of 3rd Embodiment. It is a flowchart which shows the flow which performs face authentication of 3rd Embodiment. It is a block diagram which shows the electric constitution of the mobile telephone with a camera of 4th Embodiment. It is explanatory drawing of the feature-value data table of 4th Embodiment. It is a flowchart which shows the flow which performs face authentication of 4th Embodiment.

[First Embodiment]
As shown in FIG. 1, a camera-equipped mobile phone 2 embodying the present invention includes a main body case 3, a power button 4, a call button 5, a plurality of input buttons 6, a setting button 7 for performing various settings and selections, and a release button. In addition to various operation buttons such as 8, an LCD 9 and a photographing lens 10 are provided.

  The release button 8 has a two-stage push structure. When the release button 8 is lightly pressed (half-pressed), a shooting preparation operation such as focusing is performed. When further pressed (fully pressed) in this state, a photographing operation is performed.

  Whether the mobile phone with camera 2 is used as a mobile phone, a shooting / playback mode (hereinafter referred to as shooting mode) for shooting and playback of images, and shooting for authentication, and whether or not to permit use And an authentication mode for determining whether or not. Switching between the cellular phone mode and the photographing mode is performed by operating the setting button 7. When the power is turned on by operating the power button 4 in the power-off state, the mode is switched to the authentication mode. It is possible to switch to the mobile phone mode and the photographing mode only when authentication photographing is performed in the authentication mode and use permission is given. Even in the authentication mode, an incoming call can be made.

  As shown in FIG. 2, the camera-equipped mobile phone 2 includes a CPU 11, a RAM 12, a face feature amount calculation unit 13, a clock unit 14, an authentication data selection unit 15, a face matching determination unit 16, and a communication unit 17. Yes. A plurality of telephone numbers are stored in the RAM 12, and when making a call in the mobile phone mode, the telephone number stored in the RAM 12 is read by operating the setting button 7 or the telephone number is operated by operating the input button 6. Then, the call button 5 is operated. The communication unit 17 performs wireless communication with a base station of a mobile phone company. Then, a call is made via a microphone and a speaker (not shown) built in the main body case 3.

  Behind the photographing lens 10 is a CCD 21 on which subject light that has passed through the photographing lens 10 enters. The CCD 21 photoelectrically converts the subject image to generate a time-series three-color signal and sends it to a correlated double sampling circuit (CDS) 22. A timing signal (clock signal) from a TG (timing generator) 27 controlled by the CPU 11 is input to the CCD driver 26. The CCD 21 is operated by the CCD driver 26. The three-color signal output from the CCD 21 is converted into a three-color signal that accurately corresponds to the amount of charge accumulated in each cell of the CCD 21 after the noise component is removed by the CDS 22. The three-color signal output from the CDS 22 is amplified by an amplifier (AMP) 23 according to the ISO sensitivity, and converted to image data by an A / D converter 24.

  The image input controller 25 is connected to the CPU 11 via the data bus 29, and controls the CDS 22, AMP 23, and A / D converter 24 according to instructions from the CPU 11. The image data output from the A / D converter 24 is written into the video memory 31 or the buffer memory 33.

  When the LCD 9 is used as a viewfinder, the video memory 31 temporarily stores through-image image data having a low resolution. The live view image data is sent to the LCD driver 32 via the data bus 29, and the live view image is displayed on the LCD 9. The buffer memory 33 temporarily stores captured high-resolution image data. The image data read from the buffer memory 33 is stored in the RAM 12 by a data reader driven and controlled by the memory controller 34.

  While the image data is stored in the buffer memory 33, the face feature amount calculation unit 13 detects the face of the photographed subject based on the stored image data and also detects the face parts (eyes, nose, Lip etc.) feature quantity (position, size, etc.) is calculated. Then, the calculated face feature amount data is added to the image data. Further, time data (hour, minute, second) acquired by the clock unit 14 at the time of shooting is also added to the image data. That is, the RAM 12 stores the image data to which the facial feature amount data and the time data are added as authentication data.

  As shown in FIG. 3, the facial feature amount data of the authentication data stored in the RAM 12 is classified for each time at the time of shooting. The feature amount data table includes time data when the authentication data is photographed and face feature amount data of the authentication data. In the present embodiment, the time data is divided every hour.

  The authentication data selection unit 15 uses the face feature amount of image data (hereinafter referred to as “new image data”) taken in the authentication mode and the face feature amount of the authentication data stored in the RAM 12 to perform face authentication. Operates when performing. The authentication data selection unit 15 searches the feature amount data table for time data that matches the time data acquired by the clock unit 14 when the new image data was captured, and the facial features classified into the searched time data. The amount data is selected as face feature amount data for authentication. For example, when the shooting time is 00:20, face feature amount data (A1) classified as matching time data (00: 0 to 01:00) is selected as authentication face feature amount data.

  If there is no face feature data classified as matching time data, the authentication data selection unit 15 selects the face feature data classified as the nearest time data as authentication face feature data. To do. For example, if the shooting time is 02:10 and there is no face feature data classified as matching time data (02: 0 to 03:00), the authentication data selection unit 15 uses the closest time data. The face feature data (B1) classified as (01: 0 to 02:00) is selected as authentication face feature data.

  When there are a plurality of face feature amount data classified into the matching time data, the authentication data selection unit 15 calculates an average value of the plurality of face feature amount data, and uses the calculated average value as the authentication face value. Select as feature data. For example, when the photographing time is 10:20, the average value of the face feature amount data (H1, H2, H3) classified into the matching time data (10: 0 to 11:00) is used as the authentication face feature. Select as quantity data.

  The face match determination unit 16 determines whether the face feature value calculated based on the new image data matches the face feature value of the authentication face feature data selected by the authentication data selection unit 15. judge. When the match is determined, the face match determination unit 16 outputs a match signal to the CPU 11. Upon receiving the coincidence signal, the CPU 11 switches from the authentication mode to the mobile phone mode.

  The face coincidence determination unit 16 outputs a disagreement signal to the CPU 11 when no coincidence is determined. In response to the input of the mismatch signal, the CPU 11 continues the authentication mode. That is, when the match is not determined by the face match determination unit 16, the camera-equipped mobile phone 2 cannot be used in the mobile phone mode or the photographing mode. For this reason, the user stores authentication data in the RAM 12 in advance.

  The operation of the camera-equipped mobile phone 2 configured as described above will be described with reference to the flowchart of FIG. When the power of the camera-equipped mobile phone 2 is turned on (step (hereinafter, S) 1), the authentication mode is set (S2). In the authentication mode, when the user operates the release button 8 to photograph his / her face as a subject (S3), the clock unit 14 acquires time data at the time of photographing (S4).

  The authentication data selection unit 15 searches the feature amount data table for time data that matches the time data acquired by the clock unit 14 when the new image data is captured, and the face feature amount classified as the matching time data. If there is data (Y in S5), the face feature amount data is selected as authentication face feature amount data (S6).

  In addition, when there is no face feature data classified as matching time data (N in S5), the authentication data selection unit 15 converts the face feature data classified as the nearest time data into the authentication face data. It selects as feature-value data (S7). It should be noted that the authentication data selection unit 15 calculates the average value of the plurality of face feature amount data when there are a plurality of face feature amount data in the time data in both (S6) and (S7). The calculated average value is selected as authentication face feature data.

  The face feature amount calculation unit 13 calculates a face feature amount based on the new image data, and the face match determination unit 16 selects the face feature amount of the new image data and the authentication data selection unit 15 as face authentication. It is determined whether or not the face feature amount of the authentication face feature amount data matches (S8). When a match is determined (Y in S9), the face match determination unit 16 outputs a match signal to the CPU 11. The CPU 11 receives the coincidence signal and switches from the authentication mode to the mobile phone mode (S10).

  On the other hand, the face match determination unit 16 outputs a mismatch signal to the CPU 11 when a match is not determined (N in S9). In response to the input of the mismatch signal, the CPU 11 prompts to continue the authentication mode and perform imaging again (S11). At the same time, the CPU 11 stores the new image data in the RAM 12 as authentication data (S12).

  As described above, in the authentication mode, the face feature value data classified into the time data that matches the time data acquired by the clock unit 14 at the time of shooting the new image data or the closest time data is selected from the feature value data table. In addition, since face authentication is performed using the face feature amount of the selected face feature amount data for authentication and the face feature amount of the new image data, even if the face features differ depending on the shooting time, face authentication of the same person It is possible to suppress the determination of a mismatch when performing. Thereby, the accuracy of face authentication can be improved.

  In the first embodiment, the facial feature data stored in the RAM 12 is classified every hour. However, the classification time is not limited to one hour. For example, as shown in FIG. You may classify every 3 hours. Further, as shown in FIG. 6, the face feature amount data may be classified for each season. In this case, the clock unit 14 acquires time data and date data (year, month, day) at the time of shooting. Instead of the classification for each season, the facial feature data may be classified every month, every two months, or the like.

[Second Embodiment]
In the second embodiment shown in FIGS. 7 and 8, when the face matching determination unit 16 determines that the images match, the new image data is selected to be stored in the RAM 12 as authentication data. In addition, the same code | symbol is attached | subjected to the structural member similar to the thing of 1st Embodiment, and the detailed description is abbreviate | omitted.

  As shown in FIG. 7, in the camera-equipped mobile phone 40 according to the second embodiment, the face matching determination unit 16 selects the face feature amount calculated based on the new image data and the authentication data selection unit 15. When it is determined that the face feature amount of the authentication face feature amount data matches, the CPU 11 selects a selection screen for selecting whether or not to store the new image data in the RAM 12 as authentication data on the LCD 9. 41 is displayed. The selection screen 41 includes a comment image 42 “Do you want to store a new image as an authentication image?”, A first selection image 43 “Yes”, and a second selection image 44 “No”.

  The user operates the setting button 7 to select either the first selection image 43 “Yes” or the second selection image 44 “No”. When the first selection image 43 is selected, the CPU 11 stores new image data in the RAM 12 as authentication data. When the second selected image 44 is selected, the CPU 11 deletes the new image data.

  If the match is not determined by the face matching determination unit 16 and the number of times of face authentication is less than a predetermined number (for example, three times), the authentication data selection unit 15 performs selection again. When performing the selection again, the authentication data selection unit 15 selects the face feature amount data classified into the time data closest to the matching time data as the authentication face feature amount data.

  The operation of the camera-equipped mobile phone 40 configured as described above will be described with reference to the flowchart of FIG. When the camera-equipped mobile phone 2 is turned on (S101), the authentication mode is set (S2). Henceforth, since (S103)-(S108) are the same as (S3)-(S8) of 1st Embodiment, description is abbreviate | omitted.

  When the face match determination unit 16 determines that they match (Y in S109), it outputs a match signal to the CPU 11. In response to the input of the coincidence signal, the CPU 11 displays the selection screen 41 on the LCD 9 (S110). When the user operates the setting button 7 to select the first selection image 43 “Yes” (Y in S111), the CPU 11 stores the new image data in the RAM 12 as authentication image data (S112). On the other hand, when the second selection image 44 of “No” is selected (N in S111), the CPU 11 deletes the new image data (S113). After storing or erasing the new image data, the CPU 11 switches from the authentication mode to the mobile phone mode (S114).

  On the other hand, when the match is not determined (N in S109), the face match determination unit 16 outputs a mismatch signal to the CPU 11. In response to the input of the mismatch signal, the CPU 11 determines whether it is the third face authentication (S115). If it is the first or second time and it is not the third face authentication (N in S115), the authentication data selection unit 15 performs selection again. When performing the selection again, the authentication data selection unit 15 selects the face feature data classified as the nearest time data as authentication face feature data (S107). Then, face authentication is performed using the newly selected authentication face feature data (S108).

[Third Embodiment]
In the third embodiment of FIGS. 9 to 11, the face feature data stored in the RAM 12 is classified for each photographing position. In addition, the same code | symbol is attached | subjected to the structural member similar to the thing of the 1st, 2nd embodiment, and the detailed description is abbreviate | omitted.

  As shown in FIG. 9, a GPS unit 51 is provided in the camera-equipped mobile phone 50 according to the third embodiment. The GPS unit 51 measures the current location and acquires its position data, and includes an antenna 52 and a GPS calculation unit 53. The antenna 52 receives radio waves from four or more (for example, four) GPS satellites out of 24 GPS satellites orbiting the earth. The received radio wave is sent to the GPS calculation unit 53. The GPS calculation unit 53 stores a calculation program for calculating position data (latitude / longitude) of the current location based on the input radio wave.

  When photographing is performed in the authentication mode, the CPU 11 outputs a program activation signal to the GPS calculation unit 53. When the program activation signal is input, the GPS calculation unit 53 activates the calculation program and calculates the position data of the current location based on the radio wave input from the antenna 52 (the distances from the four GPS satellites are known at the same time). To calculate the current position).

  As shown in FIG. 10, the facial feature amount data stored in the RAM 12 is classified for each position data acquired by the GPS unit 51 at the time of shooting. The feature amount data table includes shooting position (location, latitude / longitude, attribute) data and face feature amount data.

  The authentication data selection unit 15 searches the feature amount data table for position data that matches the position data acquired by the GPS unit 51 when the new image data was captured, and the facial features classified into the searched position data. The amount data is selected as face feature amount data for authentication. For example, when the shooting position is home, face feature data (A21, A22) classified as matching position data (home) is selected as authentication face feature data. Further, if there is no face feature data classified as the matching position data, the authentication data selection unit 15 selects the face feature data classified as the nearest position data as authentication face feature data. To do. At this time, the authentication data selection unit 15 first sets the position data having the same attribute as the closest position data, and if there is no position data having the same attribute, searches for the closest position data based on the latitude / longitude.

  The operation of the camera-equipped mobile phone 50 configured as described above will be described with reference to the flowchart of FIG. When the camera-equipped mobile phone 2 is turned on (S201), the authentication mode is set (S202). In the authentication mode, when the user shoots his / her face (S203), the GPS unit 51 calculates position data of the current location (S204).

  The authentication data selection unit 15 searches the feature quantity data table for position data that matches the position data acquired by the GPS unit 51 when the new image data was captured, and the face feature quantities classified into the matching position data. If there is data (Y in S205), the face feature data is selected as authentication face feature data (S206).

  In addition, when there is no face feature data classified into the matching position data (N in S205), the authentication data selection unit 15 converts the face feature data classified into the closest position data into the authentication face data. It selects as feature-value data (S207). Note that, in both cases (S206) and (S207), the authentication data selection unit 15 calculates the average value of the plurality of face feature amount data when there are a plurality of face feature amount data. The average value is selected as authentication face feature data. Henceforth, since (S208)-(S216) are the same as (S108)-(S114) of 2nd Embodiment and (S11), (S12) of 1st Embodiment, description is abbreviate | omitted.

[Fourth Embodiment]
In the fourth embodiment of FIGS. 12 to 14, the facial feature data stored in the RAM 12 is classified for each photographing time and photographing position. In addition, the same code | symbol is attached | subjected to the structural member similar to the thing of the 1st-3rd embodiment, and the detailed description is abbreviate | omitted.

  As shown in FIG. 12, both the clock unit 14 and the GPS unit 51 are provided in the camera-equipped mobile phone 60 according to the fourth embodiment.

  As shown in FIG. 13, the face feature data stored in the RAM 12 is classified into ID numbers (“1” to “24”) assigned for each shooting time and shooting position. The feature amount data table includes shooting time (name, time zone) data, shooting position data, and face feature amount data. In the present embodiment, the shooting time data is divided every 3 hours. The authentication data selection unit 15 searches the feature amount data table for an ID that matches the time data and the position data acquired when the new image data was captured, and selects the face feature amount data classified into the searched ID. , And selected as authentication face feature data. For example, when the shooting time is 05:30 and the shooting position is at home, “4” is searched for as the ID number, and the face feature data (B31) classified into this is selected as authentication face feature data. To do.

  The operation of the camera-equipped mobile phone 60 configured as described above will be described with reference to the flowchart of FIG. When the camera-equipped mobile phone 2 is turned on (S301), the authentication mode is set (S302). In the authentication mode, when the user photographs his / her face (S303), the clock unit 14 acquires time data at the time of photographing, and the GPS unit 51 calculates position data of the current location (S304).

  The authentication data selection unit 15 searches the feature amount data table for an ID that matches the time data and the position data acquired when the new image data was captured, and has face feature amount data classified into the matching ID. In the case (Y in S305), the face feature data is selected as authentication face feature data (S306).

  Further, when there is no face feature amount data classified into the matching ID (N in S305), the authentication data selection unit 15 converts the face feature amount data classified into the closest ID into the authentication face feature amount. The data is selected (S307). At this time, the authentication data selection unit 15 selects the ID of the same shooting location as the closest ID in the closest time zone. Note that, in both cases (S306) and (S307), the authentication data selection unit 15 calculates the average value of the plurality of face feature amount data when there are a plurality of face feature amount data. The average value is selected as authentication face feature data. Henceforth, since (S308)-(S316) are the same as (S208)-(S216) of 3rd Embodiment, description is abbreviate | omitted.

  In the above-described embodiment, image data shot with a camera-equipped mobile phone is stored in the RAM as authentication data. However, image data shot with another shooting device, for example, a digital camera, is used as authentication data in the camera. You may memorize | store in RAM of an attached mobile phone. Further, the authentication data may be stored in a data server connected via the Internet without storing the authentication data in the RAM.

  When a single camera-equipped mobile phone is used by multiple users, a feature data table is created for each of the multiple users and stored in the RAM. In the authentication mode, the user name to be authenticated is input. Then, face authentication may be performed using the input feature quantity data table of the user.

  Furthermore, in the fourth embodiment, when there is no face feature data classified into an ID that matches the time data and the position data acquired at the time of capturing new image data, the authentication data selection unit 15 is closest. Although the ID of the same shooting location in the time zone is selected as the closest ID, the ID of the closest shooting location in the same time zone may be selected as the closest ID. In addition, the determination result in the face authentication when the selection method of each of these two patterns is performed is stored, and the closest ID is selected by using a selection method with a high probability determined to be coincident. Good.

  Further, the present invention is not limited to a camera-equipped mobile phone, and can be implemented as long as it is an electronic device having a photographing function, for example, a digital camera.

2, 40, 50, 60 Mobile phone with camera 10 Photography lens 12 RAM (memory means)
13 face feature amount calculation unit (face detection means)
14 Clock part (situation information acquisition means)
15 Authentication data selection section (selection means)
16 Face match determination unit (determination means)
51 GPS unit (situation information acquisition means)

Claims (5)

Obtain situation information when shooting image data including the face image of the subject,
The authentication data classified into one that matches the situation information when the image data was photographed out of a plurality of authentication data classified beforehand for each situation information at the time of photographing including the face image of the subject. Selected,
Detecting the face of the subject in the selected authentication data and the face of the subject in the image data;
A face authentication method, comprising: determining whether or not the detected face of the subject in the authentication data matches the detected face of the subject in the image data.   The face authentication method according to claim 1, wherein the situation information is at least one of time information, time information, and position information when the image data or the authentication data is captured.   When the authentication data classified into the situation information that matches the situation information when the image data is captured is not among the plurality of authentication data, the closest to the situation information when the image data is captured 3. The face authentication method according to claim 1 or 2, wherein the authentication data classified as situation information is selected.   When it is determined in the determination as to whether or not they match, the authentication data classified into the situation information closest to the situation information when the image data was captured is selected and matched. 3. The face authentication method according to claim 1 or 2, wherein the determination as to whether or not the image is present is performed again. Photographing means for photographing a face image of a subject and obtaining image data;
Situation information acquisition means for acquiring situation information when the image data is imaged by the imaging means;
A plurality of authentication data including a face image of a subject is stored in advance classified and stored for each situation information at the time of shooting,
A selection means for selecting the authentication data classified into one that matches the situation information when the image data is captured from the plurality of authentication data;
Face detection means for detecting the face of the subject in the authentication data selected by the selection means and the face of the subject in the image data;
A determination unit that determines whether or not the face of the subject in the authentication data detected by the face detection unit matches the face of the subject in the image data detected by the face detection unit;
A face authentication apparatus comprising:

Images