JP2010176304A - Personal authentication apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a personal authentication apparatus that always properly authenticates a person by flexibly changing an authentication threshold. <P>SOLUTION: The personal authentication apparatus acquires a face image of a person to be authenticated, matches the acquired face image against a face image registered beforehand to compute a similarity between the two images, and compares the computed similarity with an authentication threshold to determine whether the person to be authenticated has been already registered or not. Ambient brightness information is acquired as external environment information for acquiring the face image of the person to be authenticated, and the authentication threshold is changed in accordance with the acquired ambient brightness information. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention is used in, for example, an entrance / exit management system, and acquires an image including at least a face as biometric information of a person to be authenticated, and compares the acquired face image with a pre-registered face image The present invention relates to a person authentication apparatus that determines the similarity between both images and compares the calculated similarity with an authentication threshold value to determine whether the authentication target person is a person registered in advance.

  In general, this type of person authentication apparatus acquires a face image of a person who has been permitted as an authentication target in advance using a camera, a sensor, or the like, and stores unique feature information obtained from the acquired face image as dictionary information. It is stored (registered) in the means.

  Then, at the time of authentication (at the time of face authentication), the face image of the person to be recognized is acquired again using a camera, a sensor, or the like, and unique feature information obtained from the acquired face image is stored in advance in the dictionary storage means ( The similarity between the two information is obtained by collating with the registered dictionary information, and the obtained similarity is compared with a preset threshold for authentication, so that the person to be authenticated is a person registered in advance. Whether or not there is is determined (for example, see Patent Document 1).

  Conventionally, regarding the setting of the threshold for authentication in such a person authentication device, it is possible to perform accurate authentication according to the person to be authenticated by setting the threshold for authentication in advance according to the characteristics of the person to be authenticated (individual). This method is known (for example, see Patent Document 2).

JP-A-9-319877 JP 2001-21309 A

  In the method of Patent Document 2, even if a threshold value for authentication is set in advance according to the characteristics of the person to be authenticated (individual), for example, in the case of a person authentication method such as face authentication, depending on the external environment at the time of face authentication There is a high possibility that accurate person authentication cannot be performed.

  For example, when the threshold value for authentication is set high in advance, the image quality of the face image acquired when the periphery at the time of face authentication is dark is poor, and the calculated similarity value is lower than the threshold value for authentication even for the person himself / herself. There is a possibility that it may become, and accurate person authentication cannot be performed. In addition, if the method is such that the threshold value for authentication set in advance is lowered in anticipation of this, there is a possibility of accepting a person other than the person himself / herself.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a person authentication apparatus that can flexibly change an authentication threshold and can always perform accurate person authentication.

  The personal authentication device of the present invention includes a dictionary storage unit that stores biometric information acquired from a person who has been permitted as an authentication target in advance as dictionary information, a biometric information acquisition unit that acquires biometric information from the authentication target person, Similarity calculation means for determining the similarity between the biometric information acquired by the biometric information acquisition means and the dictionary information stored in the dictionary storage means, and the similarity calculated by the similarity calculation means Is compared with a predetermined threshold for authentication to determine whether the person to be authenticated is a person registered in advance, and the biometric information of the person to be authenticated is acquired by the biometric information acquisition means. External environment information acquisition means for acquiring external environment information at the time, and the authentication threshold based on the external environment information acquired by the external environment information acquisition means And an authentication threshold changing means for changing.

  In addition, the person authentication device of the present invention includes a dictionary storage unit that stores biometric information acquired from a person permitted as an authentication target in advance as dictionary information, and a biometric information acquisition unit that acquires biometric information from the authentication target person. The similarity calculation means for determining the similarity between the biological information acquired by the biological information acquisition means and the dictionary information stored in the dictionary storage means, and the similarity calculation means By comparing the similarity with a preset threshold for authentication, it is determined whether or not the authentication target person is a person registered in advance, and the biometric information of the authentication target person is obtained by the biometric information acquisition means. External environment information acquisition means for acquiring external environment information at the time of acquisition, and when the biological information of the person to be authenticated is acquired by the biological information acquisition means, Biometric information-accompanying environment information acquisition means for acquiring environmental information associated with the report, external environment information acquired by the external environment information acquisition means, and biometric information-related environment information acquired by the biometric information-accompanying environment information acquisition means, Authentication threshold value changing means for changing the authentication threshold value based on the above.

  ADVANTAGE OF THE INVENTION According to this invention, the threshold value for authentication can be changed flexibly and the person authentication apparatus which can always perform exact person authentication can be provided.

1 is a block diagram schematically showing a configuration of a person authentication device according to a first embodiment of the present invention. The figure which shows the structure of the threshold value table which concerns on 1st Embodiment. The flowchart explaining the process of acquisition of external environment information and threshold value change which concern on 1st Embodiment. The flowchart explaining the collation process which concerns on 1st Embodiment. The figure which shows the structure of the threshold value table which concerns on the 2nd Embodiment of this invention. The flowchart explaining the process of acquisition of external environment information and threshold value change which concern on 2nd Embodiment.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
In the following description, a case where a face image is used as the biological information of a person will be described as an example. However, other biological information such as an iris pattern, a vein pattern, and a fingerprint pattern can be used in the same manner. .

  The place where the person authentication device according to the present invention is used is, for example, a place where only a passerby permitted in advance is allowed to pass, an important customer as a specific person or a person who needs attention from passers-by. A case where it is desired to recognize that a problem has occurred is assumed. For example, an entry / exit management system that manages passers-by entrances and exits of buildings and corporate buildings, or applications that recognize passers-by in places where many passers-by, such as commercial facilities, entertainment facilities, and transportation facilities, are assumed. Is done.

First, the first embodiment will be described.
FIG. 1 schematically shows the configuration of a person authentication apparatus according to the first embodiment of the present invention. This person authentication device includes a registration terminal device 11 for registering a face image as biometric information, a collation terminal device 12 for collating face images, an external environment information acquisition device 13 as external environment information acquisition means, and each of these devices. The collating server 16 is configured to be communicably connected to the devices 11 to 13 via communication lines (networks) 14 and 15.

  The registration terminal device 11 is for, for example, registering a face image of a person (hereinafter referred to as a passerby or authentication target person) M who is permitted as an authentication target in advance, and includes an image (hereinafter referred to as at least a face of the person M). A video camera (hereinafter simply referred to as a camera) 21 that captures and inputs a face image) and performs various operation guidance for the person M, or the person M performs a predetermined operation. It is configured to have an operation unit 22 or the like to perform.

  The collation terminal device 12 is for collating the face image of the passer-by M, for example, at the entrance of a building or room, and captures an image including at least the face of the passer-by M (hereinafter referred to as a face image). A video camera (hereinafter simply referred to as a camera) 23 as a biometric information acquisition means to be inputted, and an operation unit 24 for performing various operation guidance for the passer-by M and for the passer-by M to perform a predetermined operation. It is comprised.

The external environment information acquisition device 13 is installed in the vicinity of the verification terminal device 12, for example, and acquires ambient brightness information as external environment information when the camera 23 acquires a face image of the passerby M. And an illuminometer or the like is used. In this example, the brightness information is output as a numerical value from “0 to 999”.
In this example, the external environment information acquisition device 13 also has a person detection sensor function that detects a passerby M approaching the verification terminal device 12.

  The collation server 16 is mainly composed of a personal computer or the like, for example, and a dictionary storage means for storing a face image of the person M registered in the registration terminal device 11 and an ID as unique identification information as dictionary information, Similarity calculated by the similarity calculating means for determining the similarity between the face image of the passerby M acquired by the verification terminal device 12 and the dictionary information stored in the dictionary storing means, and the similarity calculated by the similarity calculating means. A determination means for determining whether or not the passer-by M is a person registered in advance by comparing the highest similarity of the degrees with a preset threshold for authentication, and an external environment information acquisition device Based on the brightness information (external environment information) acquired in step 13, the threshold value table for authentication is used to change the threshold value for authentication to change the threshold value for authentication. To.

  Further, for example, as shown in FIG. 2, the verification server 16 includes a threshold value table that stores authentication threshold values that are optimal for the brightness information in association with a plurality of types of brightness information (external environment information). Yes. That is, in this example, the threshold value “50” for the brightness information “0 to 99”, the threshold value “55” for the brightness information “100 to 199”, and the threshold value for the brightness information “200 to 299”. “60”,..., Brightness information “900 to 999” is stored in correspondence with a threshold value “95”.

Next, the operation according to the first embodiment in such a configuration will be described.
First, the registration process will be described.
The registered terminal device 11 acquires the face image of the person M to be authenticated, sends the acquired face image to the verification server 16 via the communication line 14, and stores it as dictionary information in the dictionary storage means in the verification server 16.

Next, external environment information acquisition and threshold value change processing will be described with reference to the flowchart shown in FIG.
When the external environment information acquisition device 13 detects a passerby M approaching (step S1), the external environment information acquisition device 13 acquires brightness information around the verification terminal device 12 (step S2), and transmits the acquired brightness information to the communication line 15. To the verification server 16 (step S3).

  The verification server 16 receives the brightness information sent from the external environment information acquisition device 13 and temporarily stores it in a memory (not shown) (step S4). Next, the matching server 16 refers to the threshold value table of FIG. 2 based on the temporarily stored brightness information, thereby acquiring the corresponding authentication threshold value (new threshold value) (step S5), and the acquired authentication threshold value. Based on the above, the preset threshold for authentication is updated (step S6).

  Specifically, using FIG. 2, for example, when the brightness information sent from the external environment information acquisition device 13 is “500”, the brightness information is referred to by referring to the threshold value table of FIG. Since the threshold value corresponding to “500” is “75”, the threshold value “75” is acquired and rewritten as a new authentication threshold value.

Next, the collation process will be described with reference to the flowchart shown in FIG.
As described above, when the external environment information acquisition device 13 detects the approacher M approaching, the verification terminal device 12 acquires the face image of the passer M and compares the acquired face image via the communication line 15. It transmits to the server 16 (step S11).

  The collation server 16 receives the face image sent from the collation terminal device 12, and obtains the similarity between the received passer M's face image and dictionary information stored in the dictionary storage means in advance. (Step S12).

Next, the matching server 16 compares the highest similarity among the obtained similarities with the authentication threshold set as described above, and if the obtained similarity is higher than the authentication threshold, the pass It is determined that the person M is a person registered in advance, and when the obtained similarity is lower than the threshold for authentication, it is determined that the passer-by M is not a person registered in advance, and the determination result is used as the collation terminal device 12. (Step S13).
The verification terminal device 12 displays the determination result transmitted from the verification server 16 on the operation unit 24 and notifies the passerby M (steps S14 and S15).

  As described above, according to the first embodiment described above, it is possible to change the authentication threshold flexibly by considering external environment information (peripheral brightness information) when performing person authentication. Accurate person authentication can be performed.

Next, a second embodiment will be described.
Since the system configuration according to the second embodiment is basically the same as that of the first embodiment (FIG. 1) described above, illustration is omitted, but the threshold table in the verification server 16 is different. In addition to the function of acquiring the surrounding brightness information when the external environment information acquisition device 13 acquires the face image of the passer M by the camera 23, when acquiring the face image of the passer M, It also has a function as biometric information-accompanied environment information acquisition means for acquiring brightness information of a face portion as environment information associated with the face image, peripheral brightness information and face acquired by the external environment information acquisition device 13 The difference is that it has a function as an environmental information difference calculation means for obtaining a difference from the brightness information of the part.

  FIG. 5 shows a configuration of a threshold table in the verification server 16 according to the second embodiment. This threshold value table stores authentication threshold values that are optimal for the differences corresponding to the differences between the brightness information (biological information accompanying environment information) of a plurality of types of face portions and the surrounding brightness information (external environment information). is doing. That is, in this example, the threshold “40” is set for the difference “0 or less” between the brightness information of the face portion and the surrounding brightness information, the threshold “50” is set for the difference “0-19”, and the difference “20” is set. ˜39 ”and the threshold value“ 90 ”for the difference“ 80 to 100 ”.

Next, the operation according to the second embodiment in such a configuration will be described.
Since the registration process and the collation process are the same as those in the first embodiment described above, the description thereof will be omitted, and only the external environment information acquisition and threshold value change processes different from those in the first embodiment will be described with reference to FIG. This will be described with reference to the flowchart shown in FIG.

  When the external environment information acquisition device 13 detects a passerby M approaching (step S21), the external environment information acquisition device 13 acquires brightness information around the collation terminal device 12, and also acquires brightness information of the face portion of the passerby M. Then, the acquired brightness information is transmitted to the verification server 16 via the communication line 15 (step S23).

  The collation server 16 receives the brightness information of the face part and the brightness information of the surroundings sent from the external environment information acquisition device 13, and temporarily stores them in a memory (not shown) (step S24). Next, the collation server 16 obtains a difference between the temporarily stored face brightness information and the surrounding brightness information (step S25).

  Next, the matching server 16 obtains a corresponding authentication threshold value (new threshold value) by referring to the threshold value table of FIG. 5 based on the difference between the obtained brightness information of the face portion and the surrounding brightness information. Then, based on the acquired authentication threshold value, a preset authentication threshold value is updated (step S27).

  Specifically, using FIG. 5, for example, the brightness information of the face portion sent from the external environment information acquisition device 13 is “600”, the brightness information of the surroundings is “650”, When the difference is “50”, referring to the threshold value table of FIG. 5, the threshold corresponding to “50” is “70”, so that the threshold “70” is acquired and a new authentication threshold is obtained. Can be rewritten as

  As described above, according to the second embodiment described above, the external environment information (peripheral brightness information) and the biometric information accompanying environment information (facial part brightness information) are taken into account when performing person authentication. Thus, the threshold value for authentication corresponding to the difference can be changed flexibly, and accurate person authentication can always be performed.

  DESCRIPTION OF SYMBOLS 11 ... Registration terminal device, 12 ... Collation terminal device, 13 ... External environment information acquisition device (external environment information acquisition means), 14, 15 ... Communication line (network), 16 ... Verification server, M ... Person (passerby, authentication) Subjects) 21, 23 ... Video cameras (biological information acquisition means), 22, 24 ... operation units.

Claims (4)

Dictionary storage means for storing biometric information acquired from a person who has been permitted as an authentication target in advance as dictionary information;
Biometric information acquisition means for acquiring biometric information from the person to be authenticated;
Similarity calculation means for obtaining the similarity between the biometric information acquired by the biometric information acquisition means and the dictionary information stored in the dictionary storage means,
Determining means for determining whether or not the person to be authenticated is a person registered in advance by comparing the similarity obtained by the similarity calculating means with a preset threshold for authentication;
External environment information acquisition means for acquiring external environment information when acquiring biometric information of the person to be authenticated by the biometric information acquisition means;
An authentication threshold value changing means for changing the authentication threshold value based on the external environment information acquired by the external environment information acquiring means;
A person authentication device comprising: Further comprising a threshold value table storing authentication threshold values that are optimal for the external environment information in correspondence with the plural types of external environment information,
The authentication threshold value changing unit acquires a corresponding authentication threshold value from the threshold table based on the external environment information acquired by the external environment information acquiring unit, and the authentication set in advance based on the acquired authentication threshold value The person authentication apparatus according to claim 1, wherein the threshold value for use is updated. Dictionary storage means for storing biometric information acquired from a person who has been permitted as an authentication target in advance as dictionary information;
Biometric information acquisition means for acquiring biometric information from the person to be authenticated;
Similarity calculation means for obtaining the similarity between the biometric information acquired by the biometric information acquisition means and the dictionary information stored in the dictionary storage means,
Determining means for determining whether or not the person to be authenticated is a person registered in advance by comparing the similarity obtained by the similarity calculating means with a preset threshold for authentication;
External environment information acquisition means for acquiring external environment information when acquiring biometric information of the person to be authenticated by the biometric information acquisition means;
When acquiring the biometric information of the person to be authenticated by the biometric information acquisition means, biometric information-accompanying environment information acquisition means for acquiring environmental information associated with the biometric information;
An authentication threshold value changing means for changing the authentication threshold value based on the external environment information acquired by the external environment information acquisition means and the biometric information-accompanying environment information acquired by the biometric information-accompanying environment information acquiring means;
A person authentication device comprising: Environmental information difference calculation means for obtaining a difference between the external environment information acquired by the external environment information acquisition means and the biological information-associated environment information acquired by the biological information-related environment information acquisition means;
Further comprising a threshold value table that stores a threshold value for authentication optimal for the difference corresponding to the difference between the multiple types of external environment information and the biometric information-associated environment information,
The authentication threshold value changing unit acquires a corresponding authentication threshold value from the threshold table based on the difference between the external environment information and the biological information-accompanying environment information obtained by the environment information difference calculation unit, and the acquired authentication threshold value 4. The person authentication apparatus according to claim 3, wherein the authentication threshold value set in advance is updated based on the threshold value.

Images