JP2007122493A - Authentication device and authentication method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the convenience of a user and to improve security performance. <P>SOLUTION: A wide angle camera picks up an authentication image and a monitored image, and when a human detection sensor detects an authentication object person, an authentication processing part makes a storing part store authentication record information including a personal ID on the basis of a collation result between the authentication image and registered information. The picked-up image is subjected to face collation by a monitoring processing part and then stored as image record information in the storing part, and a classification processing part is configured to classify the image record information on the basis of the personal ID, and a display processing part is configured to display the classified image record information on a display part. Furthermore, classification processing executed by the classification processing part is configured to be able to be performed at optional timing. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an authentication device and an authentication method for authenticating a person to be authenticated by recording a captured image as a monitoring image and using the captured face image as an authentication image when performing personal authentication. The present invention relates to an authentication apparatus and an authentication method capable of improving not only the time of login but also the security after login.

  2. Description of the Related Art Conventionally, there has been known an authentication device that performs personal authentication by capturing the face of a person to be authenticated using a CCD (Charge Coupled Device) camera or the like and comparing the captured face image with a registered image registered in advance. Yes.

  In recent years, such an authentication apparatus is also used for login processing to a personal computer (hereinafter referred to as “PC”). For example, in Patent Document 1, when the face of a person who wants to log in to a PC or the like is captured by a camera, and the captured face image is successfully compared with a pre-registered face image, login to the PC is performed. Permitted technology is disclosed.

  By the way, as the importance of security management for PCs and the like increases, attempts have been made to prevent data leakage when a PC user leaves the PC. For example, if the input to the PC has not been performed for a predetermined time, a so-called screen saver is activated, and only when a password set by the PC user is input, the screen saver is terminated and the normal screen display is restored. Are known.

JP 2003-67339 A

  However, the technique of Patent Document 1 described above is a technique for improving security in a login scene, and is not a technique considering security after successful login. That is, there is a possibility that data may be stolen by a suspicious person or the like when a PC user who has been successfully logged in leaves.

  For this reason, it is also conceivable to secure security when the PC user is absent by using the technique of Patent Document 1 and the above-described screen saver. However, the screen saver is a technique for preventing data leakage by prohibiting the use of the PC by another person when the PC user is absent, and is not a technique for actively acquiring information when the PC user is absent.

  For example, even when a suspicious person attempts to log in to a PC when the PC user is absent, the risk of such data theft cannot be transmitted to the PC user. That is, although there is a risk of data theft, there is a problem in that the security cannot be sufficiently secured because the risk cannot be dealt with.

  Therefore, how to realize an authentication device that can improve security not only when logging in but also when a PC user is absent is a big issue.

  The present invention has been made to solve the above-described problems caused by the prior art, and provides an authentication apparatus and an authentication method capable of improving security not only when logging in but also when a PC user is absent. The purpose is to provide.

  In order to solve the above-described problems and achieve the object, the invention according to claim 1 records the captured image as a monitoring image and, when performing personal authentication, uses the captured face image as an authentication image. An authentication device for authenticating a person to be authenticated by using the storage device, wherein identification information for identifying the person to be authenticated and a registration image of the person to be authenticated are stored in association with each other, and stored in the storage unit And a classifying unit that classifies the monitoring image and / or the authentication image into an image corresponding to the identification information and an image not corresponding to the identification image based on the identification information.

  The invention according to claim 2 is characterized in that, in the above-mentioned invention, the classification means performs the classification of the monitoring image and / or the authentication image when the identity authentication is successful.

  According to a third aspect of the present invention, in the above invention, the classification unit performs the classification of the monitoring image and / or the authentication image when recording the image.

  The invention according to claim 4 further comprises authentication recording means for recording the identification information corresponding to the person to be authenticated when the identity authentication is successful in the invention described above, wherein the classification means Based on the identification information recorded by the authentication recording means, the monitoring image and / or the authentication image is classified into an image corresponding to the identification information and an image not corresponding to the identification information.

  The invention according to claim 5 further comprises display means for displaying an image not corresponding to the identification information recorded by the authentication recording means among the images classified by the classification means in the above invention. It is characterized by.

  Further, in the invention according to claim 6, in the above invention, the display unit displays the authentication image that has failed in the personal authentication among the images classified by the classification unit, separately from the monitoring image. It is characterized by.

  According to a seventh aspect of the present invention, in the above invention, the display unit divides the image classified by the classification unit as an image corresponding to the identification information stored in the storage unit for each identification information. It is characterized by being displayed separately.

  In addition, the invention according to claim 8 further includes a deletion unit that deletes an image corresponding to the identification information recorded by the authentication recording unit from among the images classified by the classification unit. It is characterized by.

  The invention according to claim 9 is characterized in that, in the above invention, the monitoring image and the authentication image are picked up by a single camera.

  According to a tenth aspect of the present invention, in the above invention, the monitoring image and the authentication image are respectively captured by two cameras, the monitoring image capturing camera and the authentication image capturing camera. Features.

  The invention according to claim 11 is an authentication method for authenticating a person to be authenticated by recording a captured image as a monitoring image and using the captured face image as an authentication image when performing personal authentication. A storage step of associating and storing identification information for identifying the authentication target person and a registration image of the authentication target person, and the monitoring image based on the identification information stored in the storage step And / or a classification step of classifying the authentication image into an image corresponding to the identification information and an image not corresponding to the identification information.

  The invention according to claim 12 is characterized in that, in the above invention, the classification step performs the classification of the monitoring image and / or the authentication image if the identity authentication is successful.

  The invention according to claim 13 is characterized in that, in the above invention, the classification step performs the classification of the monitoring image and / or the authentication image when the image is recorded.

  The invention according to claim 14 further includes an authentication recording step of recording the identification information corresponding to the person to be authenticated when the identity authentication is successful in the invention described above, wherein the classification step includes the step of Based on the identification information recorded in the authentication recording step, the monitoring image and / or the authentication image is classified into an image corresponding to the identification information and an image not corresponding to the identification information.

  According to the first aspect of the present invention, based on the identification information for identifying the person to be authenticated and the storage means for storing the registered image of the person to be authenticated in association with each other, the monitoring is performed based on the identification information stored in the storage means. Since it is configured to include a classifying unit that classifies the image and / or the authentication image into an image corresponding to the identification information and an image not corresponding to the image, the convenience of the user is improved by reducing the burden of checking the monitoring image. In addition, the security can be improved by reducing the difficulty of confirming the suspicious person.

  According to the invention of claim 2, since the classification means is configured to classify the monitoring image and / or the authentication image if the personal authentication is successful, the monitoring image acquired when the person to be authenticated is absent or By classifying the authentication images when providing them to the person to be authenticated, there is an effect that the processing load at the time of image acquisition can be reduced.

  Further, according to the invention of claim 3, the classification means is configured to classify the monitoring image and / or the authentication image when recording the image. There is an effect that the monitoring image or the authentication image acquired when the user is absent can be promptly provided to the person to be authenticated.

  According to a fourth aspect of the present invention, the personal computer further comprises an authentication recording means for recording identification information corresponding to the person to be authenticated when the personal authentication is successful, and the classification means includes the identification information recorded by the authentication recording means. Based on this, the monitoring image and / or the authentication image is classified into an image corresponding to the identification information and an image that does not correspond, so that the image that matches the personal ID of the successful authentication person and the image that does not match are efficiently classified. By doing this, there is an effect that an image in which a person other than the successful authentication person is captured can be easily provided to the successful authentication person.

  Further, according to the invention of claim 5, since the display means for displaying an image not corresponding to the identification information recorded by the authentication recording means among the images classified by the classification means is further provided, By not displaying an image that matches the personal ID, it is possible to reduce the image check burden of a successful authentication person.

  Further, according to the invention of claim 6, since the display means is configured to display the authentication image in which the personal authentication has failed among the images classified by the classification means in distinction from the monitoring image, the personal authentication is performed. In other words, by distinguishing the image of a person who has approached the distance where personal authentication is performed from other images, the image that is strongly expected to be a suspicious person is given priority over other images. By doing so, it is possible to reduce the effort of finding a suspicious person.

  According to the invention of claim 7, the display means is configured to display the images classified as the images corresponding to the identification information stored in the storage means by the classification means so as to be distinguished for each identification information. When a personal ID is registered, it is possible to provide an easy-to-understand explanation to the PC user, such as the visit of a personal ID registrant and the approach of a personal ID unknown person who is expected to be a suspicious person. There is an effect.

  Further, according to the invention of claim 8, since the image processing apparatus is further provided with the deletion means for deleting the image corresponding to the identification information recorded by the authentication recording means among the images classified by the classification means, the personal authentication succeeds. By deleting the person's image, the storage area of the storage device can be used efficiently.

  According to the ninth aspect of the present invention, since the monitoring image and the authentication image are picked up by one camera, the manufacturing cost of the authentication device can be suppressed by reducing the cost related to the camera. There is an effect.

  According to the invention of claim 10, since the monitoring image and the authentication image are respectively captured by the two cameras of the monitoring image capturing camera and the authentication image capturing camera, the monitoring range and the authentication range can be flexibly set. There is an effect that it can be set.

  According to the invention of claim 11, based on the storage step for storing the identification information for identifying the person to be authenticated and the registered image of the person to be authenticated in association with each other, and the identification information stored in the storage step Since the monitoring image and / or the authentication image is configured to include a classification step for classifying the monitoring image and / or the authentication image into an image corresponding to the identification information and an image not corresponding to the monitoring information, the convenience of the user is reduced by reducing the burden of checking the monitoring image. It is possible to improve the security and to improve the security by reducing the difficulty of confirming the suspicious person.

  According to the twelfth aspect of the present invention, since the classification process is configured to classify the monitoring image and / or the authentication image if the personal authentication is successful, the monitoring image acquired when the person to be authenticated is absent or By classifying the authentication images when providing them to the person to be authenticated, there is an effect that the processing load at the time of image acquisition can be reduced.

  According to the invention of claim 13, since the classification step is configured to classify the monitoring image and / or the authentication image when the monitoring image is recorded, the classification process is performed at the time of image acquisition. The monitoring image or the authentication image acquired when the person is absent can be quickly provided to the person to be authenticated.

  According to the fourteenth aspect of the present invention, the method further includes an authentication recording step of recording identification information corresponding to the person to be authenticated when the personal authentication is successful, and the classification step includes the identification information recorded by the authentication recording step. Based on this, the monitoring image and / or the authentication image is classified into an image corresponding to the identification information and an image that does not correspond, so that the image that matches the personal ID of the successful authentication person and the image that does not match are efficiently classified. By doing this, there is an effect that an image in which a person other than the successful authentication person is captured can be easily provided to the successful authentication person.

  Exemplary embodiments of an authentication apparatus and an authentication method according to the present invention will be described below in detail with reference to the accompanying drawings. In the following description, an authentication apparatus that is used for authentication when logging on to a PC and that acquires a monitoring image when the PC user is absent will be described.

  Prior to detailed description of each embodiment, features of the authentication method according to the present invention will be described. FIG. 1 is a diagram showing the characteristics of the authentication method according to the present invention. In the figure, a scene where the authentication device acquires a monitoring image when the PC user is absent is shown.

  As shown in the figure, in the authentication method according to the present invention, a monitoring image acquired via a camera is recorded in a storage device such as a hard disk device (see (1) in FIG. 1). In addition, when acquiring a monitoring image, the above-described recording process is performed only when the monitoring image is changed by using the moving object detection function, but even if the amount of the recording image is reduced in this way, The recorded image becomes enormous. In many cases, only a small number of suspicious person images are included in the enormous recorded images.

  In the conventional technique, all of these recorded images are usually displayed on a display device, or the recorded images are grouped to display a representative image (thumbnail). However, with such a method, it has not been possible to effectively avoid the problem that the PC user needs a lot of labor to find the suspicious person or overlooks the suspicious person.

  Therefore, in the authentication method according to the present invention, the face image of the PC user is registered together with an ID for personal identification, and the above-described monitoring image is classified based on this ID ((2 in FIG. 1). )reference). For example, when the PC user is “Mr. A”, the face image of Mr. A and the ID of Mr. A (“001” in FIG. 1) are registered in advance. Then, by collating the face image included in the monitoring image with the face image of Mr. A, the images are classified into images whose ID could be determined as “001” and images whose ID could not be determined.

  When the PC user is only Mr. A, the above-mentioned monitoring images often include a lot of Mr. A's own images, and this large number of personal images has been an obstacle to finding suspicious person images. However, according to the authentication method according to the present invention, it is possible to efficiently classify Mr. A's image (person's image) and images other than Mr. A's (other person's image), and display only images other than Mr. A. As a result (see (3) in FIG. 1), it is possible to reduce the effort of finding a suspicious person and to improve security by preventing oversight of a suspicious person.

  When there are a plurality of registered IDs, the images are classified into images corresponding to each ID and images not corresponding to any ID. For example, when the face images and IDs of 10 people are registered, the monitoring images are classified for each ID assigned to each of the 10 people, and the monitoring images that do not correspond to any ID are “images with unknown IDs”. Will be classified. By doing in this way, it can also be used for the purpose of recording the ID of the person who visited when the PC user was absent.

  For example, in a corporate office or the like, if a personal PC is assigned to each employee and the logon authority of each PC is limited to the user himself / herself, the ID and face image of all employees can be verified on each PC. The history of the ID or name of the employee who visited the seat when the user himself / herself was absent can be created. Further, in this case, since the monitoring image in which face detection or face matching has failed is classified as an image with an unknown ID, it is possible to easily find a suspicious person.

  As described above, in the authentication method according to the present invention, the face image of the person to be authenticated and the personal identification ID are registered in advance, the PC user is authenticated when the PC logon is permitted, and the ID acquired at the time of authentication is obtained. Based on the above, the surveillance images acquired when the PC user is absent are classified. Therefore, the convenience of the user can be improved by reducing the check load of the monitoring image, and the security strength can be increased by reducing the difficulty of suspicious person confirmation.

  In the above description, the case where the monitoring image is classified based on the ID acquired at the time of authentication has been described. However, the image to be classified includes the authentication image captured at the time of authentication. Since the authentication image is an image of a person approaching the PC, there is a high possibility that a person who is trying to operate the PC when the PC user is absent is captured. Therefore, by classifying the authentication image by the same method as the classification of the monitoring image, it is possible to reduce the effort for finding a suspicious person.

  In the following first embodiment, a basic configuration of an authentication apparatus to which the above-described authentication method is applied will be described, and in a second embodiment, a modification of the authentication apparatus shown in the first embodiment will be described.

  FIG. 2 is a block diagram illustrating the configuration of the authentication apparatus according to the first embodiment. As shown in the figure, the authentication device 1 includes a human sensor 2, a wide-angle camera 3, a display unit 4, a storage unit 11, and a control unit 12. The storage unit 11 includes authentication record information. 11a, registration information 11b, and image recording information 11c are stored. The control unit 12 includes an authentication start unit 12a, an authentication processing unit 12b, an authentication recording unit 12c, a face matching unit 12d, a monitoring processing unit 12e, an image recording unit 12f, a display processing unit 12g, and a classification. And a processing unit 12h.

  The human sensor 2 is provided on the upper part of the display of the PC and detects whether or not the PC user (person to be authenticated) is within a predetermined distance by detecting (receiving) infrared rays emitted from a person. It is a device to do. The wide-angle camera 3 is provided in the upper part of the display of the PC like the human sensor 2 and images an authentication target person who is going to be authenticated (attempts to log in), and the authentication target person who has successfully logged in This is a camera for picking up the situation in front of the PC as a monitoring image when the person leaves. The wide-angle camera 3 has a wide imaging range and a deep depth of field, so that it can be focused over a wide range.

  In the authentication device 1 according to the first embodiment, the authentication image 1 and the monitoring image are picked up using the single wide-angle camera 3 as described above, so that the cost of the authentication device 1 can be suppressed. it can. The display unit 4 is a display device such as a liquid crystal display, and the above-described PC display or the like is used. For example, a login screen is displayed on the display unit 4 at the time of authentication at the time of login, a monitoring screen is displayed at the time of monitoring when the PC user is absent, and a monitoring image confirmation screen is displayed at the time of confirmation of the monitoring image.

  The storage unit 11 is a storage device such as a hard disk device, and stores authentication record information 11a, registration information 11b, and image record information 11c. The authentication record information 11a is information obtained by accumulating a logon authentication result to the PC. Here, an example of the authentication record information 11a will be described with reference to FIG. FIG. 3 is a diagram illustrating an example of the authentication record information 11a.

  As shown in FIG. 3, the authentication record information 11a includes a “time” indicating the time when the authentication was performed, an “authentication success / failure” indicating whether the authentication was successful or unsuccessful, and an individual of the authentication target person registered in advance. This information includes “personal ID” for identifying the ID. For example, the rows 31, 32, and 36 in the figure indicate that the person with the personal ID “001” has been successfully authenticated.

  Also, in FIG. 33, the authentication failed and the personal ID of the person to be authenticated could not be acquired. In FIG. 34, although the authentication failed, the personal ID ( 005) can be obtained. Furthermore, in FIG. 35, it is shown that the authentication is successful and the personal ID (002) of the person to be authenticated can be acquired.

  In FIG. 3, it is assumed that face images with personal IDs “001” to “005” are registered in association with the personal ID, and the personal IDs “001” and “002” are registered. It is assumed that logon is permitted. Further, FIG. 3 shows a case where a plurality of people are permitted to log on. However, for example, it may be set so that only a person whose personal ID is “001” is permitted to log on.

  Returning to the description of FIG. 2, the registration information 11b will be described. The registration information 11b is information including the above personal ID and a face image captured in advance for face collation, and has previously registered information on a person who is permitted to log on to the PC and a person to be monitored. Is. Instead of storing the registration information 11b in the storage unit 11 in advance, the registration information 11b may be stored in a server device on the network and downloaded to the storage unit 11 as appropriate.

  The image recording information 11c is information obtained by accumulating data related to the monitoring image. Here, an example of the image recording information 11c will be described with reference to FIG. FIG. 4 is a diagram illustrating an example of the image recording information 11c. As shown in the figure, the image recording information 11c includes a “mode” indicating whether the image was captured during monitoring or face authentication, a “time” indicating the time when the image was acquired, a face “Type” indicating success or failure of detection or authentication, “Personal ID” indicating the identified ID, “Image No.” indicating an image group forming the image group, and an image group corresponding to “Image No.” Information including “representative No.” representing The image corresponding to the “representative No.” is an image used as the “thumbnail” described above.

  For example, the data in the 45th row in FIG. 4 indicates that the data is captured in the “face authentication” mode, and the data in other rows is the data captured in the “monitoring” mode. Show. Of the data captured in the “monitor” mode, 42 to 44 indicate that face detection was successful, and 42 and 44 were able to acquire a personal ID after successful face matching. It shows that.

  Further, as shown in each row of FIG. 4, the acquired image is an image group composed of a plurality of images. For example, the acquisition of this image group is triggered by the occurrence of a change in the image captured by the wide-angle camera 3, and the acquisition is terminated when the change in the image disappears. In this way, by managing a group of a series of images, it is possible to efficiently classify related images. In the first embodiment, the case where captured images are managed for each image group has been described. However, each image (one image) may be managed.

  Returning to the description of FIG. 2, the control unit 12 will be described. The control unit 12 is a processing unit that performs a process of classifying the images acquired in the monitoring process and providing it to the user based on the personal ID acquired in the authentication process while switching between the authentication process and the monitoring process. . Specifically, the image captured by the wide-angle camera 3 is stored in the storage unit 11 as the image recording information 11c via the image recording unit 12f, and the personal ID acquired by the authentication processing unit 12b is stored in the authentication recording unit 12c. And stored in the storage unit 11 as authentication record information 11a. Then, the classification processing unit 12h classifies the image recording information 11c based on the authentication recording information 11a, and the display processing unit 12g displays the classified information on the display unit 4.

  When the human sensor 2 detects the person to be authenticated, the authentication start unit 12a gives an instruction to start the authentication process to the authentication processing unit 12b by passing the image captured by the wide-angle camera 3 to the authentication processing unit 12b. It is a processing part to output. Further, upon receiving an authentication processing start instruction from the authentication start unit 12a, the authentication processing unit 12b instructs the face matching unit 12d to perform face matching between the captured image and the registered image included in the registration information 11b. Is a processing unit that performs a process of outputting to the authentication recording unit.

  The authentication recording unit 12c is a processing unit that performs processing for storing the authentication result (see FIG. 3) received from the authentication processing unit 12b in the storage unit 11 as the authentication record information 11a. The face collating unit 12d is a processing unit that performs a process of collating the face image included in the registration information 11b with the captured image captured by the wide-angle camera 3. For example, the face matching unit 12d extracts a face image included in the captured image, corrects the orientation, position, size, and the like of the face, and calculates the feature amount of the corrected face image. Then, the face matching is performed by comparing the calculated feature amount with the feature amount of the face image included in the registration information 11b.

  When the captured image is received from the wide-angle camera 3 and a change appears in the captured image, the monitoring processing unit 12e instructs the image recording unit 12f to record the received captured image, and the change in the captured image is detected. When there is no more, the processing unit performs processing for instructing to stop recording of the captured image.

  The monitoring processing unit 12e is also a processing unit that detects a face image included in the captured image by using the face matching unit 12d and identifies a personal ID corresponding to the detected face image. The image recording unit 12f is a processing unit that performs processing for storing information including information such as a personal ID added by the monitoring processing unit 12e (see FIG. 4) in the storage unit 11 as image recording information 11c. .

  The classification processing unit 12h is a processing unit that performs processing for classifying the image recording information 11c based on the authentication recording information 11a. For example, the classification processing unit 12h extracts the personal ID of the person who has been successfully authenticated from the authentication record information 11a, and associates the personal ID with the personal ID included in the image record information 11c, thereby obtaining the same individual. “Marking” such as setting a flag on the image information related to the ID is performed. Here, this association will be described with reference to FIG. FIG. 5 is a diagram illustrating an example of association in the classification process.

  As shown in FIG. 5, the operation state of the authentication device 1 includes the above-described monitoring mode and authentication mode. When the human sensor 2 detects the authenticator, the authentication processing unit 12b starts operating according to an instruction from the authentication start unit 12a and shifts to the authentication mode. Note that the mode transition from the authentication mode to the monitoring mode is executed by using, for example, an input to a PC input device (such as a keyboard or a mouse) not being performed for a predetermined time.

  When the mode is changed from the monitoring mode to the authentication mode, an authentication face image (authentication image) is captured as indicated by 53 in FIG. When the face authentication is successful, the personal ID of the person to be authenticated is determined (“001” in FIG. 5). The found personal ID is recorded as authentication record information 11a.

  The classification processing unit 12h searches the image recording information 11c including the images (51 and 52 in FIG. 5) captured in the monitoring mode by using the found personal ID. For example, if the face matching has been successfully performed on the image 52 shown in FIG. 5 and it is already known that the personal ID of this image is 001, the above-described “marking” is performed on this image. . Further, the classification processing unit 12h similarly performs “marking” on the authentication image (53 in FIG. 5). If the deletion process is performed instead of “marking”, the storage area of the storage unit 11 can be used effectively.

  FIG. 5 shows the case where the association processing performed by the classification processing unit 12h is performed based on whether the personal IDs match. However, the association is not limited to this and may be performed based on other criteria. For example, based on the time when the face authentication of the person to be authenticated succeeded as a reference, a monitoring image or an authentication image captured within a predetermined time from this time may be associated with the personal ID of the successful authentication person. By doing in this way, the image captured when the person to be authenticated sits or leaves the seat can be efficiently classified from the images captured by other persons.

  Returning to the description of FIG. 2, the display processing unit 12g will be described. The display processing unit 12g is a processing unit that performs processing for displaying the image recording information 11c classified by the classification processing unit 12h on the display unit 4. For example, the display processing unit 12g displays the recorded image on the display unit 4 after excluding the image relating to the person who has successfully authenticated the face and logged into the PC. By doing in this way, it becomes possible to solve the difficulty of suspicious person confirmation resulting from the large number of personal images. An example of a screen displayed on the display unit 4 by the display processing unit 12g will be described later with reference to FIG.

  Next, the processing procedure of the classification process performed by the above-described classification processing unit 12h will be described with reference to FIG. FIG. 6 is a flowchart showing the processing procedure of the classification process. As shown in the figure, when the authentication processing unit 12b acquires the personal ID of the person to be authenticated (hereinafter referred to as “authentication ID”) (step S101), the classification processing unit 12h corresponds to the acquired authentication ID. A mark is added to the authentication image (step S102), and a mark is also added to the monitoring image corresponding to the authentication ID (step S103).

  Subsequently, the classification processing unit 12h classifies the authentication image and the monitoring image based on the personal ID (hereinafter referred to as “registration ID”) included in the registration information 11b (step S104). For example, when five personal IDs “001” to “005” are registered as registration IDs, the authentication image and the monitoring image are classified for each personal ID “001” to “005”, and Images that do not fall under the personal ID are classified as other images. Then, the display processing unit 12g displays the image recording information 11c classified in this way on the display unit 4 (step S105), and the processing is ended.

  Next, the image recording information 11c subjected to the classification process by the classification processing unit 12h will be described with reference to FIG. FIG. 7 is a diagram illustrating an example of data classified by the classification process. 7 shows an example of a data folder when the acquired images are classified based on the personal ID. In this way, each data folder is provided for each personal ID, and when the personal ID is unknown, an “unknown” folder is provided. Note that reference numeral 71 in FIG. 7 shows a case where two folders are provided as the “unknown” folder, when the face detection is successful and when face detection fails.

  7 shows an example of a data folder when the acquired images are classified based on the time when each image is acquired. Note that a data folder for each personal ID may be created hierarchically in each data folder indicated by 72 in FIG. 7, and a data folder for each image acquisition time is indicated hierarchically in each data folder indicated by 71 in FIG. It may be created.

  Next, a screen example displayed on the display unit 4 by the display processing unit 12g will be described with reference to FIG. FIG. 8 is a diagram illustrating an example of a screen that displays classified images. The figure shows an example of a screen that displays the authentication image and the monitoring image separately. When the authentication recording button 80d is pressed, the screen shown in 81 is displayed when the monitoring recording button 80e is pressed. Each of the screens shown in 82 is displayed.

  As indicated by 81 in FIG. 8, when Mr. X having the personal ID “001” is logged on (see 80c in FIG. 8), an authentication image other than the personal ID “001” is displayed in the image display area 80b. A representative image (thumbnail) is displayed. By this display, Mr. E (personal ID is “005”) at 10:50 and Mr. B (personal ID is “002”) at 11:01 while Mr. X is away. You can see that Also, at 10:47, it can be seen that a person (?) Whose personal ID is unknown has visited, so Mr. X can check the entire image group by clicking on the thumbnail of "?" Thus, it can be easily determined whether or not this “?” Is a suspicious person.

  Further, as indicated by 82 in FIG. 8, when Mr. X having the personal ID “001” is logged on (see 80c in FIG. 8), the image display area 80b has a monitor other than the personal ID “001”. A representative image (thumbnail) of the image is displayed. This display shows that Mr. B (personal ID is “002”) was around Mr. X's seat at 20:45 while Mr. X was away. In addition, it can be seen that the face detection failed in the monitoring image acquired at 20:40, and that the face detection was successful in the monitoring image acquired at 20:47, but the identification of the personal ID failed. . By checking the entire image group by clicking the monitoring image acquired at 20:40 or the thumbnail portion of “?”, Mr. X checks whether or not a suspicious person is captured in these monitoring images. Can be easily determined.

  Next, a processing procedure of authentication processing and monitoring processing performed in the authentication device 1 according to the first embodiment will be described with reference to FIG. FIG. 9 is a flowchart illustrating a processing procedure of authentication processing and monitoring processing performed in the authentication device 1 according to the first embodiment. Note that the flowchart of FIG. 9 shows the processing procedure of the authentication apparatus 1 in a state in which the mode is changed to the monitoring mode because the PC user leaves the seat or the like.

  While the PC user is away, the display processing unit 12g displays a monitoring screen on the display unit 4 (step S201). The authentication start unit 12a determines whether the human sensor 2 has detected a person by monitoring the output of the human sensor 2 (step S202). When the human sensor 2 detects a person (Yes at Step S202), the display processing unit 12g displays a face authentication screen on the display unit 4 (Step S203).

  On the other hand, when the human sensor 2 has not detected a person (No at Step S202), the monitoring processing unit 12e determines whether or not moving object detection has been performed based on a change in an image captured by the wide-angle camera 3 or the like. (Step S204). If a moving object is detected (step S204, Yes), the face matching unit 12d performs a face matching process (step S205), and stores a monitoring image including information on whether or not a personal ID is found as image recording information 11c. The data is stored in the unit 11 (step S206), and the processing after step S201 is repeated. On the other hand, when a moving object is not detected (step S204, No), the process after step S201 is repeated without performing a face collation process etc.

  Next, the processing procedure after step S203 will be described. First, the authentication processing unit 12e determines whether or not the face detection is successful (step S207). If the face detection is successful (step S207, Yes), then the face matching unit 12d is used to determine the face. It is determined whether or not the collation is successful (step S208). If the face matching is successful (step S208, Yes), the classification processing unit 12h performs the classification process shown in FIG. 6 (step S210).

  On the other hand, when face detection fails (step S207, No) and face matching fails (step S208, No), it is determined whether or not the failure is within a predetermined number of times (step S209). If it is within the predetermined number of times (step S209, Yes), the processing after step S207 is repeated. On the other hand, when the predetermined number of times has been exceeded (No at Step S209), the processes after Step S201 are repeated.

  Following the processing procedure of step S210, the display processing unit 12g ends the face authentication screen displayed on the display unit 4 (step S211), and displays the image display screen shown in FIG. 8 (step S212). To provide the PC user with an image of being away from the desk and finish the process.

  As described above, in the first embodiment, when the authentication image and the monitoring image are captured by the wide-angle camera and the human sensor detects the person to be authenticated, the authentication processing unit determines the verification result between the authentication image and the registration information. The authentication record information including the personal ID is stored in the storage unit. The captured image is face-matched by the monitoring processing unit and stored in the storage unit as image recording information. The classification processing unit classifies the image recording information based on the personal ID, and classifies the image recording. Since the display processing unit is configured to display the information on the display unit, it is possible to improve the convenience for the user by reducing the check load of the monitoring image, and to reduce the difficulty of confirming the suspicious person Security can be improved.

  In the following second embodiment, a modified example of the authentication apparatus shown in the first embodiment will be described. In the second embodiment, a case will be described in which two cameras, that is, an authentication camera and a monitoring camera are provided, and face matching of the monitoring image is performed after the monitoring image is acquired. In the following, differences from the first embodiment will be mainly described, and description of common points will be omitted.

  In the second embodiment, an authentication device 1a that is a modification of the authentication device 1 according to the first embodiment will be described. FIG. 10 is a block diagram illustrating the configuration of the authentication device 1a according to the second embodiment. As shown in the figure, the authentication device 1a includes two cameras, a camera 5 for authentication processing and a wide-angle camera 3 for monitoring processing. The authentication processing camera 5 is connected to the authentication processing unit 12b through the authentication processing starting unit 12a together with the human sensor 2, and the wide-angle camera 3 is connected to the monitoring processing unit 12e.

  Here, the layout of the human sensor and the camera will be described with reference to FIG. FIG. 11 is a diagram illustrating a layout example of the human sensor and the camera. The layout example shown in FIG. 111 is a layout example in the first embodiment, and the layout example shown in FIG. 112 is a layout example in the second embodiment.

  First, a layout example according to the first embodiment will be described. In the first embodiment, a single wide-angle camera 3 that is common to the monitoring process and the authentication process is used. The wide-angle camera 3 is approaching the authentication target person 110a to receive authentication, or passes near the seat or the seat. The person 110b who is going to approach is imaged. When the human sensor 2 detects the authentication target person 110a, the face image of the authentication target person 110a is captured.

  On the other hand, in the second embodiment, the wide-angle camera 3 is used as a monitoring processing camera, and a person 110b who passes near the seat or approaches the seat is imaged. Further, when the camera 5 is used as an authentication processing camera and the human sensor 2 detects the authentication target person 110a, the face image of the authentication target person 110a is captured.

  Returning to the description of FIG. 10, the description of the differences from the first embodiment will be continued. In the authentication device 1a according to the second embodiment, when the monitoring processing unit 12e receives a monitoring image captured by the wide-angle camera 3, the face recognition processing is not performed on the storage unit 11 via the image recording unit 12f. The point which memorize | stores a monitoring image differs from Example 1. FIG. Then, at a predetermined timing, for example, when a person who has succeeded in authentication logs on to the PC, the classification processing unit 12h uses the face matching unit 12d to perform face matching processing of the monitoring image included in the image recording information 11c. Do. In this way, the processing load of the monitoring image recording process can be reduced, and the face matching process can be executed at an arbitrary timing.

  As described above, in the second embodiment, since the monitoring image and the authentication image are captured by separate cameras, the imaging area to be imaged can be set flexibly. Further, since the face collation for the monitoring image can be executed at an arbitrary timing, the processing load associated with the monitoring image recording process can be reduced.

  Next, a system configuration when the authentication device 1 and the authentication device 1a according to the first and second embodiments are connected to a network will be described with reference to FIG. FIG. 12 is a diagram illustrating a network example in which the authentication device 1 or the authentication device 1a is connected. Although only the authentication device 1 is shown in the figure, an authentication device 1a may be connected instead of these authentication devices 1. 120a to 120c shown in the figure represent rooms in a corporate office or the like.

  As shown in FIG. 12, the authentication devices 1 are connected to each other via a communication line such as a LAN (Local Area Network). The LAN described above is connected to the server device via a network such as the Internet. This server device collectively manages the registration information 11b registered in the storage unit 11 of each authentication device 1, and generates a history (log) of authentication record information 11a and image record information 11c stored in the storage unit 11. It is a computer that performs processing. The server apparatus may be directly connected to the LAN shown in FIG. Moreover, you may comprise so that the predetermined | prescribed authentication apparatus 1 may perform the function equivalent to a server apparatus. With such a configuration, it is possible to monitor a wide range (for example, the entire room).

  In each of the embodiments described above, the authentication apparatus that performs the authentication process based on the face image has been described. However, the present invention is not limited to this, and the authentication is based on other biometric information (for example, fingerprint, iris, blood vessel pattern, etc.). An authentication apparatus that performs processing may be configured.

  As described above, the authentication apparatus and the authentication method according to the present invention are useful when it is desired to execute both the authentication process and the monitoring process. In particular, the monitoring image or the recorded image of the authentication image is used for the convenience of the user. Suitable when you want to provide without losing.

It is a figure which shows the characteristic of the authentication apparatus which concerns on this invention. 1 is a block diagram illustrating a configuration of an authentication device according to Embodiment 1. FIG. It is a figure which shows an example of authentication record information. It is a figure which shows an example of image recording information. It is a figure which shows an example of the correlation in a classification | category process. It is a flowchart which shows the process sequence of a classification process. It is a figure which shows the example of data classified by the classification process. It is a figure which shows the example of a screen which displays the classified image. 3 is a flowchart illustrating a processing procedure of authentication processing and monitoring processing performed in the authentication device according to the first embodiment. FIG. 6 is a block diagram illustrating a configuration of an authentication device according to a second embodiment. It is a figure which shows the example of a layout of a human sensitive sensor and a camera. It is a figure which shows the example of a network which connected the authentication apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, 1a Authentication apparatus 2 Human sensor 3 Wide-angle camera 4 Display part 5 Camera 11 Storage part 11a Authentication recording information 11b Registration information 11c Image recording information 12 Control part 12a Authentication start part 12b Authentication processing part 12c Authentication recording part 12d Face collation part 12e Monitoring processing unit 12f Image recording unit 12g Display processing unit 12h Classification processing unit

Claims (14)

An authentication device that records a captured image as a monitoring image and authenticates a person to be authenticated by using the captured face image as an authentication image when performing personal authentication,
Storage means for storing the identification information for identifying the authentication subject and the registration image of the authentication subject in association with each other;
Classification means for classifying the monitoring image and / or the authentication image into an image corresponding to the identification information and an image not corresponding to the monitoring image and / or the authentication image based on the identification information stored in the storage means Authentication device. The classification means includes
The authentication apparatus according to claim 1, wherein if the identity authentication is successful, the classification of the monitoring image and / or the authentication image is performed. The classification means includes
The authentication apparatus according to claim 1, wherein when the image is recorded, the classification of the monitoring image and / or the authentication image is performed. An authentication recording means for recording the identification information corresponding to the person to be authenticated when the identity authentication is successful;
The classification means includes
The monitoring image and / or the authentication image are classified into an image corresponding to the identification information and an image not corresponding to the monitoring image and / or the authentication image based on the identification information recorded by the authentication recording unit. 3. The authentication device according to 3.   5. The authentication apparatus according to claim 4, further comprising a display means for displaying an image that does not correspond to the identification information recorded by the authentication recording means among the images classified by the classification means. The display means includes
6. The authentication apparatus according to claim 5, wherein among the images classified by the classification unit, the authentication image that has failed in the personal authentication is displayed separately from the monitoring image. The display means includes
The authentication apparatus according to claim 5 or 6, wherein the classification unit displays an image classified as an image corresponding to the identification information stored in the storage unit for each identification information.   The image processing apparatus according to any one of claims 4 to 7, further comprising a deletion unit that deletes an image corresponding to the identification information recorded by the authentication recording unit from among the images classified by the classification unit. The authentication device described.   The authentication apparatus according to claim 1, wherein the monitoring image and the authentication image are captured by a single camera.   9. The monitoring image and the authentication image are respectively captured by two cameras, the monitoring image capturing camera and the authentication image capturing camera. Authentication device. An authentication method for authenticating a person to be authenticated by using a captured face image as an authentication image when recording a captured image as a monitoring image and performing personal authentication,
A storage step of storing the identification information for identifying the authentication subject and the registration image of the authentication subject in association with each other;
And a classification step of classifying the monitoring image and / or the authentication image into an image corresponding to the identification information and an image not corresponding to the image based on the identification information stored in the storage step. Authentication method. The classification step includes
The authentication method according to claim 11, wherein if the identity authentication is successful, the classification of the monitoring image and / or the authentication image is performed. The classification step includes
The authentication method according to claim 11, wherein the classification of the monitoring image and / or the authentication image is performed when the image is recorded. An authentication recording step of recording the identification information corresponding to the person to be authenticated when the identity authentication is successful;
The classification step includes
13. The monitoring image and / or the authentication image are classified into an image corresponding to the identification information and an image not corresponding to the monitoring image and / or the authentication image based on the identification information recorded in the authentication recording step. 14. The authentication method according to 13.

Images