JP2005092700A - Face identification device and passage control device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a face identification device that stabilizes the images acquired, thus increasing the success rate of identification while reducing identification errors. <P>SOLUTION: The face identification device takes a face image of a person to be identified using a camera, and collates the face image with dictionary information to verify whether or not the person to be identified is the true person. A lighting environment in the absence of the person to be identified is analyzed through self-diagnosis. When the result of the self-diagnosis is other than a reference value, the lighting environment is controlled so that the result is within the reference value. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention provides, for example, a face authentication device that authenticates whether or not the person to be authenticated is the person himself / herself based on the face image of the person to be authenticated, and a passerby's passage (for example, an important facility using the face authentication device) It is related with the traffic control apparatus which controls entrance / exit of a room.

Recently, for example, in security management, a face authentication device has been developed that authenticates whether or not the person to be authenticated is the person himself / herself based on the face image of the person to be authenticated.
In general, this type of face authentication device includes an operation unit including a camera and a color liquid crystal display unit for inputting a face image, and a control unit including an authentication engine and a camera control unit for performing an authentication process. While the input image from the camera of the operation unit is displayed on the liquid crystal display unit, the face image of the person to be authenticated is acquired and the feature point of the face is extracted, and the extracted feature point is dictionary information Register as

After that, each time the face image of the person to be authenticated is acquired while displaying the input image from the camera on the liquid crystal display unit, the facial feature points are extracted, and the extracted feature points are registered in advance as dictionary information. Is verified to verify whether or not the person to be authenticated is the person himself / herself.
JP 2001-236499 A JP 2000-60825 A

Such a face authentication apparatus needs to observe the following conditions in the installation specifications of the operation unit.
1. Blocking outside light
2. Throughout the day / night / year, the face of the person to be authenticated is within a certain illuminance and within a certain fluctuation range.
However, the installation environment has a wide range of needs, and it is difficult to make the illuminance within the fluctuation range. If these factors overlap, the feature point extraction data at the time of dictionary registration and the feature point extraction data at the time of collation (at the time of verification) vary greatly. As a result, there is a problem of a decrease in the authentication success rate for authenticating the person to be authenticated with the person, and conversely, the occurrence of erroneous authentication for authenticating the person to be authenticated with another person.

  Accordingly, an object of the present invention is to provide a face authentication device and a traffic control device that can stabilize an acquired image, improve an authentication success rate, and reduce erroneous authentication.

  The face authentication apparatus of the present invention is installed so as to face the person to be authenticated, and at least an image input unit that captures and inputs a face image of the person to be authenticated, and a face image input by the image input unit is registered in advance. Authentication means for verifying whether or not the person to be authenticated is the person himself / herself by collating with the dictionary information, and when there is no person to be authenticated within the imaging field of view of the image input means, at least for each predetermined time Environmental self-diagnosis means for analyzing the surrounding environment in the field of view by self-diagnosis, and when the self-diagnosis result of this environmental self-diagnosis means is other than the reference value, the self-diagnosis result is within the reference value based on the self-diagnosis result And environmental control means for controlling the surrounding environment.

  The traffic control device according to the present invention is installed so as to face a passerby, image input means for capturing and inputting at least a passer's face image, a face image input by the image input means, and a dictionary registered in advance Authentication means for authenticating whether or not the passer is the person by checking the information, passage control means for controlling the pass of the passer according to the authentication result of the authentication means, and the image input means When there is no passerby in the imaging field of view, environmental self-diagnosis means for analyzing at least the surrounding environment in the imaging field of view by self-diagnosis every predetermined time, and the self-diagnosis result of this environmental self-diagnosis means is other than the reference value And an environmental control means for controlling the surrounding environment so that the self-diagnosis result is within a reference value based on the self-diagnosis result.

  According to the present invention, an environmental self-diagnosis function for analyzing the surrounding environment when there is no person to be authenticated by self-diagnosis is provided, and the acquired image is stabilized by taking into account the result of this self-diagnosis. Therefore, it is possible to provide a face authentication device and a traffic control device that can improve the authentication success rate and reduce false authentication.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 schematically shows a configuration of a traffic control device using the face authentication device according to the present embodiment. In FIG. 1, the operation unit 1 in this example is a video camera (for example, a CCD type video camera, hereinafter) as an image input unit that captures and inputs a face image (an image including at least a face) of the person to be authenticated. , Simply abbreviated as a camera) 2, a display unit (for example, a color liquid crystal display) 3 as a guide means for displaying the face image of the person to be authenticated input by the camera 2, various operation guides, etc. A keyboard 4 as input means for inputting an ID number or a password as unique identification information, and a speaker 5 as voice guidance means for performing various voice guidance for the person to be authenticated are provided. It is installed in the part opposite to the face part.

  The operation unit 1 is connected to the control unit 6. That is, the camera 2, the display unit 3, the keyboard 4, and the speaker 5 of the operation unit 1 are connected to the main control unit (control unit) via the camera control unit 7, the display control unit 8, the key input control unit 9, and the voice control unit 10, respectively. ) 11.

  The illuminating device 12 as an illuminating unit illuminates the face portion of the person to be authenticated when the camera 2 captures a face image. For example, the illuminating device 12 includes four illumination lights 12a, 12b, 12c, and 12d. It is connected to the main control unit 11 via the unit 13.

  The main control unit 11 includes a memory 14 as a storage unit for storing various data, an authentication processing unit 15 as an authentication unit that performs authentication processing using a face image, a dictionary storage unit 16 as a dictionary storage unit, and a gate Control units 17 are connected to each other. In the dictionary storage unit 16, for example, a plurality of dictionary information is stored (registered) in advance corresponding to the ID number of the person to be authenticated.

  In such a configuration, the face image of the person to be authenticated input by the camera 2 and the ID number of the person to be authenticated input by the keyboard 4 are the camera control unit 7, the key input control unit 9, and the main control unit 11. And sent to the authentication processing unit 15. The authentication processing unit 15 selects, for example, dictionary information corresponding to the input ID number from the dictionary storage unit 16, and collates the selected dictionary information with the input face image, so that the person to be authenticated is the person himself / herself. It is authenticated whether or not there is, and the authentication result is sent to the main control unit 11.

  Based on the authentication result from the authentication processing unit 15, the main control unit 11 controls the passage of the person to be authenticated (passerby) by performing opening / closing control of the pass gate 18 via the gate control unit 17. The passage gate 18 is, for example, an entrance / exit door installed at the entrance of a house or an entrance / exit door of an important facility. Here, the gate control unit 17 and the traffic gate 18 constitute traffic control means.

  FIG. 2 shows a specific example of the external configuration of the operation unit 1 and an arrangement example of the illumination device 12 with respect to the operation unit 1. That is, the camera 2, the display unit 3, the keyboard 4, and the speaker 5 are provided on the front surface of the operation unit 1, respectively. In this case, the camera 2 is provided below the display unit 3, the keyboard 4 is provided on the side of the camera 2, and the speaker 5 is provided on the side of the display unit 3. An illuminating lamp 12a is arranged at the upper part of the operating unit 1, an illuminating lamp 12b is arranged at the lower part of the operating unit 1, an illuminating lamp 12c is arranged at the left side of the operating unit 1, and an illuminating lamp 12d is arranged at the right side of the operating unit 1. It is installed.

Next, the basic operation of the face authentication process in the above configuration will be briefly described.
First, the dictionary registration process will be described.
After the apparatus is turned on, the administrator menu is entered by performing a predetermined operation with the keyboard 4 of the operation unit 1. Here, a new registration or registration update menu is selected, and the registrant is positioned in front of the camera 2 of the operation unit 1. At this time, the registrant stands at a position 40 cm away from the camera 2 and confirms that the image captured by the camera 2 is displayed on the display unit 3 to detect the face.

  Next, the registrant inputs his / her ID number with the keyboard 4 of the operation unit 1 and then performs registration. When registration is executed, the authentication processing unit 15 extracts facial feature points from the input face image, and the extracted feature points are used as dictionary information in association with the input ID numbers. Register (memorize). This completes the registration process.

Next, authentication processing (collation processing) will be described.
The person to be authenticated (passerby) is positioned in front of the camera 2 of the operation unit 1 after inputting his / her ID number with the keyboard 4 of the operation unit 1. At this time, the person to be authenticated stands at a position 40 cm away from the camera 2. When the face is detected by the image captured by the camera 2, the authentication process is automatically started.

  When the authentication process is started, the authentication processing unit 15 extracts feature points from the face image captured by the camera 2 and selects dictionary information corresponding to the input ID number from the dictionary storage unit 16. By collating the selected dictionary information with the extracted feature points, it is authenticated whether or not the person to be authenticated is the principal, and the authentication result is sent to the main control unit 11.

  When receiving the authentication result from the authentication processing unit 15, the main control unit 11 determines whether or not the authentication is successful by referring to the authentication result. In the case of the authentication success, that is, the input face image and dictionary information If they match, a voice guidance indicating that the authentication has succeeded is output to the speaker 5 of the operation unit 1, a guidance for the end of authentication is displayed on the display unit 3, and an opening command is sent to the gate control unit 17. When the gate control unit 17 receives the opening command from the main control unit 11, the gate control unit 17 opens the passage gate 18 and allows the person to be authenticated (passerby) to pass.

  Note that if the authentication is unsuccessful, that is, if the input face image and the dictionary information do not match, the main control unit 11 keeps the passage gate 18 closed and the speaker 5 of the operation unit 1 fails to authenticate. While outputting the voice guidance, the display unit 3 displays guidance that has not been authenticated, and the authentication processing is terminated.

Next, an authentication operation centering on environmental self-diagnosis control, which is a feature of the present invention, will be described.
First, a first operation example will be described with reference to the flowchart shown in FIG. In the first operation example, the surrounding environment (luminance distribution of the background image) when there is no person to be authenticated is analyzed by self-diagnosis, and when the self-diagnosis result is other than the reference value, the self-diagnosis result is within the reference value. It controls the surrounding environment (for example, lighting environment).

In a standby state where there is no person to be authenticated in front of the camera 2 (within the field of view) (steps S1 and S2), the main control unit (environmental self-diagnosis means) 11 is set for a predetermined time (for example, 30 minutes). A self-diagnosis of the surrounding environment is performed every time (step S3).
That is, in a standby state where there is no person to be authenticated, a background image is acquired from the camera 2 and the luminance distribution of the acquired background image is analyzed. Here, the analysis of the luminance distribution refers to, for example, obtaining the luminance of each pixel (for example, 256 levels of luminance levels from 0 to 255) for the image captured from the camera 2 and calculating the average luminance of the entire image. The calculated average luminance is used as the self-diagnosis result. When the self-diagnosis result is within the reference value (no difference from the environment at the time of dictionary registration performed in advance), the diagnosis is OK, and the self-diagnosis result is other than the reference value (the environment is too bright or too dark). In this case (the difference from the environment at the time of dictionary registration performed in advance is large), the diagnosis is NG.

If the diagnosis is OK in the environmental self-diagnosis in step S3, the process returns to the standby state in step S1, and in preparation for the next self-diagnosis operation, the illuminance stabilization process is performed in the case of diagnosis NG (step S4), and then the standby in step S1. Return to the state and prepare for the next self-diagnosis operation.
The above operation is repeated every 30 minutes, for example. With this series of processing, it is possible to prepare the surrounding environment before the authentication processing.

Here, the following can be considered for the illuminance stabilization processing.
1. Stabilize the surrounding environment by controlling lighting and ambient light fluctuation factors
Control in conjunction with the illumination lamps 12a, 12b, 12c, 12d installed around the operation unit 1 so that the analysis result of the luminance distribution is within the reference value. For example, when it is diagnosed that the lighting environment is too bright, the lighting lamps 12a, 12b, 12c, and 12d are dimmed, and when it is diagnosed that the lighting environment is too dark, the lighting lights 12a, 12b, 12c, and 12d are brightened.
-Since light incidence from the window causes fluctuations in the surrounding environment, control is performed so that the analysis result of the luminance distribution is within the reference value in conjunction with a light shielding device such as a curtain.
Using both the illumination device 12 and the light shielding device, control is performed so that the analysis result of the luminance distribution is within the reference value.
2. Warning to the person to be authenticated by guidance
・ When the difference between the luminance distribution analysis result and the reference value is small, the guidance is directly displayed to the person to be authenticated to promote standardization of the surrounding environment.
・ If the analysis result of the brightness distribution deviates from the standard value, a guidance is sent to the system administrator to promote standardization of the surrounding environment in advance and prevent troubles during authentication.

  Now, in the standby state (step S1), when a person to be authenticated is detected (when a face is detected for an image captured from the camera 2) (step S2), the authentication processing unit 15 performs an authentication process (collation process) ( Step S5). Since this authentication process has already been described, a description thereof will be omitted.

  When the authentication process ends, the main control unit 11 performs a process based on the authentication result (step S6), and returns to the standby state of step S1. The processing based on the authentication result is, for example, as described above, when the authentication result is successful, the pass gate 18 is opened, and when the authentication result is unsuccessful, the pass gate 18 is kept closed.

  Next, a second operation example will be described with reference to the flowchart shown in FIG. In the second operation example, the surrounding environment (luminance distribution of the background image) when there is no person to be authenticated is analyzed by self-diagnosis, and the verification image correction parameter is set based on the self-diagnosis result. Luminance correction is performed on the input face image based on the set parameters.

In a standby state in which there is no person to be authenticated in front of the camera 2 (within the field of view) (steps S11 and S12), the main control unit 11 performs self-regulation of the surrounding environment every predetermined time (for example, 30 minutes). Diagnosis is performed (step S13).
That is, in a standby state where there is no person to be authenticated, a background image is acquired from the camera 2 and the luminance distribution of the acquired background image is analyzed. Here, the analysis of the luminance distribution refers to, for example, obtaining the luminance of each pixel (for example, 256 levels of luminance levels from 0 to 255) for the image taken from the camera 2 and calculating the average luminance of the entire image. The calculated average luminance is used as the self-diagnosis result.

Next, a collation image correction parameter is set based on the self-diagnosis result (step S14). For example, the collation image correction parameter is set to 11 levels of “0 to 10”, and the self-diagnosis result is a standard brightness within a reference value (there is no difference from the environment at the time of dictionary registration performed in advance). Sets the collation image correction parameter to “5”, and if the self-diagnosis result is other than the reference value (the difference is large from the environment at the time of dictionary registration performed in advance) and the environment is too bright, the collation image correction parameter is set. The value is set to “0 to 4” corresponding to the self-diagnosis result. If the environment is too dark, the collation image correction parameter is set to the value “6 to 10” corresponding to the self-diagnosis result.
When the collation image correction parameters are thus set, the process returns to the standby state in step S11 to prepare for the next self-diagnosis operation.

  Now, in the standby state (step S11), when the person to be authenticated is detected (when a face is detected for the image captured from the camera 2) (step S12), the authentication processing unit 15 performs an authentication process (collation process) ( Step S15). In this case, the face image captured by the camera 2 is subjected to correction processing based on the collation image correction parameters set in step S14, and feature points are extracted from the corrected face image and input. The dictionary information corresponding to the ID number is selected from the dictionary storage unit 16, and the selected dictionary information is compared with the extracted feature points to authenticate whether the person to be authenticated is the person himself / herself.

  For example, when the collation image correction parameter is set to “10”, the face image brightened as a whole is acquired by increasing the gain of the camera 2 above the normal value, and collation with dictionary information is performed. When the collation image correction parameter is set to “1”, the face image darkened as a whole is taken in by reducing the gain of the camera 2 below the normal value, and collation with dictionary information is performed. When the collation image correction parameter is set to “5”, the correction process is not performed.

  When the authentication process ends, the main control unit 11 performs a process based on the authentication result (step S16), and returns to the standby state of step S11. The processing based on the authentication result is, for example, as described above, when the authentication result is successful, the pass gate 18 is opened, and when the authentication result is unsuccessful, the pass gate 18 is kept closed.

  Next, a third operation example will be described with reference to the flowchart shown in FIG. In the third operation example, the surrounding environment (luminance distribution of the background image) when there is no person to be authenticated is analyzed by self-diagnosis, and statistical distribution data for a certain period of the self-diagnosis result is created. The ambient environment (for example, lighting environment) is controlled by referring to the statistical distribution data.

In a standby state where there is no person to be authenticated in front of the camera 2 (within the field of view) (steps S21 and S22), the main control unit 11 performs self-regulation of the surrounding environment every predetermined time (for example, 30 minutes). Diagnosis is performed (step S23).
That is, in a standby state where there is no person to be authenticated, a background image is acquired from the camera 2 and the luminance distribution of the acquired background image is analyzed. Here, the analysis of the luminance distribution refers to, for example, obtaining the luminance of each pixel (for example, 256 levels of luminance levels from 0 to 255) for the image taken from the camera 2 and calculating the average luminance of the entire image. The calculated average luminance is used as the self-diagnosis result.

  Next, based on the self-diagnosis result, statistical distribution data (database) for a certain period is created (updated) (step S24). For example, when the environmental correction parameter is set to 11 levels of “0 to 10” and the self-diagnosis result is a standard brightness within the reference value (there is no difference from the environment at the time of dictionary registration performed in advance). If the environment correction parameter is set to “5” and the environment is too bright because the self-diagnosis result is other than the reference value (large difference from the environment at the time of dictionary registration performed in advance), the environment correction parameter is The value is set to “0 to 4” according to the result. When the environment is too dark, the environment correction parameter is set to the value “6 to 10” according to the self-diagnosis result.

  A database as shown in FIG. 6 is created based on such environmental correction parameters. In the example of FIG. 6, one day is data every 3 hours, and the data for 3 hours is an average value of 6 data (3 × 2 = 6 because self-diagnosis is performed every 30 minutes). ing. For example, if it is currently 16:00 on August 9, the hatched portion a in FIG. 6 is updated based on the self-diagnosis result (overwriting the average value of data from 15:00 to 16:00).

Further, the average value column b of the database shown in FIG. 6 stores the average value of all the data for every three hours, and the data in this column b is referred to during the authentication process. By creating such a database, it is possible to know the lighting environment in each time zone of the day.
Thus, when statistical distribution data (database) for a certain period is created (updated), the process returns to the standby state in step S21 to prepare for the next self-diagnosis operation.

  Now, in the standby state (step S21), when a person to be authenticated is detected (when a face is detected for an image captured from the camera 2) (step S22), the authentication processing unit 15 performs an authentication process (collation process) ( Step S25). In this case, first, the illumination environment is controlled by referring to a parameter corresponding to the current time of the database created in step S24.

  That is, for example, if it is currently 9 o'clock on September 5, the average value c corresponding to the time zone (6: 00 to 9:00) including 9 o'clock in the average value column b in the database of FIG. Refer to When the referenced average value c is “0-4”, the lighting lamps 12a, 12b, 12c, 12d are darkened, and when the referenced average value c is “5”, the lighting lamps 12a, 12b, 12c are darkened. , 12d is set to the standard brightness, and when the average value c referred to is “6 to 10”, the illumination lamps 12a, 12b, 12c, 12d are brightened.

  In this way, after controlling the lighting environment, the face image is captured from the camera 2, the feature points are extracted from the captured face image, and dictionary information corresponding to the input ID number is selected from the dictionary storage unit 16, By comparing the selected dictionary information with the extracted feature points, it is authenticated whether or not the person to be authenticated is the person himself / herself.

  When the authentication process ends, the main control unit 11 performs a process based on the authentication result (step S26), and returns to the standby state of step S21. The processing based on the authentication result is, for example, as described above, when the authentication result is successful, the pass gate 18 is opened, and when the authentication result is unsuccessful, the pass gate 18 is kept closed.

  In the third operation example, the case where the illumination environment (the brightness of the illumination lights 12a, 12b, 12c, and 12d) is controlled as the control of the surrounding environment has been described, but the following control is also conceivable. .

1. Stabilize the surrounding environment by controlling lighting and ambient light fluctuation factors
-Since light incidence from the window causes fluctuations in the surrounding environment, control is performed so that the analysis result of the luminance distribution is within the reference value in conjunction with a light shielding device such as a curtain.
Using both the illumination device 12 and the light shielding device, control is performed so that the analysis result of the luminance distribution is within the reference value.
2. Warning to the person to be authenticated by guidance
・ When database parameters are close to standard values (for example, “4” or “6” when the standard value is “5” as shown in FIG. 6), the guidance is directly sent to the person to be authenticated. And promote environmental standardization.
-When the database parameter deviates significantly from the standard value (for example, “0” or “10” when the standard value is “5” as shown in FIG. 6), guidance is sent to the system administrator. Promote standardization of the surrounding environment in advance and prevent troubles during certification.

  Next, a fourth operation example will be described. In the fourth operation example, the surrounding environment (luminance distribution of the background image) when there is no person to be authenticated is analyzed by self-diagnosis, and statistical distribution data for a certain period of the self-diagnosis result is created. By referring to this statistical distribution data, luminance correction is performed on the input face image.

  That is, the fourth operation example is different from the above-described third operation example only in the authentication process in step S25 in FIG. 5, and the other operations are the same as the third operation example. Only described below.

  In the case of this operation example, first, luminance correction is performed on the input face image by referring to a parameter corresponding to the current time of the database created in step S24. That is, for example, if it is currently 9 o'clock on September 5, the average value c corresponding to the time zone (6: 00 to 9:00) including 9 o'clock in the average value column b in the database of FIG. Refer to If the referenced average value c is “0 to 4”, the face image darkened as a whole is captured by lowering the gain of the camera 2 below the normal value, and the referenced average value c is “5”. If the average value c of the camera 2 is “6 to 10”, the gain of the camera 2 is increased above the normal value, so that a face image with a bright overall image is taken. Include.

  In this way, after performing the brightness correction, the face image is taken from the camera 2, the feature point is extracted from the taken face image, and dictionary information corresponding to the input ID number is selected from the dictionary storage unit 16, By comparing the selected dictionary information with the extracted feature points, it is authenticated whether or not the person to be authenticated is the person himself / herself.

  In the first to fourth operation examples, the average luminance of the entire image is obtained as the environmental self-diagnosis. However, other than this, for example, whiteout or blackout area diagnosis may be performed. That is, for the image captured from the camera 2, the brightness of each pixel (for example, 256 levels of brightness levels from 0 to 255) is obtained. Then, a blackout area (for example, a pixel whose luminance level is “5” or lower is blackened) and a whiteout area (for example, a pixel whose luminance level is “250” or higher is whitened) are extracted. Is the self-diagnosis result.

  When this self-diagnosis result is within the reference value (when whiteout, no blackout area is small or small), the diagnosis is OK, and when the above self-diagnosis result is other than the standard value (when whiteout is present or above a certain value) (When there is a ratio of black / blackout or when there is a ratio of a certain value or more), the diagnosis is NG. For example, when the overexposure (blackout) is 5% or less for the entire image, the diagnosis is OK, and when it exceeds 5%, the diagnosis is NG. The reason may be that the environment may not be appropriate as a whole, and in terms of overexposure, the diagnosis is NG even when an unexpectedly strong spotlight is incident on the camera.

It is also possible to implement a diagnosis that combines the above-described diagnosis based on the average brightness of the entire image and the above-described diagnosis of overexposure / blackout area.
Furthermore, in both the diagnosis based on the average luminance of the entire image and the above-described whiteout / blackout area diagnosis, a specific portion (for example, a central portion of an image where a face image is expected to be present) is expected. It is also possible to make a self-diagnosis only for a nearby area.

The block diagram which shows schematically the structure of the traffic control apparatus using the face authentication apparatus which concerns on embodiment of this invention. The front view which shows the specific example of the external appearance structure of an operation part, and the example of arrangement | positioning of the illuminating device with respect to an operation part. The flowchart explaining a 1st operation example. The flowchart explaining a 2nd operation example. The flowchart explaining the 3rd operation example. The figure which shows an example of the statistical distribution data (database) in the fixed period of a self-diagnosis result.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Operation part, 2 ... Video camera (image input means), 3 ... Display part (guidance means), 4 ... Keyboard (input means), 5 ... Speaker (voice guidance means), 6 ... Control part, 7 ... Camera control 11, main control unit (control unit, environmental self-diagnosis unit), 12 illuminating device (illuminating unit), 13 illuminating control unit, 14 memory (storage unit), 15 authentication processing unit (authentication unit), 16... Dictionary storage unit (dictionary storage unit), 17... Gate control unit (passage control unit), 18.

Claims (10)

An image input means that is installed to face the person to be authenticated and captures and inputs at least a face image of the person to be authenticated;
Authentication means for authenticating whether or not the person to be authenticated is the principal by collating the face image input by the image input means with dictionary information registered in advance;
When there is no person to be authenticated in the imaging field of view of the image input unit, environmental self-diagnosis unit that analyzes at least the surrounding environment in the imaging field of view by self-diagnosis every predetermined time;
When the self-diagnosis result of the environmental self-diagnosis means is other than the reference value, the environment control means for controlling the surrounding environment based on the self-diagnosis result so that the self-diagnosis result is within the reference value;
A face authentication apparatus comprising: An image input means that is installed to face the person to be authenticated and captures and inputs at least a face image of the person to be authenticated;
Authentication means for authenticating whether or not the person to be authenticated is the principal by collating the face image input by the image input means with dictionary information registered in advance;
When there is no person to be authenticated in the imaging field of view of the image input unit, environmental self-diagnosis unit that analyzes at least the surrounding environment in the imaging field of view by self-diagnosis every predetermined time;
Parameter setting means for setting a parameter for collation image correction based on the self-diagnosis result of the environmental self-diagnosis means;
An image correction unit for performing luminance correction on the face image input by the image input unit based on the parameter set by the parameter setting unit during the authentication process in the authentication unit;
A face authentication apparatus comprising: An image input means that is installed to face the person to be authenticated and captures and inputs at least a face image of the person to be authenticated;
Authentication means for authenticating whether or not the person to be authenticated is the principal by collating the face image input by the image input means with dictionary information registered in advance;
When there is no person to be authenticated in the imaging field of view of the image input unit, environmental self-diagnosis unit that analyzes at least the surrounding environment in the imaging field of view by self-diagnosis every predetermined time;
Based on the self-diagnosis result of the environmental self-diagnosis means, a database creation means for creating an environment database in which environmental correction parameters are set for each day of self-diagnosis by the environmental self-diagnosis means and every time the self-diagnosis is performed When,
By referring to parameters corresponding to the current date and time of the environment database created by the database creation means during the authentication process in the authentication means, the surroundings so that an optimal face image can be obtained from the image input means Environmental control means for controlling the environment;
A face authentication apparatus comprising: Further comprising illumination means for illuminating at least the face portion of the person to be authenticated when photographing the face of the person to be authenticated by the image input means;
The environment control means controls the lighting means so as to obtain an appropriate lighting environment by referring to parameters corresponding to the current date and time of the environment database created by the database creating means. The face authentication apparatus according to claim 3. An image input means that is installed to face the person to be authenticated and captures and inputs at least a face image of the person to be authenticated;
Authentication means for authenticating whether or not the person to be authenticated is the principal by collating the face image input by the image input means with dictionary information registered in advance;
When there is no person to be authenticated in the imaging field of view of the image input unit, environmental self-diagnosis unit that analyzes at least the surrounding environment in the imaging field of view by self-diagnosis every predetermined time;
Statistical distribution data creating means for creating statistical distribution data for a certain period of the self-diagnosis result based on the self-diagnosis result of the environmental self-diagnosis means;
An image correction unit for performing luminance correction on the face image input by the image input unit by referring to the statistical distribution data generated by the statistical distribution data generation unit during the authentication process in the authentication unit;
A face authentication apparatus comprising:   The environmental self-diagnosis unit acquires a background image from the image input unit every predetermined time when there is no person to be authenticated within the imaging field of view of the image input unit, and the brightness of the background image acquired every predetermined time 6. The face authentication apparatus according to claim 1, wherein the distribution is analyzed.   The environmental self-diagnosis means is based on the output of an illuminometer installed in the vicinity of the image input means, and when there is no person to be authenticated in the imaging field of view of the image input means, the image input means The face authentication apparatus according to claim 1, wherein an illuminance distribution in the vicinity is analyzed. Further comprising illumination means for illuminating at least the face portion of the person to be authenticated when photographing the face of the person to be authenticated by the image input means;
The environment control means controls the illumination means so that an appropriate lighting environment is obtained based on the self-diagnosis result when the self-diagnosis result of the environment self-diagnosis means is other than a reference value. The face authentication apparatus according to claim 6 or 7. An image input means that is installed to face the passerby and captures and inputs at least a passer's face image;
Authentication means for authenticating whether or not the passerby is the person by comparing the face image input by the image input means with dictionary information registered in advance;
A passage control means for controlling the passage of the passer according to the authentication result of the authentication means;
When there is no passerby in the imaging field of view of the image input unit, environmental self-diagnosis unit that analyzes at least the surrounding environment in the imaging field of view by self-diagnosis every predetermined time;
When the self-diagnosis result of the environmental self-diagnosis means is other than the reference value, the environment control means for controlling the surrounding environment based on the self-diagnosis result so that the self-diagnosis result is within the reference value;
A traffic control device comprising: An image input means that is installed to face the passerby and captures and inputs at least a passer's face image;
Authentication means for authenticating whether or not the passerby is the person by comparing the face image input by the image input means with dictionary information registered in advance;
A passage control means for controlling the passage of the passer according to the authentication result of the authentication means;
When there is no passerby in the imaging field of view of the image input unit, environmental self-diagnosis unit that analyzes at least the surrounding environment in the imaging field of view by self-diagnosis every predetermined time;
Parameter setting means for setting a parameter for collation image correction based on the self-diagnosis result of the environmental self-diagnosis means;
An image correction unit for performing luminance correction on the face image input by the image input unit based on the parameter set by the parameter setting unit during the authentication process in the authentication unit;
A traffic control device comprising:

Images