JP4283126B2 - Biometric authentication system and biometric authentication method - Google Patents

Description

  The present invention relates to a biometric authentication system and a biometric authentication method.

  In recent years, a system for authenticating a person based on physical characteristics such as a fingerprint, an iris, a face, and a voiceprint has become widespread. For example, application to identity authentication in electronic commerce via the Internet is being studied. In this biometric authentication system, since biometric data to be used is important personal information, the prevention of leakage of biometric data is positioned as an important issue in order to protect personal information. For this reason, it has been proposed to avoid transmitting or storing biometric data or an authentication key itself as it is, using conversion using a one-way hash function or conversion using encryption.

Here, when determining whether or not authentication is possible based on the match / mismatch of the conversion values by the one-way hash function, for example, as in password authentication, valid converted data (valid registered password data and valid authenticated password) Data) is a perfect match. However, in the biometric data, the measurement value fluctuates at every measurement, and the biometric data at the time of registration of the same person and the biometric data at the time of authentication are not always completely coincident. For this reason, in the conventional biometric authentication system, the fluctuation of the password generated from the biometric data is corrected by error correction using a linear code such as a Reed-Solomon code (for example, see Patent Document 1).
JP 2003-23422 A

  However, in the error correction using the linear code described above, it is not easy to completely correct the fluctuation of the password generated from the biometric data, and the realization of the biometric authentication system using the conversion using the one-way hash function described above. There is a problem that is difficult. For example, in order to improve error correction capability, it is necessary to increase the code length, but the processing is rapidly complicated, resulting in a decrease in processing speed, and can be applied to authentication services that require immediacy. Disappear. In addition, if the error correction results in an incomplete result and the fluctuation of the password generated from the biometric data cannot be completely corrected, correct authentication cannot be performed.

  The present invention has been made in consideration of such circumstances, and its purpose is to perform authentication processing while keeping biometric data, which is personal information of the user, concealed, thereby preventing leakage of biometric data. It is an object of the present invention to provide a biometric authentication system and a biometric authentication method that can be easily realized.

  In order to solve the above-described problems, a biometric authentication system according to the present invention includes a biometric data measuring device and a biometric authentication system that authenticates based on biometric data measured by the measuring device. The measurement apparatus includes a conversion unit that converts the measured biometric data using a conversion function including non-linear conversion, and the authentication device includes a storage unit that stores registration data obtained by converting the measured biometric data by the conversion unit; , The similarity calculation means for calculating the similarity between the authentication target data obtained by converting the measured biometric data by the conversion means and the registration data, the similarity is determined, and authentication is performed according to the determination result It is characterized by having an authenticating means.

  In the biometric authentication system of the present invention, the conversion function expands the amount of information while performing nonlinear conversion.

  In the biometric authentication system of the present invention, the conversion function stores the normalized Hamming distance between the converted data.

  In the biometric authentication system of the present invention, the measurement device and the authentication device each include communication means for transmitting and receiving data via a communication line, and the registration data and the subject to be authenticated by an encryption method using a disposable key. It is characterized by encrypting data, concealing the data on the communication line, and preventing impersonation by others.

  In the biometric authentication system of the present invention, the storage means stores the encrypted registration data.

  The biometric authentication method of the present invention is a biometric authentication method for authenticating based on biometric data, wherein biometric data is converted by a conversion function including non-linear conversion and stored as registered data, and biometric data is converted by the conversion function. Converting the data to be authenticated, calculating the similarity between the registered data and the data to be authenticated, and determining the similarity and authenticating according to the determination result. It is a feature.

  In the biometric authentication method of the present invention, the conversion function expands the amount of information while performing nonlinear conversion.

  In the biometric authentication method of the present invention, the conversion function stores the normalized Hamming distance between the converted data.

  In the biometric authentication method of the present invention, the registration data is encrypted by an encryption method using a disposable key and then sent from the user to the authenticator via a communication line; And a process of sending the data from the user to the authenticator via a communication line after being encrypted by an encryption method using a disposable key.

  In the biometric authentication method of the present invention, the encrypted registration data is stored.

  According to the present invention, since the converted data after biometric data conversion by the conversion function is used for registration and authentication, the biometric data itself that is important personal information of the user is not used as it is for the authentication process. Thereby, leakage of biometric data can be prevented. Furthermore, since authentication is performed according to the similarity of the converted data, it is possible to allow fluctuations in biometric data that occur at each measurement, and it is possible to realize with a simple configuration.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the present embodiment, as a specific example of the biometric authentication system, a system that authenticates the person by the feature of the iris will be described.

  FIG. 1 is a block diagram showing a configuration of a biological data measuring apparatus 1 according to an embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of the authentication device 2 according to an embodiment of the present invention. The biometric authentication system according to the present embodiment includes the measurement device 1 and the authentication device 2.

In the measurement apparatus 1 illustrated in FIG. 1, the imaging unit 11 includes a camera and acquires an iris image from an image of the user's eyes. For example, the iris image is acquired by capturing the eyes of the user with a camera. Next, for the image of the eye, the scleral side boundary, the pupil side boundary, and the upper and lower eyelid side boundaries are determined by using the change in image brightness, the iris region is specified, and only the iris image is cut out. When acquiring an iris image, living body detection is performed to confirm that the image is acquired from a living human eye.
The feature extraction unit 12 extracts features from iris image data (biological measurement data) to obtain an iris code (biological data). This feature extraction is performed by paying attention to the physical feature amount unique to the iris. For example, eight annular analysis zones are allocated to the iris region, and the iris code is acquired by scanning the annular analysis zone.
In addition, as an acquisition method of an above-mentioned iris code, the thing of patent 3307936 gazette can be utilized, for example.

The conversion unit 13 converts the iris code using a predetermined conversion function. Details of this conversion function will be described later.
The encryption unit 14 encrypts the converted data after the iris code conversion by the conversion unit 13 using a disposable key.
The communication unit 15 is connected to a communication network such as the Internet and transmits / receives data.

  In the authentication device 2 shown in FIG. 2, the database 21 stores and accumulates user registration data. The stored registration data is encrypted data that is further encrypted by the encryption unit 14 after the registration iris code measured by the measuring device 1 is converted by the conversion unit 13. This registration data is provided with an expiration date. The expired registration data is deleted. This expiration date is set in consideration of safety and the like.

The similarity calculation unit 22 includes the data to be authenticated and the decrypted data of the registration data in the database 21 (conversion data after the registration iris code measured by the measurement device 1 is converted by the conversion unit 13). Calculate similarity. The authentication target data is converted data after the authentication iris code measured by the measuring device 1 is converted by the conversion unit 13.
Further, the similarity obtained by the similarity computing unit 22 is suitable for evaluating the physical feature amount extracted by the feature extracting unit 12 of the measuring apparatus 1. This similarity calculation is performed by a predetermined method corresponding to the feature extraction method of the feature extraction unit 12. For example, the normalized Hamming distance can be used as the similarity for iris authentication. Also, Euclidean distance can be used. The feature extraction unit 12 and the similarity calculation unit 22 can be realized by using commercially available products that can be combined with each other. In addition, the imaging unit 11 may be combined.
The authentication unit 23 determines the similarity that is the calculation result of the similarity calculation unit 22 and authenticates according to the determination result.

The key generation unit 24 generates a disposable key. This key is provided to the measuring apparatus 1 and used for encryption of the converted data after the iris code conversion.
The decryption unit 25 decrypts the encrypted data received from the measurement device 1 with the corresponding key. The decryption unit 25 corresponds to the encryption unit 14 of the measurement apparatus 1.
The communication unit 26 is connected to a communication network such as the Internet and transmits / receives data.

Next, the operation of the biometric authentication system according to the present embodiment will be described with reference to FIGS. 3 and 4 are diagrams for explaining the flow of data in the biometric authentication system according to the present embodiment. FIG. 3 is for user registration, and FIG. 4 is for user authentication. 5 and 6 are flowcharts showing the flow of processing in the biometric authentication system according to the present embodiment. FIG. 5 is for user registration, and FIG. 6 is for user authentication.
In the following description, it is assumed that the iris code is coded as n bits (X ₁ ... X _n ). However, X _i (i = 1,..., N) is 0 or 1. Further, it is assumed that an n-bit iris code is converted into a 2n-bit converted iris code (X ₁ ... X _2n ) by the conversion of the conversion unit 13. The iris code length n is mainly 2048 bits or 4096 bits.

First, the operation at the time of user registration will be described with reference to FIGS.
First, the user makes a registration request to the authentication device 2. In response to this registration request, the key generation unit 24 of the authentication device 2 generates a 2n-bit challenge code c (disposable key). Then, the challenge code c is transmitted to the measuring device 1, and the measuring device 1 receives the challenge code c (step S101). This challenge code c is held in the decoding unit 25 until decoding is completed for decoding.

Next, the user images the eyes using the camera 11a shown in FIG. Thereby, the imaging part 11 of the measuring apparatus 1 acquires an iris image from imaging data (step S102). Next, the feature extraction unit 12 acquires an n-bit iris code A = (A ₁ ... A _n ) from the iris image data (step S103). Next, the conversion unit 13 generates a random number R as a conversion parameter (step S104). This random number R is stored for authentication. Details of the random number R will be described later.

Next, in step S105, the conversion unit 13 converts the iris code A by the conversion function f using the random number R that is the conversion parameter. The converted iris code after this conversion is expressed as f (A, R) = A ′ = (A ₁ ′... A _n ′). Further, the encryption unit 14 encrypts the converted iris code A ′ using the challenge code c received from the authentication device 2 (step S106). This encrypted data is represented as E _k (A ′, c). However, E _k is an encryption function, and its inverse function E _k ⁻¹ exists. For example, E _k is a public key encryption function using the public key _k of the authentication device 2. Thereby, transmission data E _k (A ′, c) is obtained.

Then, the measurement device 1 is transmission data _E k (A ', c) to create a registration ID of the ID and transmission data _E k (A', c) the transmission to the authentication device 2. The ID is stored for authentication. Further, the iris code A, the converted iris code A ′, the transmission data E _k (A ′, c), and the challenge code c are deleted.
Next, the authentication device 2 receives the ID and transmission data E _k (A ′, c) (step S107). Next, the authentication device 2 registers the received ID and transmission data E _k (A ′, c) in the database 21 in association with each other (step S108). The database 21 stores the transmission data E _k (A ′, c) in an encrypted state. As a result, the user's iris information is more securely concealed.

Next, the operation at the time of user authentication will be described with reference to FIGS.
First, as in the case of user registration described above, generation and delivery of a challenge code c ′ (step S201), acquisition of an iris image (step S202), acquisition of an iris code B = (B ₁ ... B _n ). (Step S203), the conversion iris code B ′ = (B ₁ ′... B _n ′) is calculated and the transmission data E _k (B ′, c ′) is obtained by encryption (Step S204). The conversion parameter R used in step S204 is the one saved at the time of user registration.

Next, the measurement apparatus 1 transmits the transmission data E _k (B ′, c ′) and the ID stored at the time of user registration to the authentication apparatus 2 and is received by the authentication apparatus 2 (step S205). Note that the measuring apparatus 1 erases the iris code B, the converted iris code B ′, the transmission data E _k (B ′, c ′), and the challenge code c ′ in the same manner as at the time of registration.

Next, the decryption unit 25 of the authentication device 2 decrypts the received E _k (B ′, c ′) (step S206). The decrypted data B ′, that is, the converted iris code B ′ = f (B, R) based on the conversion function f is the user's authentication data.

Next, the decryption unit 25 of the authentication device 2 obtains the corresponding registration data E _k (A ′, c) from the database 21 based on the ID received from the measurement device 1 together with the transmission data E _k (B ′, c ′). Obtained and decoded (step S207). The decoded data A ′, that is, the converted iris code A ′ = f (A, R) by the conversion function f becomes the reference data of the user.

Next, the similarity calculation unit 22 calculates the similarity between the reference data A ′ and the authenticated data B ′ (step S208). In this embodiment, the normalized Hamming distance HD _pro between the reference data A ′ and the authenticated data B ′ is calculated by the following equation.

When the calculation of the normalized Hamming distance HD _pro is completed, the authentication device 2 uses the decrypted converted iris code A ′, B ′ and the encrypted converted iris code E _k (B ′, c ′) received for authentication. to erase.

Next, the authentication unit 23 compares the normalized Hamming distance HD _pro that is the calculation result of the similarity calculation unit 22 with a predetermined threshold (step S209). As a result of the comparison, if the calculation result is equal to or less than the threshold value, it is determined that the user is the user (step S210). As a result, authentication is successful. On the other hand, if the calculation result exceeds the threshold, it is determined that the user is not himself (step S211). This results in authentication failure.

  Note that the threshold used in step S209 takes into account the statistically obtained identity rejection rate (the rate that the person is mistakenly regarded as another person) and the acceptance rate (the rate at which the other person is mistakenly regarded as the person). Set by the administrator.

As described above, according to the present embodiment, the conversion data after the iris code conversion by the conversion function is transmitted, registered, and used for authentication in both cases of user registration and user authentication. The iris code itself, which is important personal information of the user, is not transmitted as it is or used for authentication processing. Thereby, leakage of the iris code (biological data) can be prevented.
Furthermore, since the authentication is performed according to the similarity of the converted data, the fluctuation of the iris code that occurs at each measurement can be allowed, and a simple configuration can be realized.

  Further, since the registration data and the data to be authenticated are encrypted with the challenge code, it is possible to conceal the data on the communication line and to prevent impersonation by others.

Next, the conversion function f for converting the above-described iris code will be described in detail.
This conversion function stores similar characteristics between iris codes (converted data) before conversion. Hereinafter, an example of a conversion function for storing the normalized Hamming distance between the registration iris code and the authentication iris code will be described.

FIG. 7 is a diagram for explaining a first example of a conversion procedure using a conversion function according to the present invention. Example 1 will be described with reference to FIG.
In FIG. 7, n-bit iris code X = (X ₁ ... X _n ) is converted into 2n-bit converted iris code X ′ = (X ₁ ′... X _2n ′) by the following procedure. Perform the conversion. However, X _i (i = 1,..., N) and X _j ′ (j = 1,..., 2n) are 0 or 1.
First, using a random number generation algorithm that guarantees safety,
Generate a random number R of {n + log ₂ 2n + (2 log ₂ 2n + 2) · 3n / 2} bits. This random number R is
R = (R _e R _p R ₁ ... R _{3n / 2} ).
However, the element R _{e of the} random number R is n bits, the element R _p is log ₂ 2n bits,
The element R _j is (2log ₂ 2n + 2) bits, and j = 1,..., 3n / 2.
Also, element R _j is
R _j = (u _j v _j θ _j )
However, u _j and v _j are log ₂ 2n bits, and θ _j is 2 bits.
The generated random number R is concealed by being safely stored on the measuring device 1 side. The random number R is changed for each authentication device 2 that registers the iris code.

Next, in step S301, the element R _e (n bits) of the random number R is connected to the iris code X (n bits). By this enlargement process, a 2n-bit enlarged iris code Y is obtained.
Y = (Y ₁ ... Y _2n )
= (X || R _e ) = (X ₁ ... X _n R _e1 ... R _en )
However, the symbol “||” represents concatenation.

Then, in step S302, in accordance with the sort rule to be selected by the element R _p of the random number R, sort the elements of 2n bits enlarged iris code Y. There are 2n rearrangement rules, which are safely stored as a conversion table. By this rearrangement process, a 2n-bit rearrangement iris code Z is obtained.
Z = (Z ₁ ... Z _2n )

Next, in step S303, the rearrangement iris code Z is rotated. In this rotation process, first, a 2n-dimensional vector W = (W ₁ ,..., W _2n ) is generated from the rearranged iris code Z by the following equation.

Next, the processes of steps S303-1 and S303-2 shown below are repeated. The processes in steps S303-1 and S303-2 are started by setting the initial value of j to 1, and thereafter “j is (3n / 2) or less” while adding 1 to j. Repeat while being satisfied.
In step S303-1, the element R _j = (u _j v _j θ _j ) of the random number R is used, and the ([u _j ] ₁₀ +1) th element from the 2n-dimensional vector W and ([v _j ] ₁₀ +1) The two-dimensional vector Wa is constructed by extracting the th element. However, [x] ₁₀ is the value of the decimal representation of x.

  This two-dimensional vector Wa has coordinates of any of 45 degrees, 135 degrees, 225 degrees, and 315 degrees on the two-dimensional circumference of radius 1 when the horizontal axis positive direction is 0 degrees at the center of the origin (0, 0). Will be located.

In step S303-2, the two-dimensional vector Wa ′ is obtained by rotating the two-dimensional vector Wa counterclockwise by an angle θ _j ′ around the origin. However, the angle θ _j ′ = 90 · [θ _j ] ₁₀ (the unit is degrees). Here, since θ _j is 2 bits, the angle θ _j ′ is one of 0 degree, 90 degrees, 180 degrees, and 270 degrees.

  That is, the two-dimensional vector Wa is converted into a two-dimensional vector Wa ′ by the following equation.

Then, the two elements of the two-dimensional vector Wa ′ are converted into corresponding elements of the transformed 2n-dimensional vector W ′ = (W ₁ ′,..., W _2n ′), that is, the ([u _j ] ₁₀ +1) -th element. And ([v _j ] ₁₀ +1) th element. Here, if the corresponding element already exists, it is replaced.

  In step 303-1, there is a possibility that elements of the 2n-dimensional vector W are selected in an overlapping manner. In this case, instead of the 2n-dimensional vector W, the converted 2n-dimensional vector W ' Extract and use elements.

Next, in step S304, a converted iris code X ′ = (X ₁ ′... X _2n ′) is generated from the converted 2n-dimensional vector W ′ obtained by the above rotation process using the following equation.

FIG. 8 is a diagram for explaining a second example of the conversion procedure using the conversion function according to the present invention. Embodiment 2 will be described with reference to FIG.
Also in this second embodiment, with respect to the iris of the n-bit code _{X =} (X 1 · · · _{X n),} by the following procedure, converting iris code X 2n-bit _{'= (X 1' ··· X} 2n Convert to ').

First, using a random number generation algorithm that guarantees safety,
(3log ₂ n + 6) An n-bit random number R is generated. This random number R is
R = (R ₁ ... R _j ... R _n )
Where element R _j is
R _j = (u _j v _j z _j θ _j φ _j e _j d _j )
However, u _j , v _j and e _j are log ₂ n bits, θ _j and φ _j are 2 bits, and z _j and d _j are 1 bit.
The generated random number R is concealed by being safely stored on the measuring device 1 side. The random number R is changed for each authentication device 2 that registers the iris code.

Next, the following steps S401 to S405 are repeated. The processing of steps S401 to S405 is started by setting the initial value of j to 1, and thereafter, while 1 is added to j, it is repeated while satisfying the condition that “j is n or less”.
In step S401, from the iris code X, (n + j−1) -dimensional vector W _{n + j−1} = (W ₁ ,..., W _{n + j−1} ) and temporary (n + j) -dimensional vector W _{n + j} ′ = (W ₁ ′,..., W _{n + j−1} ′, 0) are generated.

Then, in step S402, using elements _R j ₌ random number _{R (u j v j z j} θ j φ j e j d j), (n + j-1) from dimensional vector _{W n + j-1 ([} u j] _{The 10} + 1) th element and the ([ _vj ] ₁₀ + 1) th element are extracted. Also, a 1-bit random value z _j is converted by the following equation.

Then, a three-dimensional vector Wa is constructed from the two elements extracted from the (n + j−1) -dimensional vector W _{n + j−1} and the converted value W _{n + j of the} random value z _j .

  The three-dimensional vector Wa is a vector in the xyz space, and is an element of x coordinate, y coordinate, and z coordinate in order.

Next, in step S403, the angle θ _j ′ = 90 · [θ _j ] ₁₀ (in degrees) around the x-axis and the angle φ _j ′ = 90 · [φ _j ] around the y-axis about the three-dimensional vector Wa. Each is rotated by ₁₀ (in degrees) to obtain a transformed three-dimensional vector Wa ′.

  That is, the three-dimensional vector Wa is converted into a converted three-dimensional vector Wa ′ by the following equation.

However, rotation matrices G and H around the x-axis and y-axis are expressed by the following equations.

Next, in step S404, the three elements of the transformed three-dimensional vector Wa ′ are replaced with the corresponding elements of the temporary (n + j) -dimensional vector W _{n + j} ′, that is, the ([u _j ] ₁₀ +1) th element and ([v _j ] ₁₀ +1) element and (n + j) element. Here, if the corresponding element already exists, it is replaced.

Next, in step S405, the obtained temporary (n + j) dimensional vector W _{n + j} ′ is moved to the left if “d _j = 1”, and to the right if “d _j = 0” (([ e _j ] ₁₀ +1) bits are cyclically shifted.

Thereby, an (n + j) dimensional vector W _{n + j} is obtained. Next, the process returns to step S401, and 1 is added to j.

Then, when j is 2n-dimensional vector _{W 2n} is obtained by repeating the processing of steps S401~S405 to 1 to n, the following equation from the 2n-dimensional vector _{W 2n,} converting iris code X 2n-bit '= ( X ₁ ′... X _2n ′) is generated.

  In both the first and second embodiments described above, the information amount is expanded using a random number, and nonlinear conversion such as rotation and rearrangement based on the random number is performed, and the n-bit iris code X is converted into the 2n-bit converted iris code X ′. Convert to This makes it very difficult to obtain the user's iris information from the converted iris code X ′, which can contribute to the realization of protection of the user's personal information. Further, in the second embodiment, the amount of information is expanded while nonlinearly converting the iris code. Therefore, the probability that the iris code itself is nonlinearly converted is higher than that of the first embodiment, which is more preferable.

The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes design changes and the like within a scope not departing from the gist of the present invention.
For example, in the above-described embodiment, the iris code is sent from the measurement device 1 to the authentication device 2 via the communication line, but the method of delivering the iris code (biometric data) is not limited to this. For example, the user may be recorded on a recording medium such as a flexible disk and delivered to the authenticator.

  In addition, the present invention can be applied to various types of biometric authentication regardless of the type of living body, based on a predetermined physical feature amount unique to the living body. For example, it can be applied to authentication using physical features such as irises, fingerprints, palm prints, handprints, veins such as backs of hands, faces, voiceprints, retinas, body odors, DNA, handwriting, keystrokes, and walking.

  It is also possible to construct a system for authenticating a person by combining a plurality of types of biometric authentication.

It is a block diagram which shows the structure of the measuring apparatus 1 of the biometric data which concerns on one Embodiment of this invention. It is a block diagram which shows the structure of the authentication apparatus 2 which concerns on the same embodiment. It is a figure for demonstrating the flow of the data at the time of user registration in the biometric authentication system which concerns on one Embodiment of this invention. It is a figure for demonstrating the flow of the data at the time of user authentication in the biometric authentication system which concerns on the embodiment. It is a flowchart which shows the flow of the process at the time of user registration in the biometric authentication system which concerns on the embodiment. It is a flowchart which shows the flow of a process at the time of user authentication in the biometric authentication system which concerns on the embodiment. It is a figure for demonstrating the 1st example of the conversion procedure by the conversion function which concerns on this invention. It is a figure for demonstrating the 2nd example of the conversion procedure by the conversion function which concerns on this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Measurement apparatus of iris code (biological data), 2 ... Authentication apparatus, 11 ... Imaging part (biological measuring means), 12 ... Feature extraction part, 13 ... Conversion part, 14 ... Encryption part, 15, 26 ... Communication , 21 ... database (storage means), 22 ... similarity calculation part, 23 ... authentication part, 24 ... key generation part, 25 ... decryption part.

Claims (6)

In a biometric authentication system comprising a biometric data measurement device and an authentication device that authenticates based on biometric data measured by the measurement device,
The measurement apparatus includes a conversion unit that performs conversion having non-linearity configured by combining linear conversion and random numbers on the measured biological data,
The authentication device
Storage means for storing registration data obtained by converting the measured biological data by the conversion means;
Similarity calculation means for calculating the degree of similarity between the data to be authenticated in which the measured biometric data is converted by the conversion means and the registered data based on a normalized Hamming distance ;
Authentication means for determining the similarity and authenticating according to the determination result,
The converting means includes
Random numbers (R = (R _e R _p R ₁ ... R _j ... R _{3n / 2} ), j = 1,..., 3n / 2, {n + log ₂ 2n + (2 log ₂ 2n + 2) · 3n / 2 } Bits, R _e is n bits, R _p is log ₂ 2n bits, R _j is (2 log ₂ 2n + 2) bits, R _j = (u _j v _j θ _j ), u _j , v _j are log ₂ 2n bits, θ _j is 2 bits) and random number generating means for generating
Data connection means for connecting the element (R _e ) of the random number (R) to biometric data (X, n bits);
A conversion table that holds a plurality (2n) of data rearrangement rules;
Wherein selecting one of sorting rules from the conversion table based on the element (R _p) of the random number (R), the connection data (Y, 2n bits) bit rearrangement rearranging bits according sorting rules selected Means,
First vector generating means for generating a 2n-dimensional vector (W) composed of predetermined elements corresponding to each bit of the rearranged data (Z, 2n bits);
Among the elements of the 2n-dimensional vector (W), from the elements corresponding to the first part (u _j ) and the second part (v _j ) of each element (R _j ) of the random number (R) , the random number ( Second vector generation means for generating a two-dimensional vector (Wa) corresponding to each element (R _j ) of R );
Rotation processing means for performing a predetermined rotation process on the two-dimensional vector (Wa) according to the third part (θ _j ) of each element (R _j ) of the random number (R) ;
Third vector generation means for generating a 2n-dimensional vector (W ′) from the elements of the two-dimensional vector (Wa ′) after the rotation processing;
Conversion data generation means for generating biometric data (X ′, 2n bits) after conversion consisting of predetermined bits corresponding to each element of the 2n-dimensional vector (W ′),
A biometric authentication system characterized by that. In a biometric authentication system comprising a biometric data measurement device and an authentication device that authenticates based on biometric data measured by the measurement device,
The measurement apparatus includes a conversion unit that performs conversion having non-linearity configured by combining linear conversion and random numbers on the measured biological data,
The authentication device
Storage means for storing registration data obtained by converting the measured biological data by the conversion means;
Similarity calculation means for calculating the degree of similarity between the data to be authenticated in which the measured biometric data is converted by the conversion means and the registered data based on a normalized Hamming distance ;
Authentication means for determining the similarity and authenticating according to the determination result,
The converting means includes
Random number (R = (R ₁ ... R _j ... R _n ), j = 1,..., N, (3log ₂ n + 6) n bits, R _j = (u _j v _j z _j θ _{j j} φ _j e _j d _j ), u _j , v _j , e _j are log ₂ n bits, θ _j , φ _j are 2 bits, and z _j , d _j are 1 bit),
Using a predetermined element corresponding to each bit of the biometric data (X, n bits), the (n + j−1) dimension vector (W _{n + j−1} = W ₁ ,..., W _{n + j−1} ) and the first First vector generation means for generating temporary (n + j) dimensional vectors (W _{n + j} ′ = W ₁ ′,..., W _{n + j−1} ′, 0);
Of the elements of the (n + j-1) -dimensional vector (W _{n + j-1} ), the elements corresponding to the first part (u _j ) and the second part (v _j ) of the element (R _j ) of the random number (R) A three-dimensional vector (Wa) composed of (x component, y component) and a predetermined element (z component) corresponding to the third part (z _j ) of the element (R _j ) of the random number (R ) is generated. Second vector generating means;
Rotation processing means for performing predetermined rotation processing on the three-dimensional vector (Wa) according to the fourth part (θ _j ) and the fifth part (φ _j ) of the element (R _j ) of the random number (R) ,
The elements (x component, y component) of the three-dimensional vector (Wa ′) after the rotation processing are replaced with corresponding elements of the temporary (n + j) dimensional vector (W _{n + j} ′), and the three-dimensional vector (Wa ′). ) Element (z component) as the (n + j) th element of the temporary (n + j) dimensional vector (W _{n + j} ′);
With respect to the element of the temporary (n + j) dimensional vector (W _{n + j} ′) after the replacement , the seventh part (in the direction corresponding to the value of the sixth part (e _j ) of the element (R _j ) of the random number (R) ( d _j ) is cyclically shifted by the number of bits corresponding to the value of j j, and a temporary (n + j) dimensional vector (W _{n + j} ′ = W ₁ ′,..., W _{n + j−1} ′, 0) when 1 is added to _j Bit shift means to obtain
Biological data after conversion (X ′, 2n bits) composed of predetermined bits corresponding to each element of the 2n-dimensional vector (W _2n ) after the cyclic shift generated while j is added one by one from 1 to n. Conversion data generation means for generating,
A biometric authentication system characterized by that. The measuring device and the authentication device each include a communication means for transmitting and receiving data via a communication line,
The registration data and the by the encryption method using a key that is disposable encrypts the authentication data, to perform the confidentiality of data on the communication line, and claim 1 or claim, characterized in that to prevent spoofing by others Item 3. The biometric authentication system according to Item 2 . The biometric authentication system according to claim 3 , wherein the storage unit stores the encrypted registration data. A biometric authentication method in a biometric authentication system comprising a biometric data measurement device and an authentication device that authenticates based on biometric data measured by the measurement device,
The measuring device has conversion means,
The authentication device includes a storage unit, a similarity calculation unit, and an authentication unit,
When registering a user, the conversion means performs conversion having non-linearity configured by combining linear conversion and random numbers on the biological data measured by the measurement device, and the storage means uses the conversion means. Store the converted registration data,
When authenticating a user, the conversion means performs conversion having non-linearity configured by combining linear conversion and random numbers on the biometric data measured by the measuring device, and the similarity calculation means performs the conversion The degree of similarity between the data to be authenticated converted by the means and the registered data stored in the storage means is calculated by the normalized Hamming distance, and the authentication means determines the similarity and authenticates according to the determination result. ,
The converting means includes a random number generating means, a data connecting means, a bit rearranging means, a first vector generating means, a second vector generating means, a rotation processing means, a third vector generating means, a converted data generating means, A conversion table holding a plurality (2n) of bit rearrangement rules,
When converting biometric data,
The random number generation means generates a random number (R = (R _e R _p R ₁ ... R _j ... R _{3n / 2} ), j = 1,..., 3n / 2, {n + log ₂ 2n + (2 log ₂ 2n + 2 ) · 3n / 2} bits, R _e is n bits, R _p is log ₂ 2n bits, R _j is (2 log ₂ 2n + 2) bits, R _j = (u _j v _j θ _j ), u _j and v _j are log ₂ 2n bits, θ _j is 2 bits)
The data concatenation means concatenates the element (R _e ) of the random number (R ) to biometric data (X, n bits);
The bit rearranging unit selects one of the sorting rules from the conversion table based on the element (R _p) of the random number (R), the bit of the connection data in accordance with the sort rule selected (Y, 2n bits) Sort
The first vector generating means generates a 2n-dimensional vector (W) including predetermined elements corresponding to each bit of the rearranged data (Z, 2n bits);
The second vector generating means corresponds to the first part (u _j ) and the second part (v _j ) of each element (R _j ) of the random number (R) among the elements of the 2n-dimensional vector (W). A two-dimensional vector (Wa) corresponding to each element (R _j ) of the random number (R ) is generated from each element
The rotation processing means performs a predetermined rotation process on the two-dimensional vector (Wa) according to the third part (θ _j ) of each element (R _j ) of the random number (R) ,
The third vector generating means generates a 2n-dimensional vector (W ′) from the elements of the rotated two-dimensional vector (Wa ′);
The converted data generating means generates converted biometric data (X ′, 2n bits) composed of predetermined bits corresponding to each element of the 2n-dimensional vector (W ′);
A biometric authentication method. A biometric authentication method in a biometric authentication system comprising a biometric data measurement device and an authentication device that authenticates based on biometric data measured by the measurement device,
The measuring device has conversion means,
The authentication device includes a storage unit, a similarity calculation unit, and an authentication unit,
When registering a user, the conversion means performs conversion having non-linearity configured by combining linear conversion and random numbers on the biological data measured by the measurement device, and the storage means uses the conversion means. Store the converted registration data,
When authenticating a user, the conversion means performs conversion having non-linearity configured by combining linear conversion and random numbers on the biometric data measured by the measuring device, and the similarity calculation means performs the conversion The degree of similarity between the data to be authenticated converted by the means and the registered data stored in the storage means is calculated by the normalized Hamming distance, and the authentication means determines the similarity and authenticates according to the determination result. ,
The conversion means includes random number generation means, first vector generation means, second vector generation means, replacement means, bit shift means, and conversion data generation means,
When converting biometric data,
The random number generation means is a random number (R = (R ₁ ... R _j ... R _n ), j = 1,..., N, (3log ₂ n + 6) n bits, R _j = (u _j v _j z _j θ _j φ _j e _j d _j ), u _j , v _j , e _j are log ₂ n bits, θ _j , φ _j are 2 bits, and z _j , d _j are 1 bit)
The first vector generating means uses a predetermined element corresponding to each bit of the biometric data (X, n bits), and uses a (n + j−1) -dimensional vector (W _{n + j−1} = W ₁ ,..., W _{n + j−1} ) and the first temporary (n + j) dimensional vector (W _{n + j} ′ = W ₁ ′,..., W _{n + j−1} ′, 0),
The second vector generation means includes a first part (u _j ) and a second part ( R _j ) of the element (R _j ) of the random number (R) among the elements of the (n + j−1) dimensional vector (W _{n + j−1} ). 3 consisting of each element (x component, y component) corresponding to v _j ) and a predetermined element (z component) corresponding to the third part (z _j ) of the element (R _j ) of the random number (R) Generate a dimension vector (Wa)
The rotation processing means performs a predetermined rotation process on the three-dimensional vector (Wa) according to the fourth part (θ _j ) and the fifth part (φ _j ) of the element (R _j ) of the random number (R). Done
The replacement means replaces the elements (x component, y component) of the three-dimensional vector (Wa ′) after the rotation processing with corresponding elements of the temporary (n + j) dimensional vector (W _{n + j} ′), and the three-dimensional Let the element (z component) of the vector (Wa ′) be the (n + j) th element of the temporary (n + j) dimensional vector (W _{n + j} ′),
The direction according to the value of the sixth part (e _j ) of the element (R _j ) of the random number (R) with respect to the element of the temporary (n + j) dimensional vector (W _{n + j} ′) after the bit shift means A cyclic (n + j) -dimensional vector (W _{n + j} ′ = W ₁ ′,..., W _{n + j−} when a cyclic shift is performed by the number of bits corresponding to the value of the seventh part (d _j ) and 1 is added to _{j. 1} ', 0)
Biometric data (X after conversion) composed of predetermined bits corresponding to each element of the 2n-dimensional vector (W _2n ) after the cyclic shift generated by the conversion data generation means (j is incremented by 1 from 1 to n ) ', 2n bits)
A biometric authentication method.

Images