JP2003044858A - Device and method for personal identification - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device for personal identification with sufficiently high reliability which corresponds to accuracy fluctuations due to an individual physical condition change, the change with time and use environment difference. SOLUTION: This device for personal identification is provided with a biological information 1 inputting means 1 and a biological information 2 inputting means 4 for acquiring first and second biological information, a biological information 1 reliability determining means 2 for determining the reliability of the first biological information on the basis of the first biological information and its input environment information, a biological information 1 collating means 3 for collating the fist biological information with biological information 1 registered data 8, a biological information 2 reliability determining means 5 for determining the reliability of the second biological information on the basis of the second biological information and its input environment information, a biological information 2 collating means 6 for collating the second biological information with biological information 2 registered data, and a compound deciding means 7 for performing person himself/herself decision on the basis of the reliabilities determined by the biological information 1 reliability determining means 2 and the biological information 2 reliability determining means 5 and collation results by the biological information 1 collating means 3 and the biological information 2 collating means 6.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a personal authentication device and method for identifying an individual by using biometric information, and in particular,
The present invention relates to an individual authentication device and method capable of coping with changes in physical condition of an individual, changes over time, and changes in use environment.

[0002]

2. Description of the Related Art Conventionally, a personal authentication device using biometric information such as a fingerprint, an iris, a voiceprint, a face, and a palm type has been adopted for applications such as entry / exit management and personal computer access. Authentication using biometric information is an excellent authentication method in terms of security and ease of use because it is less likely to be forgotten, lost, or stolen than authentication using non-biometric information such as a password or an IC card.

Although this biometric information is unique and unique to an individual, the same characteristics cannot always be extracted. In particular, the characteristics may change greatly due to changes in the physical condition of the individual, changes over time, and differences in the usage environment. As a personal authentication method using biometric information, the entered biometric information is compared with biometric information registration data that has been registered in advance, and a similarity representing how similar is calculated is calculated. It is a general method to judge that the person is the person if the set threshold value is exceeded, and to judge another person if the threshold value is not exceeded.

Therefore, in the prior art, it is general that the same authentication method is used for individual authentication in any case, without taking into consideration changes in physical condition of the individual, changes over time, and differences in use environment.

[0005]

However, the conventional personal authentication method does not take into consideration changes in the physical condition of the individual, changes over time, and differences in the usage environment, so that the authentication accuracy cannot be sufficiently guaranteed, and the authentication is rejected. There is a problem that even if there are easy people, it is not possible to respond appropriately. The present invention has been made to solve such a problem, and an object of the present invention is to provide a highly reliable personal authentication device that can cope with accuracy changes due to changes in physical condition of the individual, changes over time, and differences in use environment.

[0006]

A personal identification device of the present invention comprises biometric information 1 input means for acquiring first biometric information,
Biometric information 1 reliability determination means for determining the reliability of the first biometric information based on the first biometric information acquired by the biometric information 1 input means and the input environment information of the first biometric information; A biometric information 1 collating unit that collates the first biometric information acquired by the biometric information 1 input unit with biometric information 1 registration data in which information including the first biometric information is registered in advance; The reliability of the second biometric information based on the biometric information 2 input means for acquiring the biometric information, the second biometric information acquired by the biometric information 2 input means, and the input environment information of the second biometric information. The biometric information 2 reliability determining means for determining the degree, the biometric information 2 registration data in which the second biometric information acquired by the biometric information 2 input means and the information including the second biometric information are registered in advance. Biological information to match And 2 comparing means, reliability is determined by the biometric information 1 reliability determining means, the biometric information 2
And a composite determination means for performing identity determination, which is determination of whether or not the person is the identity, based on the reliability determined by the reliability determination means, the verification result by the biometric information 1 verification means, and the verification result by the biometric information 2 verification means. It has a configuration including. With this configuration, the reliability of a plurality of input biometric information is determined for each authentication, and by setting a threshold value according to this reliability, the physical condition change or secular change of an individual,
Considering the difference in usage environment, high authentication accuracy and reliability can be realized, and even if the user who is easily rejected the authentication based on one biometric information, the reliability of the authentication based on the other biometric information is high. It is possible to realize a personal authentication device that can be authenticated and can reduce the number of persons who cannot be authenticated and make it easy to use.

Further, the personal authentication apparatus of the present invention includes biometric information 1 input means for acquiring first biometric information, first biometric information acquired by the biometric information 1 input means, and the first biometric information. Biometric information 1 reliability determination means for determining the reliability of the first biometric information based on the input environment information of the first biometric information, the first biometric information acquired by the biometric information 1 input means, and the first biometric information The information including the information acquired by the biometric information 1 collating means for collating the preregistered biometric information 1 registration data, the biometric information 2 input means for acquiring the second biometric information, and the biometric information 2 input means. Biometric information 2 for determining the reliability of the second biometric information based on the second biometric information and the input environment information of the second biometric information.
Biometric information 2 collating means for collating the reliability determining means with the biometric information 2 registration data in which the second biometric information acquired by the biometric information 2 input means and the information including the second biometric information are registered in advance. And the reliability determined by the biometric information 1 reliability determination means, the reliability determined by the biometric information 2 reliability determination means, the verification result by the biometric information 1 verification means, and the biometric information 2 verification means. And a composite determination means for determining the identity based on the collation result, wherein the first biometric information and the second biometric information are fingerprints, voiceprints, or faces, respectively. Of different information. With this configuration, by using different information about the fingerprint, voiceprint, or face as the biometric information,
It is possible to realize an individual authentication device having high authentication accuracy and reliability.

Further, the personal authentication apparatus of the present invention includes biometric information 1 input means for acquiring first biometric information, first biometric information acquired by the biometric information 1 input means, and the first biometric information. Biometric information 1 reliability determination means for determining the reliability of the first biometric information based on the input environment information of the first biometric information, the first biometric information acquired by the biometric information 1 input means, and the first biometric information The information including the information acquired by the biometric information 1 collating means for collating the preregistered biometric information 1 registration data, the biometric information 2 input means for acquiring the second biometric information, and the biometric information 2 input means. Biometric information 2 for determining the reliability of the second biometric information based on the second biometric information and the input environment information of the second biometric information.
Biometric information 2 collating means for collating the reliability determining means with the biometric information 2 registration data in which the second biometric information acquired by the biometric information 2 input means and the information including the second biometric information are registered in advance. And the reliability determined by the biometric information 1 reliability determination means, the reliability determined by the biometric information 2 reliability determination means, the verification result by the biometric information 1 verification means, and the biometric information 2 verification means. And a composite determination unit that performs identity determination, which is determination of whether or not the user is the identity based on the collation result, using fingerprint information as the first biometric information or the second biometric information, and the fingerprint information is The biometric information 1 reliability determining unit or the biometric information 2 reliability determining unit that is input determines the reliability according to the number of feature points including the branch points and end points of the input fingerprint. It has. With this configuration, the reliability of the fingerprint information input for each authentication is determined, and by setting the threshold value according to this reliability, consideration is given to changes in individual physical condition, secular change, and differences in usage environment. As a result, it is possible to realize a personal authentication device having high authentication accuracy and reliability.

Further, the personal identification device of the present invention includes biometric information 1 input means for obtaining the first biometric information, the first biometric information obtained by the biometric information 1 input means, and the first biometric information. Biometric information 1 reliability determination means for determining the reliability of the first biometric information based on the input environment information of the first biometric information, the first biometric information acquired by the biometric information 1 input means, and the first biometric information The information including the information acquired by the biometric information 1 collating means for collating the preregistered biometric information 1 registration data, the biometric information 2 input means for acquiring the second biometric information, and the biometric information 2 input means. Biometric information 2 for determining the reliability of the second biometric information based on the second biometric information and the input environment information of the second biometric information.
Biometric information 2 collating means for collating the reliability determining means with the biometric information 2 registration data in which the second biometric information acquired by the biometric information 2 input means and the information including the second biometric information are registered in advance. And the reliability determined by the biometric information 1 reliability determination means, the reliability determined by the biometric information 2 reliability determination means, the verification result by the biometric information 1 verification means, and the biometric information 2 verification means. And a composite determination unit that determines whether the person is a person based on a collation result. The voiceprint information is used as the first biometric information or the second biometric information. The biometric information 1 reliability determining means or the biometric information 2 reliability determining means that is input determines the reliability according to the background noise of the input voiceprint and the age of the voiceprint registration data. The it has. With this configuration, the reliability of the voiceprint information input for each authentication is determined, and by setting the threshold value according to this reliability, consideration is given to changes in individual physical condition, secular changes, and differences in usage environment. As a result, it is possible to realize a personal authentication device having high authentication accuracy and reliability.

Further, the personal identification device of the present invention includes biometric information 1 input means for obtaining the first biometric information, the first biometric information obtained by the biometric information 1 input means, and the first biometric information. Biometric information 1 reliability determination means for determining the reliability of the first biometric information based on the input environment information of the first biometric information, the first biometric information acquired by the biometric information 1 input means, and the first biometric information The information including the information acquired by the biometric information 1 collating means for collating the preregistered biometric information 1 registration data, the biometric information 2 input means for acquiring the second biometric information, and the biometric information 2 input means. Biometric information 2 for determining the reliability of the second biometric information based on the second biometric information and the input environment information of the second biometric information.
Biometric information 2 collating means for collating the reliability determining means with the biometric information 2 registration data in which the second biometric information acquired by the biometric information 2 input means and the information including the second biometric information are registered in advance. And the reliability determined by the biometric information 1 reliability determination means, the reliability determined by the biometric information 2 reliability determination means, the verification result by the biometric information 1 verification means, and the biometric information 2 verification means. And a composite determination unit that performs identity determination, which is determination of whether or not the person is the identity based on the collation result, using face information as the first biometric information or the second biometric information, and the face information is The biometric information 1 reliability determining means or the biometric information 2 reliability determining means that is input has a configuration that determines the reliability according to the input face illumination level and the age of the face registration data. That. With this configuration, the reliability of the face information input for each authentication is determined, and by setting the threshold value according to this reliability, consideration is given to changes in individual physical condition, secular changes, and differences in usage environment. As a result, it is possible to realize a personal authentication device having high authentication accuracy and reliability.

Further, the personal identification device of the present invention includes biometric information 1 input means for obtaining the first biometric information, the first biometric information obtained by the biometric information 1 input means, and the first biometric information. Biometric information 1 reliability determination means for determining the reliability of the first biometric information based on the input environment information of the first biometric information, the first biometric information acquired by the biometric information 1 input means, and the first biometric information The information including the information acquired by the biometric information 1 collating means for collating the preregistered biometric information 1 registration data, the biometric information 2 input means for acquiring the second biometric information, and the biometric information 2 input means. Biometric information 2 for determining the reliability of the second biometric information based on the second biometric information and the input environment information of the second biometric information.
Biometric information 2 collating means for collating the reliability determining means with the biometric information 2 registration data in which the second biometric information acquired by the biometric information 2 input means and the information including the second biometric information are registered in advance. And the reliability determined by the biometric information 1 reliability determination means, the reliability determined by the biometric information 2 reliability determination means, the verification result by the biometric information 1 verification means, and the biometric information 2 verification means. A composite determination means for determining the identity based on the collation result, and a biometric information 1 reliability determination means when the identity is determined in the identity determination by the composite determination means. The biometric information 1 registration data is updated with the acquired first biometric information based on the reliability, and based on the reliability determined by the biometric information 2 reliability determining means. There have has a structure in which a registration update means for updating the biometric information 2 registered data in the acquired second biometric information. With this configuration, the reliability of multiple pieces of biometric information that has been input for each authentication is determined, and by setting the threshold value according to this reliability, changes in individual physical condition, secular change, and differences in usage environment can be detected. In consideration of,
Even if a user who can achieve high authentication accuracy and reliability and is easily rejected from authentication based on one biometric information, authentication can be performed if the reliability of authentication based on the other biometric information is high. A personal authentication device that can be reduced and made easier to use, and by comparing reliability and updating old registration data with new input data to prevent accuracy deterioration due to aging and to achieve high reliability. Can be realized.

Further, the personal identification device of the present invention includes biometric information 1 input means for acquiring first biometric information, first biometric information acquired by the biometric information 1 input means, and the first biometric information. Biometric information 1 reliability determination means for determining the reliability of the first biometric information based on the input environment information of the first biometric information, the first biometric information acquired by the biometric information 1 input means, and the first biometric information The information including the information acquired by the biometric information 1 collating means for collating the preregistered biometric information 1 registration data, the biometric information 2 input means for acquiring the second biometric information, and the biometric information 2 input means. Biometric information 2 for determining the reliability of the second biometric information based on the second biometric information and the input environment information of the second biometric information.
Biometric information 2 collating means for collating the reliability determining means with the biometric information 2 registration data in which the second biometric information acquired by the biometric information 2 input means and the information including the second biometric information are registered in advance. And the reliability determined by the biometric information 1 reliability determination means, the reliability determined by the biometric information 2 reliability determination means, the verification result by the biometric information 1 verification means, and the biometric information 2 verification means. A composite determination means for determining the identity based on the collation result, and a biometric information 1 reliability determination means when the identity is determined in the identity determination by the composite determination means. The biometric information 1 registration data is updated with the acquired first biometric information based on the reliability, and based on the reliability determined by the biometric information 2 reliability determining means. And updating the biometric information 2 registration data with the acquired second biometric information, wherein the registration updating means has the reliability determined by the biometric information 1 reliability determining means. Biometric information 1
It is configured to be updated when the reliability is higher than the reliability of the registration data, and to be updated when the reliability determined by the biometric information 2 reliability determining means is higher than the reliability of the biometric information 2 registration data. . With this configuration, the reliability of multiple pieces of biometric information that has been input for each authentication is determined, and by setting the threshold value according to this reliability, changes in individual physical condition, secular change, and differences in usage environment can be detected. Considering this, it is possible to realize high authentication accuracy and reliability, and even users who are easily rejected authentication based on one biometric information will be able to perform authentication if the reliability of authentication based on the other biometric information is high. It is possible to reduce the number of impossible users and make it easier to use, and by comparing the reliability and updating the old registration data with new input data, it is possible to prevent accuracy deterioration due to aging and to achieve high reliability, A personal authentication device can be realized.

In the personal authentication method of the present invention, the biometric information 1 input step for obtaining the first biometric information, the first biometric information obtained in the biometric information 1 input step, and the first biometric information Biometric information 1 reliability determination step of determining the reliability of the first biometric information based on the input environment information of the first biometric information, and the first biometric information 1 acquired in the biometric information 1 input step.
Biometric information 1 collating step of collating biometric information of 1) with biometric information 1 registration data in which information including the first biometric information is registered in advance; biometric information 2 input step of acquiring second biometric information; A biometric information 2 reliability determination step of determining the reliability of the second biometric information based on the second biometric information acquired in the biometric information 2 input step and the input environment information of the second biometric information; Biometric information 2 collating step of collating the second biometric information acquired in the biometric information 2 input step with biometric information 2 registration data in which information including the second biometric information is registered in advance;
The reliability determined in the reliability determination step, the reliability determined in the biometric information 2 reliability determination step, the verification result in the biometric information 1 verification step, and the biometric information 2
And a composite determination step of performing identity determination, which is determination of whether or not the user is the identity, based on the result of the matching in the matching step. With this configuration, the reliability of multiple pieces of biometric information that has been input for each authentication is determined, and by setting the threshold value according to this reliability, changes in individual physical condition, secular change, and differences in usage environment can be detected. Considering this, it is possible to realize high authentication accuracy and reliability, and even users who are easily rejected authentication based on one biometric information will be able to perform authentication if the reliability of authentication based on the other biometric information is high. It is possible to realize a personal authentication method that can reduce the number of impossible persons and make it easier to use.

In the personal authentication method of the present invention, a fingerprint information input step of obtaining fingerprint information, a fingerprint feature point calculation step of calculating feature points from the fingerprint information obtained in the fingerprint information input step, and the feature The fingerprint information reliability determination step of determining the reliability of the fingerprint information according to the number of points, the fingerprint information acquired in the fingerprint information input step and the fingerprint information registration data in which the fingerprint information is registered in advance are collated. A fingerprint information collating step for performing a voiceprint information input step for obtaining voiceprint information, a background noise measuring step for measuring background noise using a microphone, and a collation with the voiceprint information obtained in the voiceprint information input step; A registration elapsed time calculating step of calculating a registration elapsed time of voiceprint information registration data in which voiceprint information is registered in advance, the background noise and the registration elapsed time A voiceprint information reliability determining step for determining the reliability of the voiceprint information based on the above, a voiceprint information collating step for collating the voiceprint information acquired in the voiceprint information input step with the voiceprint information registration data, and the fingerprint Based on the reliability determined in the information reliability determination step, the reliability determined in the voiceprint information reliability determination step, the verification result in the fingerprint information verification step, and the verification result in the voiceprint information verification step. And a composite determination step for determining whether or not it is included. With this configuration, the reliability of the fingerprint information and voiceprint information input for each authentication is determined, and by setting the threshold value according to this reliability, changes in physical condition and secular change of individuals, differences in usage environment, etc. In consideration of the above, it is possible to realize an individual authentication method having high authentication accuracy and reliability.

In the personal authentication method of the present invention, the voiceprint information input step of acquiring voiceprint information, the background noise measuring step of measuring background noise using a microphone, and the voiceprint information acquired in the voiceprint information input step. The registration elapsed time calculation step of calculating the registration elapsed time of the voiceprint information registration data in which the voiceprint information is registered in advance, and the reliability of the voiceprint information is determined based on the background noise and the registration elapsed time. A voiceprint information reliability determination step, a voiceprint information collation step of collating the voiceprint information acquired in the voiceprint information input step with the voiceprint information registration data, a face information input step of obtaining face information, and using a camera A face for which the face information is registered in advance, which is to be compared with the face information acquired in the face information inputting step and an illumination measuring step for measuring the illumination. Registration elapsed time calculation step for calculating the registration elapsed time of the report registration data; face information reliability determination step for determining the reliability of the face information based on the illumination degree and the registration elapsed time; and the face information input. Face information collation step of collating the face information acquired in step with the face information registration data, reliability determined in the voiceprint information reliability determination step, reliability determined in the face information reliability determination step , A composite determination step of determining whether or not the person is the person based on the verification result of the voice print information verification step and the verification result of the face information verification step. With this configuration, the reliability of the voiceprint information and face information input for each authentication is determined, and by setting the threshold value according to this reliability, changes in individual physical condition, changes over time, and differences in usage environment. In consideration of the above, it is possible to realize an individual authentication method having high authentication accuracy and reliability.

In the personal authentication method of the present invention, the biometric information 1 input step for obtaining the first biometric information, the first biometric information obtained in the biometric information 1 input step, and the first biometric information. Biometric information 1 reliability determination step of determining the reliability of the first biometric information based on the input environment information of the first biometric information, and the first biometric information 1 acquired in the biometric information 1 input step.
Biometric information 1 collating step of collating biometric information of 1) with biometric information 1 registration data in which information including the first biometric information is registered in advance; biometric information 2 input step of acquiring second biometric information; A biometric information 2 reliability determination step of determining the reliability of the second biometric information based on the second biometric information acquired in the biometric information 2 input step and the input environment information of the second biometric information; Biometric information 2 collating step of collating the second biometric information acquired in the biometric information 2 input step with biometric information 2 registration data in which information including the second biometric information is registered in advance;
The reliability determined in the reliability determination step, the reliability determined in the biometric information 2 reliability determination step, the verification result in the biometric information 1 verification step, and the biometric information 2
A composite determination step of performing identity determination to determine whether or not the user is the identity based on the verification result in the verification step; and the biometric information 1 reliability determination when the identity is determined in the identity determination in the composite determination step. The biometric information 1 registration data is updated with the acquired first biometric information based on the reliability determined in the step, and the biometric information is updated based on the reliability determined in the biometric information 2 reliability determining step. Registration update step of updating two registration data with the acquired second biometric information.
With this configuration, the reliability of multiple pieces of biometric information that has been input for each authentication is determined, and by setting the threshold value according to this reliability, changes in individual physical condition, secular change, and differences in usage environment can be detected. Considering this, it is possible to realize high authentication accuracy and reliability, and even users who are easily rejected authentication based on one biometric information will be able to perform authentication if the reliability of authentication based on the other biometric information is high. It is possible to reduce the number of impossible users and make it easier to use, and by comparing the reliability and updating the old registration data with new input data, it is possible to prevent accuracy deterioration due to aging and to achieve high reliability,
A personal authentication method can be realized.

[0017]

BEST MODE FOR CARRYING OUT THE INVENTION A personal authentication device according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a personal authentication device that uses reliability for personal authentication according to the first embodiment of this invention. In FIG. 1, the personal authentication device 100 includes a biometric information 1 input unit 1,
Biometric information 1 reliability determination means 2, biometric information 1 collation means 3,
Biometric information 2 input means 4, biometric information 2 reliability determination means 5,
It is composed of biometric information 2 collating means 6 and composite judging means 7.

The biometric information 1 input means 1 acquires any one of biometric information such as fingerprints, voiceprints and faces necessary for authentication. Hereinafter, the acquired information will be referred to as first biometric information.

The biometric information 1 reliability determination means 2 determines the first biometric information from the first biometric information acquired by the biometric information 1 input means 1, the use environment at the time of biometric information input, the information of reference registration data, and the like. The reliability a (0 ≦ a ≦ 1) is determined. At this time, when there are a plurality of (n) evaluation items regarding the reliability of the first biometric information, the reliability a ₁ , a ₂ , ..., A _n (0 ≦ a ₁ , A ₂ , ..., a _n ≦
1) is calculated, and the reliability a of the first biometric information is a = a ₁ ×
It is defined as a ₂ × ... × a _n .

The biometric information 1 collating means 3 extracts the characteristic amount necessary for collation from the first biometric information acquired by the biometric information 1 input means 1, and collates with the biometric information 1 registration data 8 registered in advance. Then, the similarity of the first biometric information is calculated. The biometric information 2 input unit 4 acquires information other than the first biometric information acquired by the biometric information 1 input unit 1 among biometric information such as fingerprints, voiceprints, and faces necessary for authentication. Hereinafter, the information acquired by the biometric information 2 input means 4 will be referred to as second biometric information.

The biometric information 2 reliability determination means 5 determines the second biometric information from the second biometric information acquired by the biometric information 2 input means 4, the use environment at the time of biometric information input, the information of reference registration data, and the like. Determine the reliability b (0 ≤ b ≤ 1). At this time, when a plurality of (n) evaluation items exist for the reliability of the second biometric information, the reliability for each evaluation item
Calculate b ₁ , b ₂ , ..., b _n (0 ≤ b ₁ , b ₂ , ..., b _n ≤ 1),
The reliability b of the second biometric information is defined as b = b ₁ × b ₂ × ... × b _n .

The biometric information 2 collating means 6 extracts the characteristic amount required for collation from the second biometric information acquired by the biometric information 2 input means 4, and collates with the biometric information 2 registration data 9 registered in advance. Then, the degree of similarity of the second biometric information is calculated. The composite determination unit 7 comprehensively determines whether or not the person is the person, using the reliability and similarity of the first biometric information and the reliability and similarity of the second biometric information.

The operation of the composite judging means 7 will be described in detail below. In order to perform the composite determination, the reliability a of the first biometric information and the reliability b of the second biometric information determined by the biometric information 1 reliability determination unit 2 and the biometric information 2 reliability determination unit 5 are used, A threshold value for determining whether the person is the person is determined. FIG. 2 is a diagram showing a two-dimensional similarity plane having coordinate axes of the similarity x of the first biometric information and the similarity y of the second biometric information.

When setting a linear threshold value on the two-dimensional similarity plane, the threshold value is set as shown in equation (1). ax + by = t (1) Here, t is a parameter. The sum of the false acceptance rate (the rate at which others are mistakenly recognized as the person) FAR and the false rejection rate (the rate at which the person is mistakenly recognized as the other person) FRR obtained by using the learning sample as the value of parameter t By setting the value t ₀ that minimizes, it is possible to set a threshold value that places importance on both security and ease of use.

Further, if the value of the parameter t is set to t> t ₀ , the FAR becomes small, so that the threshold value is set with emphasis on security. If the value of parameter t is set to t0, FRR
Becomes smaller, the threshold value is set with emphasis on ease of use. The value of the parameter t may be set according to the required accuracy of the device used. The value of the parameter t may be calculated for each authentication, but the value of the parameter t is calculated in advance for the combination of the reliability a and b, and the table is referred to for each authentication to perform the authentication process. Time can be shortened.

As shown in FIG. 2, the slope of the straight line representing the threshold value is determined by the reliability a of the first biometric information and the reliability b of the second biometric information. Here, in the case of a> b, the inclination becomes steep like the straight line 113 shown in FIG.
When a = b, the slope is a straight line with a slope of -1, like the straight line 112 shown in FIG. When a <b, the straight line 111 in FIG.
The inclination becomes gentle like.

FIG. 3 is a diagram showing an area determined to be the original person and an area determined to be another person on the two-dimensional similarity plane. As shown in FIG. 3, when the relation of a> b is satisfied, the slope of the straight line becomes steep, and the similarity x of the first biological information and the second
The coordinate point specified by the similarity y of the biometric information of
The person is determined to be the person when the person is located in the area A of the straight line 114 or more, and is determined to be the other person when the person is located in the area B of the straight line 114 or less.

In the case of the threshold setting shown in FIG. 3, even if the similarity of the second biometric information is low, if the reliability and the similarity of the first biometric information are high, it is possible to determine the person. Therefore, it is possible to determine that the user who has a low degree of similarity in one of the biometric information always is the user without rejecting. As the threshold setting method in the personal identification device of the first embodiment, the case where the threshold is set by the straight line represented by the equation (1) has been described. Alternatively, it is possible to set the threshold value by a curve represented by another function of x and y such as x ² and y ² .

As described above, the personal identification device according to the first embodiment of the present invention determines the reliability of a plurality of input biometric information for each authentication, and sets the threshold value according to this reliability. By setting the, changes in physical condition and changes over time,
High authentication accuracy and reliability can be realized in consideration of the difference in usage environment. Further, even a user who is easily rejected by authentication based on one biometric information can be authenticated if the reliability of the authentication based on the other biometric information is high, and it is possible to reduce unauthenticated persons and make it easier to use.

Next, a personal authentication device according to a second embodiment of the present invention will be described with reference to the drawings. FIG. 4 is a block diagram showing the configuration of a personal authentication device that uses the reliability of fingerprint information for personal authentication according to the second embodiment of this invention.
In FIG. 4, the personal authentication device 200 includes a fingerprint information input unit 10, a fingerprint information reliability determination unit 11, a fingerprint information collation unit 12, a fingerprint sensor 13, and a composite determination unit 7 (not shown).

The fingerprint information input means 10 is a fingerprint sensor 1
3 acquires the fingerprint information collected. The fingerprint information reliability determination means 11 determines the reliability a (0 ≦ a ≦ 1) using the fingerprint information acquired by the fingerprint information input means 10.
Hereinafter, the reliability determined by the fingerprint information reliability determining means 11 is referred to as fingerprint reliability. FIG. 7 is an explanatory diagram of a method for determining the fingerprint reliability. Fingerprint information reliability determination means 11
Detects the branch points and end points, which are the characteristic points of the fingerprint, from the fingerprint information acquired by the fingerprint information input means 10 and obtains the total number.

As shown in FIG. 7, when the number of characteristic points is too small, the characteristic points are not accurately detected because the fingers are dry or are wet and crushed by sweat, and the finger placement is Since there may be a problem, the fingerprint information reliability determination means 11 sets the reliability to low. On the contrary, when the number of feature points is too large, it is considered that a lot of noise is detected, and thus the reliability is set low in this case as well. In the case of an intermediate number of feature points, there is a high possibility that the feature points have been correctly detected, so the reliability is set high.

The fingerprint information collating means 12 extracts the characteristic amount required for collation from the fingerprint information acquired by the fingerprint information input means 10 and collates it with the fingerprint information registration data 14 that has been registered in advance to obtain the fingerprint information. Calculate the similarity.

Here, the fingerprint information input means 10 is
Corresponding to the biometric information 1 input means 1 or biometric information 2 input means 4 shown in FIG. 1, the fingerprint information reliability determining means 11 corresponds to the biometric information 1 reliability determining means 2 or the biometric information 2 reliability determining means 5. The fingerprint information collating means 12 corresponds to the biometric information 1 collating means 3 or the biometric information 2 collating means 6.
The composite determination process is performed after the process by the fingerprint information collating unit 12, and the same process as each process described in the first embodiment of the present invention is performed for this composite determination process.

As described above, the personal identification device according to the second embodiment of the present invention determines the reliability of the fingerprint information input for each authentication, and sets the threshold value according to this reliability. By performing the above, it is possible to realize high authentication accuracy and reliability in consideration of changes in physical condition of the individual, changes over time, and differences in use environment.

Next, a personal authentication device according to a third embodiment of the present invention will be described with reference to the drawings. FIG. 5 is a block diagram showing the configuration of a personal authentication device that uses the reliability of voiceprint information for personal authentication according to the third embodiment of this invention.
In FIG. 5, the personal authentication device 300 includes a voiceprint information input unit 15, a voiceprint information reliability determination unit 16, a voiceprint information collation unit 17, a microphone 18, and a composite determination unit 7 (not shown).

The voiceprint information input means 15 acquires the voiceprint information collected by the microphone 18. Voiceprint information reliability determination means 1
6 determines the reliability a (0 ≦ a ≦ 1) based on the background noise and the registration elapsed time of the reference registration data. Hereinafter, the reliability determined by the voiceprint information reliability determining unit 16 is referred to as voiceprint reliability. FIG. 8 is an explanatory diagram of a method for determining the voiceprint reliability.

The voiceprint information reliability determining means 16 uses the microphone 1
Based on the background noise sampled by 8, the reliability a ₁ (0 ≦ a ₁ ≦ 1) due to the background noise is determined as shown in FIG. That is, when the background noise is large, the features of the voiceprint are drowned out, so the reliability is set low, and when the background noise is small, the features of the voiceprint are easily detected, so the reliability is set high.

The voiceprint information reliability determining means 16 calculates the registration elapsed time based on the date and time when the reference data in the voiceprint information registration data 19 is registered, and the reliability a ₂ is calculated according to the registration elapsed time. (0 ≦ a ₂ ≦ 1) is determined. As shown in FIG. 8, when the registration elapsed time is long, it is considered that the feature difference due to the secular change may occur, so the reliability is set to be low, and when the registration elapsed time is short, the feature difference due to the secular change is low. Since it is considered to be small, the reliability is set high. Voiceprint reliability a, which is the final reliability of the voiceprint, is a
= a ₁ × a ₂ .

The voiceprint information collating means 17 extracts the characteristic amount necessary for collation from the voiceprint information acquired by the voiceprint information inputting means 15, and collates it with the voiceprint information registration data 19 registered in advance to obtain the voiceprint information. Calculate the similarity.

Here, the voiceprint information input means 15 is
Corresponding to the biometric information 1 input means 1 or the biometric information 2 input means 4 shown in FIG. 1, the voiceprint information reliability determining means 16 corresponds to the biometric information 1 reliability determining means 2 or the biometric information 2 reliability determining means 5. The voice print information collating means 17 corresponds to the biometric information 1 collating means 3 or the biometric information 2 collating means 6.
The composite determination processing is performed after the processing by the voice print information collating means 17, and the composite determination processing is similar to the processing described in the first embodiment of the present invention.

As described above, the personal identification device according to the third embodiment of the present invention determines the reliability of the voiceprint information input for each authentication, and sets the threshold value according to this reliability. By performing the above, it is possible to realize high authentication accuracy and reliability in consideration of changes in physical condition of the individual, changes over time, and differences in use environment.

Next, a personal authentication device according to a fourth embodiment of the present invention will be described with reference to the drawings. FIG. 6 is a block diagram showing the configuration of a personal authentication device that uses the reliability of face information for personal authentication according to the fourth embodiment of this invention. In FIG. 6, the personal authentication device 400 includes a face information input means 2
0, face information reliability determination means 21, face information matching means 22,
The camera 23 and the composite determination means 7 (not shown) are used.

The face information input means 20 acquires the face information collected by the camera 23. The face information reliability determination means 21
The reliability a (0 ≦ a ≦ 1) is determined based on the illumination level and the registration elapsed time of the reference registration data. Hereinafter, the reliability determined by the face information reliability determining means 21 is referred to as face reliability. FIG. 9 is an explanatory diagram of a method of determining the face reliability.

The face information reliability determining means 21 is a camera 23.
By using the image collected by, the illuminance of the background image is measured by image processing, and the reliability a ₁ (0 ≦ a ₁ ≦ 1) according to the illuminance is measured.
Are determined as shown in FIG. That is, if the illumination is low or high, the contrast is difficult to obtain,
Since it is possible that the face elements are not cut out accurately, the reliability is set low, and when the illumination is in the middle, contrast is easily obtained and facial features are easily detected. Set higher.

Further, the face information reliability determining means 21 calculates the registration elapsed time based on the date and time when the reference target data in the face information registration data 24 is registered, and the reliability a ₂ (0 ≦ a ₂ ≦ 1) is determined. As shown in FIG. 9, when the registration elapsed time is long, the reliability is set to be low because the feature difference due to the secular change may occur. When the registration elapsed time is short, the feature difference due to the secular change is low. Since it is considered to be small, the reliability is set high. The face reliability a, which is the final face reliability, is a = a ₁ × a
Calculate with ₂ .

The face information collating means 22 is the face information input means 2
The feature amount necessary for matching is extracted from the face information acquired by 0, and the face information registration data 24 registered in advance is compared to calculate the similarity of the face information.

The face information input means 20 corresponds to the biometric information 1 input means 1 or the biometric information 2 input means 4 shown in FIG. 1, and the face information reliability determination means 21 is the biometric information 1 reliability. Determining means 2 or biometric information 2 reliability determining means 5
Corresponding to, the face information collating means 22 corresponds to the biometric information 1 collating means 3 or the biometric information 2 collating means 6. The composite determination processing is performed after the processing by the face information matching unit 22, and the composite determination processing is similar to the processing described in the first embodiment of the present invention.

As described above, the personal identification device of the fourth embodiment of the present invention determines the reliability of the face information input for each authentication, and sets the threshold value according to this reliability. By performing the above, it is possible to realize high authentication accuracy and reliability in consideration of changes in physical condition of the individual, changes over time, and differences in use environment.

Next, a personal authentication device according to a fifth embodiment of the present invention will be described with reference to the drawings. FIG. 10 is a block diagram showing a configuration of a personal authentication device having a registration data registration / update function according to the fifth embodiment of the present invention.
In FIG. 10, the personal authentication device 500 includes a biometric information 1 input unit 25, a biometric information 1 reliability determination unit 26, a biometric information 1 collation unit 27, a biometric information 2 input unit 28, and a biometric information 2
The reliability determining unit 29, the biometric information 2 collating unit 30, the composite determining unit 31, and the registration updating unit 32 are included.

Each of the components from the biometric information 1 input means 25 to the composite determination means 31 that constitute the personal authentication device 500 respectively constitutes the biometric information 1 input means 1 and the biometric information 1 reliability that constitute the personal authentication device 100. Corresponding to the degree determining means 2, the biometric information 1 collating means 3, the biometric information 2 inputting means 4, the biometric information 2 reliability determining means 5, the biometric information 2 collating means 6, and the composite judging means 7, and the first aspect of the present invention. Processing similar to each processing described in the embodiment is performed.

When the composite determining means 31 determines that the person is the person himself / herself, the registration updating means 32 determines that the biometric information 1 is higher than the reliability at the time of registration when the reliability determined by the biometric information 1 reliability determining means 26 is higher. The corresponding data stored in the biometric information 1 registration data 33 is updated with the biometric information input by the input means 25.

Similarly, the registration update means 32 uses the biometric information 2
When the reliability determined by the reliability determining unit 29 is higher than the reliability at the time of registration, the corresponding data stored in the biological information 2 registration data 34 is updated with the biological information input by the biological information 2 input unit 28. To do. The reliability is compared by the reliability of the number of feature points in the case of fingerprint authentication, the reliability of background noise in the case of voiceprint authentication, and the reliability of the degree of illumination in the case of face authentication. Remember.

As described above, the personal authentication device according to the fifth embodiment of the present invention determines the reliability of a plurality of input biometric information for each authentication and sets the threshold value according to this reliability. By setting the, changes in physical condition and changes over time,
High authentication accuracy and reliability can be realized in consideration of the difference in usage environment.

Further, even a user who is likely to be denied authentication based on one biometric information can be authenticated if the reliability of the authentication based on the other biometric information is high, and it is possible to reduce unauthenticated persons and make it easier to use. it can. Furthermore, by comparing the reliability and updating the old registration data with the new input data, it is possible to prevent a decrease in accuracy due to aging and to realize high reliability.

The process in the personal authentication method according to the sixth embodiment of the present invention will be described below with reference to the drawings.
The personal authentication device according to the sixth embodiment is the same as the personal authentication device 100 according to the first embodiment, and description thereof will be omitted. FIG. 11 is a flowchart showing the flow of processing in the personal authentication method according to the sixth embodiment of this invention. In step S101, the biometric information 1 input unit 1 acquires any one of biometric information such as a fingerprint, a voiceprint, and a face, which is necessary for authentication. Hereinafter, the acquired information will be referred to as first biometric information.

In step S102, the biometric information 1 reliability determining unit 2 determines the first from the first biometric information acquired by the biometric information 1 input unit 1, the use environment at the time of biometric information input, the information of reference registration data, and the like. Biometric information reliability a (0 ≦ a ≦ 1)
To decide. At this time, when there are a plurality (n) of evaluation items regarding the reliability of the first biometric information, the reliability a ₁ , a ₂ , ..., A _n (0 ≦
a _1, a _2, ..., and calculates a _n ≦ 1), the reliability a first biometric information is defined as _{a = a 1 × a 2 ×} ... × a n.

In step S103, the biometric information 1 collating means 3 extracts the characteristic amount required for collation from the first biometric information acquired by the biometric information 1 inputting means 1, and preregisters the biometric information 1 registration data 8 And the similarity of the first biometric information is calculated. In step S104, the biometric information 2 input unit 4 acquires information other than the first biometric information acquired by the biometric information 1 input unit 1 among biometric information such as fingerprints, voiceprints, and faces necessary for authentication. Hereinafter, the information acquired by the biometric information 2 input means 4 will be referred to as second biometric information.

In step S105, the biometric information 2 reliability determination means 5 determines whether the second biometric information acquired by the biometric information 2 input means 4, the use environment at the time of biometric information input, the information of reference registration data, etc. The reliability b (0 ≦ b ≦ 1) of the biometric information of is determined. At this time, when there are a plurality of (n) evaluation items regarding the reliability of the second biometric information, the reliability b ₁ , b ₂ , ..., B _n (0 ≦ b ₁ , b ₂ ,
, B _n ≦ 1), and the reliability b of the second biometric information is
It is defined as b = b ₁ × b ₂ × ... × b _n .

In step S106, the biometric information 2 collating means 6 extracts the characteristic amount necessary for collation from the second biometric information acquired by the biometric information 2 input means 4, and preregisters the biometric information 2 registration data 9 in advance. And the similarity of the second biometric information is calculated. In step S107, the composite determination unit 7 comprehensively determines whether or not the person is the original person by using the reliability and similarity of the first biometric information and the reliability and similarity of the second biometric information.

As described above, the personal authentication method according to the sixth embodiment of the present invention determines the reliability of a plurality of input biometric information for each authentication and sets the threshold value according to this reliability. By setting the, changes in physical condition and changes over time,
High authentication accuracy and reliability can be realized in consideration of the difference in usage environment. Further, even a user who is easily rejected by authentication based on one biometric information can be authenticated if the reliability of the authentication based on the other biometric information is high, and it is possible to reduce unauthenticated persons and make it easier to use.

The processing in the personal authentication method according to the seventh embodiment of the present invention will be described below with reference to the drawings.
The personal authentication device according to the seventh embodiment is the same as the personal authentication device 200 according to the second embodiment, and the description thereof is omitted. FIG. 12 is a flowchart showing the flow of processing in the personal authentication method according to the seventh embodiment of this invention. In step S110, the fingerprint information input means 1
0 acquires the fingerprint information collected by the fingerprint sensor 13.

In step S111, the fingerprint information reliability determination means 11 determines the reliability a (0 ≦ a ≦ 1) using the fingerprint information acquired by the fingerprint information input means 10.
Hereinafter, the reliability determined by the fingerprint information reliability determining means 11 is referred to as fingerprint reliability. FIG. 7 is an explanatory diagram of a method for determining the fingerprint reliability. Fingerprint information reliability determination means 11
Detects the branch points and end points, which are the characteristic points of the fingerprint, from the fingerprint information acquired by the fingerprint information input means 10 and obtains the total number.

As shown in FIG. 7, when the number of feature points is too small, the feature points cannot be detected accurately because the finger is dry or is wet and crushed by sweat, and the finger placement is Since there may be a problem, the fingerprint information reliability determination means 11 sets the reliability to low. On the contrary, when the number of feature points is too large, it is considered that a lot of noise is detected, and thus the reliability is set low in this case as well. In the case of an intermediate number of feature points, there is a high possibility that the feature points have been correctly detected, so the reliability is set high.

In step S112, the fingerprint information collating means 1
A reference numeral 2 extracts a feature amount required for collation from the fingerprint information acquired by the fingerprint information input means 10 and collates it with the fingerprint information registration data 14 registered in advance to calculate the similarity of the fingerprint information. In step S107, the composite determination unit 7 comprehensively determines whether or not the user is the original, using the reliability and the similarity of the fingerprint information.

As described above, the personal authentication method according to the seventh embodiment of the present invention determines the reliability of the fingerprint information input for each authentication, and sets the threshold value according to this reliability. By performing the above, it is possible to realize high authentication accuracy and reliability in consideration of changes in physical condition of the individual, changes over time, and differences in use environment.

The process of the personal authentication method according to the eighth embodiment of the present invention will be described below with reference to the drawings.
The personal authentication device according to the eighth embodiment is the same as the personal authentication device 300 according to the third embodiment, and a description thereof will be omitted. FIG. 13 is a flowchart showing the flow of processing in the personal authentication method according to the eighth embodiment of the present invention. In step S115, the voice print information input means 1
5 acquires the voice print information collected by the microphone 18.

In step S116, the voiceprint information reliability determining means 16 determines the reliability a (0≤a≤1) based on the background noise and the elapsed registration time of the reference registration data. Hereinafter, the reliability determined by the voiceprint information reliability determining unit 16 is referred to as voiceprint reliability. FIG. 8 is an explanatory diagram of a method for determining the voiceprint reliability.

The voiceprint information reliability determining means 16 uses the microphone 1
Based on the background noise sampled by 8, the reliability a ₁ (0 ≦ a ₁ ≦ 1) due to the background noise is determined as shown in FIG. That is, when the background noise is large, the features of the voiceprint are drowned out, so the reliability is set low, and when the background noise is small, the features of the voiceprint are easily detected, so the reliability is set high.

The voiceprint information reliability determination means 16 calculates the registration elapsed time based on the date and time when the reference data in the voiceprint information registration data 19 is registered, and the reliability a ₂ is calculated according to the registration elapsed time. (0 ≦ a ₂ ≦ 1) is determined. As shown in FIG. 8, when the registration elapsed time is long, it is considered that the feature difference due to the secular change may occur, so the reliability is set to be low, and when the registration elapsed time is short, the feature difference due to the secular change is low. Since it is considered to be small, the reliability is set high. Voiceprint reliability a, which is the final reliability of the voiceprint, is a
= a ₁ × a ₂ .

In step S117, the voice print information collating means 1
Reference numeral 7 extracts a feature amount required for matching from the voiceprint information acquired by the voiceprint information input means 15, and compares it with preregistered voiceprint information registration data 19 to calculate the similarity of the voiceprint information. In step S107, the composite determination unit 7 comprehensively determines whether or not the user is the original, using the reliability and the similarity of the voiceprint information.

As described above, the personal authentication method according to the eighth embodiment of the present invention determines the reliability of the voiceprint information input for each authentication, and sets the threshold value according to this reliability. By performing the above, it is possible to realize high authentication accuracy and reliability in consideration of changes in physical condition of the individual, changes over time, and differences in use environment.

The process of the personal authentication method according to the ninth embodiment of the present invention will be described below with reference to the drawings.
The personal authentication device according to the ninth embodiment is the same as the personal authentication device 400 according to the fourth embodiment, and a description thereof will be omitted. FIG. 14 is a flowchart showing the flow of processing in the personal authentication method according to the ninth embodiment of the present invention.

In step S120, the face information input means 20
Acquires the face information collected by the camera 23. In step S121, the face information reliability determination means 21 determines the reliability a (0) based on the illumination level and the registration elapsed time of the reference registration data.
≦ a ≦ 1) is determined. Hereinafter, face information reliability determination means 2
The reliability determined by 1 is called face reliability. Figure 9
FIG. 7 is an explanatory diagram of a method of determining face reliability.

The face information reliability determining means 21 is a camera 23.
By using the image collected by, the illuminance of the background image is measured by image processing, and the reliability a ₁ (0 ≦ a ₁ ≦ 1) according to the illuminance is measured.
Are determined as shown in FIG. That is, if the illumination is low or high, the contrast is difficult to obtain,
Since it is possible that the face elements are not cut out accurately, the reliability is set low, and when the illumination is in the middle, contrast is easily obtained and facial features are easily detected. Set higher.

Further, the face information reliability determining means 21 calculates the registration elapsed time based on the date and time when the reference data in the face information registration data 24 is registered, and the reliability a ₂ (0 ≦ a ₂ ≦ 1) is determined. As shown in FIG. 9, when the registration elapsed time is long, the reliability is set to be low because the feature difference due to the secular change may occur. When the registration elapsed time is short, the feature difference due to the secular change is low. Since it is considered to be small, the reliability is set high. The face reliability a, which is the final face reliability, is a = a ₁ × a
Calculate with ₂ .

In step S122, the face information collating means 22
The feature amount required for matching is extracted from the face information acquired by the face information input means 20, and the feature amount is compared with the face information registration data 24 registered in advance to calculate the degree of similarity of the face information. In step S107, the composite determination unit 7 comprehensively determines whether the person is the person using the reliability and similarity of the face information.

As described above, in the personal authentication method according to the ninth embodiment of the present invention, the reliability of the face information input for each authentication is determined, and the threshold value is set according to this reliability. By performing the above, it is possible to realize high authentication accuracy and reliability in consideration of changes in physical condition of the individual, changes over time, and differences in use environment.

The processing in the personal identification method according to the tenth embodiment of the present invention will be described below with reference to the drawings. The personal authentication device of the tenth embodiment is the same as the personal authentication device 500 of the fifth embodiment,
The description is omitted. FIG. 15 is a flow chart showing the flow of processing in the personal authentication method according to the tenth embodiment of the present invention.

In step S125, the biometric information 1 input means 25 acquires any one of biometric information such as fingerprints, voiceprints, faces, etc. necessary for authentication. Hereinafter, the acquired information will be referred to as first biometric information.

In step S126, the biometric information 1 reliability determining means 26 determines the first biometric information from the first biometric information acquired by the biometric information 1 input means 25, the use environment at the time of biometric information input, the information of reference registration data, and the like. Biometric information reliability a (0 ≦ a ≦
Determine 1). At this time, when there are a plurality of (n) evaluation items regarding the reliability of the first biometric information, the reliability a ₁ , a ₂ , ..., A _n (0
≦ a ₁ , a ₂ , ..., _An ≦ 1) is calculated, and the reliability a of the first biometric information is defined as a = a ₁ × a ₂ × ... × a _n .

In step S127, the biometric information 1 collating unit 27 extracts the characteristic amount necessary for collation from the first biometric information acquired by the biometric information 1 input unit 25, and preregistered biometric information 1 registration data 33. First to check with
The degree of similarity of the biological information of is calculated. In step S128,
The biometric information 2 input unit 4 acquires information other than the first biometric information acquired by the biometric information 1 input unit 25 among biometric information such as fingerprints, voiceprints, and faces necessary for authentication. Hereinafter, the information acquired by the biometric information 2 input means 28 will be referred to as second biometric information.

In step S129, the biometric information 2 reliability determining means 29 determines the second biometric information acquired by the biometric information 2 input means 28, the use environment at the time of biometric information input, the information of the reference registration data, and the like. Biometric information reliability b (0 ≤ b ≤ 1)
To decide. At this time, when there are a plurality of (n) evaluation items regarding the reliability of the second biometric information, the reliability b ₁ , b ₂ , ..., B _n (0 ≦ b ₁ ,
b ₂ , ..., b _n ≦ 1) is calculated, and the reliability b of the second biometric information b
_Is defined as b = b ₁ × b ₂ × ... × b _n .

In step S130, the biometric information 2 collating means 30 extracts the characteristic amount required for collation from the second biometric information acquired by the biometric information 2 input means 4, and preregisters the biometric information 2 registration data 34. And the similarity of the second biometric information is calculated. In step S131, the composite determination unit 31 comprehensively determines whether or not the person is the original person by using the reliability and similarity of the first biometric information and the reliability and similarity of the second biometric information.

In step S132, the registration update means 32
Is input by the biometric information 1 input unit 25 when the composite determination unit 31 determines that the person is the person and the reliability determined by the biometric information 1 reliability determination unit 26 is higher than the reliability at the time of registration. The relevant data stored in the biometric information 1 registration data 33 is updated with the biometric information.

As described above, the personal authentication method according to the tenth embodiment of the present invention determines the reliability of a plurality of input biometric information for each authentication and sets the threshold value according to this reliability. By setting the above, it is possible to realize high authentication accuracy and reliability in consideration of changes in physical condition of the individual, changes over time, and differences in use environment.

[0087]

As described above, according to the present invention, it is possible to realize a highly reliable personal authentication device that can cope with changes in accuracy due to changes in physical condition of the individual, secular changes, and differences in use environment.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a personal authentication device that uses reliability for personal authentication according to a first embodiment of this invention.

FIG. 2 is a diagram showing a two-dimensional similarity plane having coordinate axes of the similarity x of the first biometric information and the similarity y of the second biometric information.

FIG. 3 is a diagram showing an area determined to be a person and an area determined to be another person on a two-dimensional similarity plane.

FIG. 4 is a block diagram showing the configuration of a personal authentication device that uses the reliability of fingerprint information for personal authentication according to the second embodiment of this invention.

FIG. 5 is a block diagram showing the configuration of a personal authentication device that uses the reliability of voiceprint information for personal authentication according to the third embodiment of this invention.

FIG. 6 is a block diagram showing a configuration of a personal identification device that uses face information reliability for personal identification according to a fourth embodiment of the present invention.

FIG. 7 is an explanatory diagram of a method for determining fingerprint reliability.

FIG. 8 is an explanatory diagram of a method of determining voiceprint reliability.

FIG. 9 is an explanatory diagram of a method for determining face reliability.

FIG. 10 is a block diagram showing a configuration of a personal authentication device having a registration data registration / update function according to a fifth embodiment of the present invention.

FIG. 11 is a flowchart showing the flow of processing in a personal authentication method according to the sixth embodiment of the present invention.

FIG. 12 is a flowchart showing the flow of processing in a personal authentication method according to the seventh embodiment of the present invention.

FIG. 13 is a flowchart showing the flow of processing in the personal authentication method according to the eighth embodiment of the present invention.

FIG. 14 is a flowchart showing the flow of processing in a personal authentication method according to the ninth embodiment of the present invention.

FIG. 15 is a flowchart showing the flow of processing in a personal authentication method according to the tenth embodiment of the present invention.

[Explanation of symbols]

1, 25 biometric information 1 input means 2, 26 biometric information 1 reliability determination means 3, 27 biometric information 1 collation means 4, 28 biometric information 2 input means 5, 29 biometric information 2 reliability determination means 6, 30 biometric information 2 Collation means 7, 31 Composite determination means 8, 33 Biometric information 1 registration data 9, 34 Biometric information 2 registration data 10 Fingerprint information input means 11 Fingerprint information reliability determination means 12 Fingerprint information collation means 13 Fingerprint sensor 14 Fingerprint information registration data 15 Voiceprint information input means 16 Voiceprint information reliability determination means 17 Voiceprint information matching means 18 Microphone 19 Voiceprint information registration data 20 Face information input means 21 Face information reliability determination means 22 Face information matching means 23 Camera 24 Face information registration data 32 Registration update Means 100, 200, 300, 400, 500 Personal Authentication Device 111 Reliability of First Biometric Information <Second Biometric Information Reliability threshold straight line 112 First biometric information reliability = Second biometric information reliability threshold threshold lines 113, 114 First biometric information reliability> The straight line of the threshold when it becomes the reliability of the second biometric information

Claims (11)

[Claims] 1. Based on biometric information 1 input means for acquiring first biometric information, first biometric information acquired by the biometric information 1 input means, and input environment information of the first biometric information. Biometric information 1 reliability determining means for determining the reliability of the first biometric information, first biometric information acquired by the biometric information 1 input means, and information including the first biometric information are registered in advance. Biometric information 1 collating means for collating the registered biometric information 1 registered data, biometric information 2 input means for acquiring second biometric information, second biometric information acquired by the biometric information 2 input means, and Biometric information 2 reliability determination means for determining the reliability of the second biometric information based on the input biometric information 2 and the second biometric information acquired by the biometric information 2 input means Contains second biometric information Biometric information 2 collating means for collating biometric information 2 registration data in which information is registered in advance, reliability determined by the biometric information 1 reliability determining means, reliability determined by the biometric information 2 reliability determining means Every time,
A personal identification device comprising: a biometric information 1 collating unit and a composite discriminating unit that makes a personal identification based on the biometric information 2 collating unit. . 2. The personal authentication device according to claim 1, wherein the first biometric information and the second biometric information are different information regarding any one of a fingerprint, a voiceprint, and a face. 3. The personal information authentication device uses fingerprint information as the first biometric information or the second biometric information, and the biometric information 1 reliability determining means, to which the fingerprint information is input, or 2. The personal authentication device according to claim 1, wherein the biometric information 2 reliability determination unit determines the reliability according to the number of feature points including the branch point and the end point of the input fingerprint. 4. The biometric information 1 reliability determining means, wherein the personal identification device uses voiceprint information as the first biometric information or the second biometric information, and the voiceprint information is input. 2. The personal authentication device according to claim 1, wherein the biometric information 2 reliability determining unit determines the reliability according to the background noise of the input voiceprint and the age of the voiceprint registration data. 5. The personal authentication device uses face information as the first biometric information or the second biometric information,
The biometric information 1 reliability determining unit or the biometric information 2 reliability determining unit to which the face information is input determines the reliability according to the input face illumination level and the age of the face registration data. The personal authentication device according to claim 1, wherein 6. The biometric information 1 according to the personal authentication device, further based on the reliability determined by the biometric information 1 reliability determination means when the identity is determined by the composite determination means. The registration data is updated with the acquired first biometric information, and the biometric information 2 registration data is acquired based on the reliability determined by the biometric information 2 reliability determination means. The personal authentication device according to any one of claims 1 to 5, further comprising a registration renewal means for renewing. 7. The registration updating means updates when the reliability determined by the biometric information 1 reliability determining means is higher than the reliability of the biometric information 1 registration data, and determines the biometric information 2 reliability. 7. The personal authentication device according to claim 6, wherein the reliability is updated when the reliability determined by the means is higher than the reliability of the biometric information 2 registration data. 8. Based on the biometric information 1 input step of acquiring first biometric information, the first biometric information acquired in the biometric information 1 input step, and the input environment information of the first biometric information. Information including the biometric information 1 reliability determining step for determining the reliability of the first biometric information, the first biometric information acquired in the biometric information 1 input step, and the first biometric information is registered in advance. Biometric information 1 collation step of collating the biometric information 1 registration data, biometric information 2 input step of acquiring second biometric information, and the biometric information 2
Biometric information 2 reliability determining step of determining the reliability of the second biometric information based on the second biometric information acquired in the input step and the input environment information of the second biometric information; Biometric information 2 collating step of collating the second biometric information acquired in the 2 input step with biometric information 2 registration data in which information including the second biometric information is registered in advance, and the biometric information 1 reliability determination Is the person based on the reliability determined in the step, the reliability determined in the biometric information 2 reliability determining step, the verification result in the biometric information 1 verification step, and the verification result in the biometric information 2 verification step. A personal authentication method, comprising: a composite determination step of determining whether the person is a person. 9. A fingerprint information input step of obtaining fingerprint information, a fingerprint feature point calculation step of calculating feature points from the fingerprint information obtained in the fingerprint information input step, and the fingerprint according to the number of the feature points. A fingerprint information reliability determining step of determining the reliability of information, a fingerprint information collating step of collating the fingerprint information acquired in the fingerprint information input step with the fingerprint information registration data in which the fingerprint information is registered in advance, A voice print information input step for obtaining voice print information,
Registration for calculating the registration elapsed time of voiceprint information registration data in which the voiceprint information is registered in advance, which is compared with the background noise measurement step of measuring background noise using a microphone and the voiceprint information acquired in the voiceprint information input step. An elapsed time calculating step, a voiceprint information reliability determining step for determining the reliability of the voiceprint information based on the background noise and the registered elapsed time, and the voiceprint information and the voiceprint information acquired in the voiceprint information input step. A voiceprint information collating step for collating registered data, a reliability determined in the fingerprint information reliability determining step, a reliability determined in the voiceprint information reliability determining step, a collation result in the fingerprint information collating step, And a composite determination step of determining whether or not the user is the person based on the verification result in the voiceprint information verification step. Personal authentication method. 10. A voiceprint information input step of obtaining voiceprint information, a background noise measurement step of measuring background noise using a microphone, and the voiceprint information to be compared with the voiceprint information obtained in the voiceprint information input step, A registration elapsed time calculating step of calculating a registration elapsed time of pre-registered voiceprint information registration data, a voiceprint information reliability determining step of determining a reliability of the voiceprint information based on the background noise and the registration elapsed time, A voiceprint information collating step of collating the voiceprint information obtained in the voiceprint information input step with the voiceprint information registration data, a face information input step of obtaining face information, and an illumination degree measurement of measuring an illumination degree using a camera. Step, and the registration elapsed time of the face information registration data in which the face information is registered in advance, which is compared with the face information acquired in the face information input step. A registration elapsed time calculating step for calculating, a face information reliability determining step for determining the reliability of the face information based on the illumination degree and the registration elapsed time, and face information acquired in the face information input step. And a face information matching step of matching the face information registration data with
The reliability determined in the voiceprint information reliability determination step,
A reliability determination determined in the face information reliability determination step, a verification result in the voiceprint information verification step, and a composite determination step of determining whether or not the person is the person based on the verification result in the face information verification step. Personal authentication method characterized in that. 11. The personal authentication method further includes, when the person is judged in the person judgment in the composite judgment step, the biometric information based on the reliability determined in the biometric information 1 reliability determination step. 1 registration data is updated with the acquired first biometric information, and the biometric information 2 registration data is acquired based on the reliability determined in the biometric information 2 reliability determination step. 9. The personal authentication method according to claim 8, further comprising a registration update step of updating with information.