JP2023002170A - Biometric registration device, biometric authentication device, computer program, and biometric authentication method - Google Patents

Abstract

Kind Code: A1 In one-to-many biometric authentication processing, a distance index method is used to speed up processing and improve authentication accuracy.
A reference data set generation unit that performs principal component analysis on templates of biometric data of a plurality of registered users and generates a set of reference data for a distance index method from the results of the principal component analysis; an index recording unit that calculates a degree of similarity between the template and the data, and records an index of the degree of similarity between the template and the reference data based on the result of the degree of similarity calculation.
[Selection drawing] Fig. 2

Description

The present invention relates to a biometric registration device, a biometric authentication device, a computer program, and a biometric authentication method.

Conventionally, as a technique for speeding up one-to-many biometric authentication processing, for example, the technique described in Non-Patent Document 1 is known. The technology described in Non-Patent Document 1 uses the distance index method to select templates (search targets) for matching samples (biometric information at the time of authentication) from templates (biometric information at the time of registration) of all registered users. Narrow down.

Specifically, in the biometric information registration stage, a part (n pieces) of templates (x pieces) of all registered users are defined as pivots (reference data), and the distances between all the templates and the pivots are calculated in advance. (or similarity) is calculated, and an index indicating the distance (or similarity) between each template and each pivot is recorded. Then, at the time of authentication, the distance (or similarity) between the sample and the pivot is calculated, and the template to be searched from the correlation between the distance (or similarity) between the sample and each pivot and the registered index are narrowed down, and matching processing (one-to-one biometric authentication processing) with a sample is performed only for the templates resulting from the narrowing down. This reduces the number of times the matching process between the template and the sample is executed. In addition, the permutation index method is used to further reduce the number of times the matching process is executed.

"Proposal of Sequential Index Fusion Judgment for Large-Scale ID-less Biometric Authentication", Transactions of the Institute of Electronics, Information and Communication Engineers, Vol. J96-A No.12, pp.801-814, December 2013

However, in the technique described in Non-Patent Document 1, although a plurality of templates of different registered users are used for pivots, the distance (or similarity) between each sample pivot is concentrated in a narrow range. Therefore, there is a possibility that authentication accuracy will be insufficient. If the number of pivots is increased to deal with this problem, the number of times to calculate the distance (or similarity) between the sample and the pivot increases, which reduces the effect of speeding up the one-to-many biometric authentication process. .

The present invention has been made in consideration of such circumstances, and its object is, for example, in one-to-many biometric authentication processing related to biometric authentication such as palm print authentication, fingerprint authentication, face authentication, and iris authentication, in which the distance index method The object is to speed up the processing and improve the accuracy of authentication by means of the method.

(1) One aspect of the present invention is to generate a reference data set by performing principal component analysis on biometric data templates of a plurality of registered users and generating a set of reference data for a distance index method from the results of the principal component analysis. an index recording unit that calculates the degree of similarity between each template and the reference data, and records an index of the degree of similarity between the template and the reference data based on the result of calculating the degree of similarity; It is a biometric registration device comprising:
(2) An aspect of the present invention is the biometric registration device according to (1) above, wherein the reference data set generator performs principal component analysis on a reduced template obtained by reducing the image size of the template to a predetermined size. .
(3) According to one aspect of the present invention, a degree of similarity between each of a plurality of biometric data of one registered user and the reference data is calculated; (1) or ( 2) is any of the biometric registration devices.

(4) One aspect of the present invention is a biometrics authentication device corresponding to the biometrics registration device according to any one of (1) to (3) above, wherein a set of reference data for the distance index method generated by the biometrics registration device and a reference data similarity calculation unit that calculates the similarity between each reference data included in the authentication user's biometric data sample, and the biometric registration device records the similarity calculation result of the sample a determination candidate selection unit that selects a determination candidate template from templates included in the index based on an index; and a determination unit that determines whether the determination candidate template matches the sample. , is a biometric authentication device.
(5) In one aspect of the present invention, the determination candidate selection unit compares the result of calculating the similarity of the sample with the similarity of each template in the index recorded by the biometric registration device. ) is a biometric authentication device.
(6) In one aspect of the present invention, the determination candidate selection unit performs a similarity permutation between the sample and each reference data based on the similarity calculation result of the sample, and each of the templates in the index. The biometric authentication device according to (4) above, which compares the permutation of similarity between reference data.
(7) In one aspect of the present invention, each of a plurality of biometric data of one authenticated user is matched against other biometric data, and the matching score obtained as a result of the matching is The biometric authentication device according to any one of (4) to (6) above, further comprising a sample acquiring unit that acquires a sample of the authenticated user from maximum biometric data.

(8) According to one aspect of the present invention, a computer of a biometric registration device performs principal component analysis on biometric data templates of a plurality of registered users, and from the results of the principal component analysis, sets of reference data for the distance index method. a reference data set generating step for generating a reference data set, calculating a similarity between the template and the reference data for each template, and calculating a similarity index between the template and the reference data based on the result of the similarity calculation a recording index recording step; and a computer program for executing a recording index recording step.
(9) According to one aspect of the present invention, a set of reference data for the distance index method generated by the biometric registration device is stored in a computer of a biometric authentication device corresponding to the biometric registration device according to any one of (1) to (3) above. and a reference data similarity calculation step of calculating the similarity between each reference data included in the biometric data sample of the authenticated user; a determination candidate selection step of selecting a determination candidate template from templates included in the index based on the similarity of the index; and determining whether or not the determination candidate template matches the sample. A computer program for executing a determination step.

(10) In one aspect of the present invention, the biometric registration device performs principal component analysis on biometric data templates of a plurality of registered users, and generates a set of reference data for the distance index method from the results of the principal component analysis. and a step of generating a reference data set, wherein the biometric registration device calculates a degree of similarity between each template and the reference data, and calculates the similarity between the template and the reference data based on the result of the calculation of the degree of similarity. an index recording unit that records a degree index; a similarity calculation step between reference data for calculating a similarity; a determination candidate selection step of selecting a determination candidate template from among the templates that are provided; and a determination step of determining whether the biometric authentication device matches the sample with respect to the determination candidate template. authentication method.

Advantageous Effects of Invention According to the present invention, in one-to-many biometric authentication processing, it is possible to obtain the effect that the processing can be speeded up by the distance index method and the authentication accuracy can be improved.

1 is a block diagram showing a configuration example of a palmprint authentication system according to an embodiment; FIG. 1 is a block diagram showing a configuration example of a palmprint registration device according to an embodiment; FIG. FIG. 4 is a flow diagram showing an example of procedures in a registration stage of a palm print authentication method according to an embodiment; 1 is a block diagram showing a configuration example of a palmprint authentication device according to an embodiment; FIG. FIG. 4 is a flow diagram showing an example of the procedure of the authentication stage of the palm print authentication method according to one embodiment;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that, in the embodiments shown below, palm print authentication will be described as an example of biometric authentication.
FIG. 1 is a block diagram showing a configuration example of a palm print authentication system according to one embodiment. In FIG. 1, a palmprint authentication system 1 includes a palmprint registration device 10 and a palmprint authentication device 30 . The palm print authentication system 1 and the user terminal 2 communicate via the communication network NW. The communication network NW is, for example, a mobile phone network, a wireless communication network such as a wireless LAN (Local Area Network), or the Internet.

The user terminal 2 may be, for example, a mobile communication terminal device such as a smartphone or a tablet computer (tablet PC), or a stationary communication terminal device (for example, a stationary personal computer, etc.). There may be.

The user terminal 2 holds palmprint data P. FIG. The palm print data P is image data of a captured image in which the user's palm is captured as a subject. Therefore, the palmprint data P is image data including an image of the user's palmprint. The palm print data P may be a captured image captured by a camera included in the user terminal 2 or a captured image captured by a camera separate from the user terminal 2 . The user terminal 2 transmits palmprint data P (P_a) for registration to the palmprint authentication system 1 in the registration stage. The user terminal 2 transmits palmprint data P (P_b) for authentication to the palmprint authentication system 1 in the authentication stage. The palmprint data P corresponds to biometric data.

In the registration stage, the palmprint registration device 10 of the palmprint authentication system 1 obtains an index for the distance index method from templates of all registered users, which are templates respectively registered from palmprint data P (P_a) for registration of all registered users. create and record. In the authentication stage, the palmprint authentication device 30 of the palmprint authentication system 1 uses the index recorded by the palmprint registration device 10 to complete all the samples based on the palmprint data P (P_b) for authentication of one authenticated user. Narrows down the templates to be matched (search targets) for samples from the templates of registered users.

In this embodiment, the result of principal component analysis (principal component analysis image) for templates of all registered users is used as the pivot (reference data) for the distance index method. Specifically, n principal component analysis images from the first principal component to the n-th principal component are used as pivots to obtain n pivots. A principal component analysis image is an image that maximizes the variance of features in each dimension. For this reason, the similarity distribution between the principal component analysis image and the palmprint images of different people tends to spread over a wide range, so the accuracy of palmprint image identification is improved compared to when templates are used as pivots. can be made In addition, since palmprint images can be identified with high accuracy using only low-order principal component analysis images, the number of pivots can be reduced compared to when templates are used for pivots. and the pivot can be reduced. Furthermore, since a low-order principal component analysis image is a low-resolution image, it reduces the amount of computation for calculating the similarity between the pivot and the sample compared to using a high-resolution template for the pivot. be able to. As described above, in the present embodiment, by using the result of principal component analysis (principal component analysis image) for templates of all registered users as the pivot (reference data) for the distance index method, distance The index method is used to speed up processing and improve authentication accuracy.

The present embodiment will be described in detail below.

[Palmprint registration device]
FIG. 2 is a block diagram showing a configuration example of the palm print registration device according to this embodiment. 2, the palm print registration device 10 includes a template acquisition unit 11, a template registration database (template registration DB) 12, a pivot set generation unit (reference data set generation unit) 13, and an index recording unit .

Each function of the palmprint registration device 10 is realized by the palmprint registration device 10 having computer hardware such as a CPU (Central Processing Unit) and memory, and the CPU executing a computer program stored in the memory. be done. The palmprint registration device 10 may be configured using a general-purpose computer device, or may be configured as a dedicated hardware device. For example, the palm print registration device 10 may be configured using a server computer connected to a communication network such as the Internet. Further, each function of the palm print registration device 10 may be realized by cloud computing. Further, the palmprint registration device 10 may be implemented by a single computer, or may be implemented by distributing the functions of the palmprint registration device 10 to a plurality of computers. Alternatively, the palmprint registration device 10 may be configured to open a website using, for example, the WWW system.

The template acquisition unit 11 acquires a registered user's template from the registered user's registration palm print data P_a transmitted from the user terminal 2 . The template acquisition unit 11 registers the acquired template of the registered user in the template registration DB 12 . The template may be an image of the entire palm-print image area of the palm-print data P_a, or may be an image of a part of the area.

The template registration DB 12 stores templates for each registered user.

The pivot set generation unit 13 performs principal component analysis on the templates of all registered users registered in the template registration DB 12, and from the result of the principal component analysis (principal component analysis image), the set of pivots ( pivot set). Specifically, the pivot set generation unit 13 is composed of n pivots (principal component analysis images) in which n principal component analysis images from the first principal component to the n-th principal component are pivots respectively. generates a pivot set that The palm print registration device 10 records the pivot set.

The index recording unit 14 calculates the degree of similarity between the template and each pivot in the pivot set for each template of all registered users registered in the template registration DB 12 . The index recording unit 14 records an index indicating the degree of similarity between each template and each pivot based on the result of calculating the degree of similarity.

In this embodiment, NCC (Normalized Cross Correlation), which is an example of a similarity calculation method between images, is used as a method for calculating a similarity between a template and a pivot. Therefore, the index recording unit 14 calculates the NCC value between the template and each pivot in the pivot set for each template of all registered users registered in the template registration DB 12 . The index recording unit 14 records an index indicating the NCC value between each pivot for each template based on the result of the calculation of the NCC value.

Next, the procedure of the registration stage of the palmprint authentication method according to the present embodiment will be described with reference to FIG. FIG. 3 is a flow chart showing an example of procedures in the registration stage of the palmprint authentication method according to the present embodiment.

(Step S101) The palm print registration device 10 performs principal component analysis on the templates of all registered users registered in the template registration DB 12, and uses the result of the principal component analysis (principal component analysis image) to perform pivoting using the distance index method. (pivot set). The pivot set is composed of n pivots (principal component analysis images) "p1, p2, . . . , pn".

(Step S102) The palm print registration device 10 performs template-by-template and pivot set in template registration DB 12 for all registered user templates "t1, t2, t3, . . . , tx". The degree of similarity between each pivot “p1, p2, . . . , pn” is calculated (collation processing in FIG. 3). The palm print registration device 10 records an index indicating the degree of similarity between each template and each pivot based on the result of the degree of similarity calculation (index generation processing in FIG. 3). As a result, as shown in FIG. 3, the index recording unit 14 stores each pivot "p1, p2, , pn” is recorded.

Next, some modifications of the palm print registration device 10 according to this embodiment will be described below.

(Modification 1-1)
The pivot set generation unit 13 may perform principal component analysis on a reduced template obtained by reducing the image size of the template to a predetermined size. For example, principal component analysis is performed on a reduced template obtained by reducing a template of 160 pixels×160 pixels to 8 pixels×8 pixels.

(Modification 1-2)
The template acquisition unit 11 collects a plurality of palmprint data of one registered user (for example, the A plurality of palmprint images included in video data in which palmprints are photographed, etc.) are calculated for each pivot in the pivot set, and the plurality of palmprint data are obtained from the result of calculating the similarity. A permutation of similarity between the palmprint data and each pivot is created for each pivot, and the permutation corresponding to the permutation that minimizes the sum of the distances between the permutation of one palmprint data and the permutation of other palmprint data A template of the registered user may be obtained from the palm print data.

[Palmprint authentication device]
FIG. 4 is a block diagram showing a configuration example of a palm print authentication device according to this embodiment. In FIG. 4 , the palmprint authentication device 30 includes a sample acquisition unit 31 , an inter-pivot similarity calculation unit (inter-reference data similarity calculation unit) 32 , a determination candidate selection unit 33 , and a determination unit 34 .

Each function of the palmprint authentication device 30 is realized by the palmprint authentication device 30 having computer hardware such as a CPU and memory, and the CPU executing a computer program stored in the memory. The palmprint authentication device 30 may be configured using a general-purpose computer device, or may be configured as a dedicated hardware device. For example, the palmprint authentication device 30 may be configured using a server computer connected to a communication network such as the Internet. Further, each function of the palmprint authentication device 30 may be realized by cloud computing. Further, the palmprint authentication device 30 may be implemented by a single computer, or may be implemented by distributing the functions of the palmprint authentication device 30 to a plurality of computers. Alternatively, the palmprint authentication device 30 may be configured to open a website using, for example, the WWW system.

The sample acquiring unit 31 acquires a sample of the authenticated user from palmprint data P_b for authentication of the authenticated user transmitted from the user terminal 2 . The sample may be an image of the entire area of the palmprint image area of the palmprint data P_b, or may be an image of a part of the area. For example, the sample is obtained from the same region as the region from which the template is obtained, among the palmprint image regions of the palmprint data P_b.

The inter-pivot similarity calculation unit 32 calculates the similarity between each pivot in the pivot set generated by the palm print registration device 10 and a sample of the authenticated user. The method of calculating the similarity between the sample and the pivot is the same as the method of calculating the similarity between the template and the pivot, and NCC is used in this embodiment. Therefore, the inter-pivot similarity calculation unit 32 calculates the NCC value between each pivot in the pivot set generated by the palm print registration device 10 and the sample of the authenticated user.

The determination candidate selection unit 33 selects a determination candidate template from templates included in the index based on the result of calculating the similarity of the sample of the authenticated user and the index recorded by the palm print registration device 10 .

The determination unit 34 performs matching processing (one-to-one palm print authentication processing) between the template of the determination candidate and the sample of the authenticated user, and based on the result of the matching processing, the template of the determination candidate matches the sample of the authenticated user. It is determined whether (authentication success) or not (authentication failure). The determination result (authentication success or authentication failure) is output from the palmprint authentication device 30 .

Next, with reference to FIG. 5, the procedures of the authentication stage of the palmprint authentication method according to the present embodiment will be described. FIG. 5 is a flow chart showing an example of the authentication stage procedure of the palmprint authentication method according to the present embodiment.

(Step S<b>201 ) The palmprint authentication device 30 acquires an authenticated user's sample (authentication sample) a 1 from the authenticated user's authentication palmprint data P_b transmitted from the user terminal 2 .

(Step S202) The palmprint authentication device 30 calculates the degree of similarity between each pivot "p1, p2, ..., pn" in the pivot set generated by the palmprint registration device 10 and the authentication sample a1 ( authentication pivot matching score set calculation processing in FIG. 5). As a result, a set of similarities (authentication pivot matching score set) between the authentication sample a1 and each pivot "p1, p2, . . . , pn" is generated.

(Step S203) Based on the authentication pivot matching score set and the index recorded by the palmprint registration device 10, the palmprint authentication device 30 identifies the template "t1, t2, t3, ..., tx" included in the index. Select a judgment candidate template from among them. Specifically, the palmprint authentication device 30 calculates the similarity in the authentication pivot matching score set and the similarity of each template “t1, t2, t3, . . . , tx” in the index recorded by the palmprint registration device 10 Compare with degrees. Here, for each template “t1, t2, t3, . degree set” (similarity calculation processing in FIG. 5). Next, the palmprint authentication device 30 selects a predetermined number (m) of templates having high similarity in the calculation result from among the templates “t1, t2, t3, . . . , tx” included in the index. , is selected as a judgment candidate template (candidate selection processing in FIG. 5).

(Step S204) The palmprint authentication device 30 performs matching processing (one-to-one palmprint authentication processing) between the judgment candidate template and the authentication sample a1 for each judgment candidate template for m judgment candidate templates. (matching process in FIG. 5). Next, based on the result of the matching process, the palmprint authentication device 30 determines whether the template of the determination candidate matches the authentication sample a1 (authentication success) or not (authentication failure) (the determination process in FIG. 5). ). For example, the palmprint authentication device 30 determines that the authentication is successful when the matching score as a result of the matching process exceeds a predetermined threshold value, and determines that the authentication is unsuccessful otherwise.

Next, some modifications of the palm print authentication device 30 according to this embodiment are shown below.

(Modification 2-1)
The determination candidate selection unit 33 selects a permutation of the degree of similarity between the authentication sample a1 and each pivot "p1, p2, ..., pn" in the authentication pivot matching score set, and each template "t1, t2, t3, . Specifically, the determination candidate selection unit 33 selects the “similarity permutation between the authentication sample a1 and the pivots “p1, p2, . permutation of similarity between "p2, ..., pn" and the template "t1, t2, t3, ..., tx", and the template " t1, t2, t3, . According to this modified example 2-1, since the permutation order is determined only by the information as to which is larger or smaller, it is possible to obtain the effect of improving the robustness against fluctuations of the authenticated user and improving the authentication accuracy.

(Modification 2-2)
The sample acquisition unit 31 obtains one palmprint data for each of a plurality of palmprint data of one authenticated user (for example, a plurality of palmprint images included in moving image data in which the palmprint of one authenticated user is photographed). A sample of the authenticated user may be acquired from the palmprint data having the highest matching score as a result of matching by matching with other palmprint data.

According to the above-described embodiment, in the one-to-many palmprint authentication process, the distance The indexing method can speed up the processing and improve the accuracy of authentication.

Note that the palmprint registration device 10 and the palmprint authentication device 30 may be configured as separate devices, or may be configured as the same device.

The palmprint authentication system 1 according to the present embodiment can be applied, for example, to personal authentication processing in online payment services.

As a result, it will be possible to improve overall service quality in personal authentication services that use, for example, a palm print authentication system. , promote sustainable industrialization, and contribute to the expansion of innovation.

Although the embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and design changes and the like are included within the scope of the present invention.

In the above-described embodiment, palm print authentication is used as an example of biometric authentication, but the present invention can also be applied to biometric authentication such as fingerprint authentication, face authentication, and iris authentication.
Alternatively, a computer program for realizing the functions of the devices described above may be recorded in a computer-readable recording medium, and the program recorded in the recording medium may be read and executed by the computer system. Note that the “computer system” referred to here may include hardware such as an OS and peripheral devices.
In addition, "computer-readable recording medium" includes writable nonvolatile memories such as flexible discs, magneto-optical discs, ROMs and flash memories, portable media such as DVDs (Digital Versatile Discs), and computer system built-in media. A storage device such as a hard disk that

Furthermore, "computer-readable recording medium" means a volatile memory (e.g., DRAM (Dynamic Random Access Memory)), which holds the program for a certain period of time, is also included.
Further, the above program may be transmitted from a computer system storing this program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in a transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line.
Further, the program may be for realizing part of the functions described above. Further, it may be a so-called difference file (difference program) that can realize the above-described functions in combination with a program already recorded in the computer system.

1... palmprint authentication system, 10... palmprint registration device, 11... template acquisition section, 12... template registration database (template registration DB, 13... pivot set generation section (reference data set generation section), 14... index recording section, 30... Palmprint authentication device 31 Sample acquisition unit 32 Inter-pivot similarity calculation unit (reference data similarity calculation unit) 33 Judgment candidate selection unit 34 Judgment unit 2 User terminal NW Communication network

Claims (10)

a reference data set generation unit that performs principal component analysis on templates of biometric data of a plurality of registered users and generates a set of reference data for the distance index method from the results of the principal component analysis;
an index recording unit that calculates the degree of similarity between each template and the reference data, and records an index of the degree of similarity between the template and the reference data based on the result of the calculation of the degree of similarity;
A biometric registration device comprising: The reference data set generation unit performs principal component analysis on a reduced template obtained by reducing the image size of the template to a predetermined size.
The biometric registration device according to claim 1. calculating the degree of similarity between each of a plurality of biometric data of one registered user and the reference data; further comprising a template acquisition unit that creates a permutation and acquires the template of the registered user from biometric data that minimizes the sum of distances between the permutations;
The biometric registration device according to claim 1 or 2. A biometric authentication device corresponding to the biometric registration device according to any one of claims 1 to 3,
a similarity calculation unit between reference data that calculates a similarity between each reference data included in a set of reference data for the distance index method generated by the biometric registration device and a sample of biometric data of an authenticated user;
a determination candidate selection unit that selects a determination candidate template from templates included in the index based on the result of calculating the similarity of the sample and the index recorded by the biometric registration device;
a determination unit that determines whether or not the template of the determination candidate matches the sample;
A biometric authentication device. The determination candidate selection unit compares the result of calculating the similarity of the sample with the similarity of each template in the index recorded by the biometric registration device.
The biometric authentication device according to claim 4. The determination candidate selection unit selects a similarity permutation between the sample and each reference data based on the result of calculating the similarity of the sample, and a similarity permutation between each reference data of each template in the index. compare,
The biometric authentication device according to claim 4. For each of a plurality of biometric data of one authenticated user, one biometric data is compared with other biometric data, and the biometric data with the maximum matching score as a result of the matching is selected from the biometric data of the authenticated user. further comprising a sample acquisition unit that acquires a sample;
The biometric authentication device according to any one of claims 4 to 6. In the computer of the biometric registration device,
a reference data set generation step of performing principal component analysis on templates of biometric data of a plurality of registered users and generating a set of reference data for the distance index method from the results of the principal component analysis;
an index recording step of calculating the degree of similarity between each template and the reference data, and recording an index of the degree of similarity between the template and the reference data from the result of the calculation of the degree of similarity;
A computer program for executing In a computer of a biometric authentication device corresponding to the biometric registration device according to any one of claims 1 to 3,
a reference data similarity calculation step of calculating a similarity between each reference data included in a set of reference data for the distance indexing method generated by the biometric registration device and a sample of biometric data of an authenticated user;
a determination candidate selection step of selecting a determination candidate template from templates included in the index based on the result of calculating the similarity of the sample and the similarity of the index recorded by the biometric registration device;
a determination step of determining whether the determination candidate template matches the sample;
A computer program for executing a reference data set generation step in which the biometric registration device performs principal component analysis on templates of biometric data of a plurality of registered users, and generates a set of reference data for the distance index method from the results of the principal component analysis;
The biometric registration device calculates the degree of similarity between each template and the reference data, and records an index of the degree of similarity between the template and the reference data based on the result of calculating the degree of similarity. Department and
A biometrics authentication device calculates similarity between reference data for calculating a similarity between each reference data included in a set of reference data for the distance index method generated by the biometrics registration device and a sample of biometric data of an authenticated user. a step;
The biometric authentication device selects a determination candidate template from templates included in the index based on the result of calculating the similarity of the sample and the similarity of the index recorded by the biometric registration device. a selection step;
a determination step in which the biometric authentication device determines whether or not the template of the determination candidate matches the sample;
Biometric authentication methods including.

Images