JP2002063576A - Device and method for personal identification - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a personal identification device capable of performing authentication processing with high accuracy even in a relatively short time even when the input condition of the ecological characteristics information on an object to be identified is not favorable. SOLUTION: This device is provided with an ecological characteristics information input device 2 for inputting the ecological characteristics information, an ecological characteristics collating device 4 for collating ecological characteristics information preliminarily registered in an ecological characteristics information database 3 with the ecological characteristics information inputted by the input device 2 and calculating the rate of concordance rate of these pieces of ecological characteristics information, and a comparison operation processing 7 of the rate of stranger acceptance, which respectively calculates the rates of concordance in regard to two or more pieces of ecological characteristics information inputted by using the input device 2 and the collating device 4, respectively calculates the rate of stranger acceptance corresponding to two or more rates of concordance by using the correspondence relation between the rate of concordance preliminarily set in a rate of stranger acceptance reference value file 5 and the rate of stranger acceptance, and judges the success/failure of authentication on the basis of the accumulated value of the calculated rate of each stranger acceptance.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a personal authentication apparatus and method for performing personal authentication using ecological characteristic information.

[0002]

2. Description of the Related Art In recent years, personal identification techniques using ecological characteristic information such as fingerprints and irises have been developed. For example,
In the conventional authentication method that uses fingerprints as ecological feature information, when multiple fingerprint information of one finger are registered and the matching rate at the time of fingerprint matching satisfies the matching rate that should be the reference There are those that make authentication successful. However, in this method, the state of the finger to be collated is poor, and if the matching rate to be a reference cannot be satisfied, authentication fails. As another method, there is a method in which fingerprint information of a plurality of fingers is registered first, and if the matching rate at the time of fingerprint matching of any of the fingers satisfies a matching rate to be a reference, authentication is successful. . However, if the condition of all the fingers to be collated is bad and even one finger cannot satisfy the matching rate that should be the reference,
Authentication fails.

An example of a conventional personal authentication method using a fingerprint is disclosed in Japanese Patent Application Laid-Open No. H10-154231, “Personal authentication apparatus and method using biological information”, Japanese Patent Application Laid-Open No. 2000-2001.
No. 13, "Principal rejection recovery method, device for executing the method, and medium recording the processing program" are described.

[0004]

In the conventional authentication system as described above, there is a problem that if the state of the finger to be verified is not good, authentication may not be performed. Also, when a plurality of fingers are used, authentication cannot be performed if the state of all fingers to be collated is poor. In addition, there is a problem that the authentication processing time becomes long because the collation processing of a plurality of fingers is required.

[0005] The present invention has been made in view of the above circumstances, and it is possible to perform the authentication process in a relatively short time and with high accuracy even when the input conditions of the biological characteristic information to be authenticated are poor. It is an object of the present invention to provide a personal authentication device and method that can perform the authentication.

[0006]

Means for Solving the Problems In order to solve the above problems, the invention according to claim 1 comprises an input means for inputting ecological characteristic information, an ecological characteristic information registered in advance, and an ecological characteristic information input by the input means. A matching unit that matches feature information and obtains a matching rate between them; and a matching ratio determined for each of two or more input ecological feature information using the input unit and the matching unit. Using the correspondence between the rate and the false acceptance rate, the false acceptance rates corresponding to the two or more coincidence rates are determined, and the success or failure of the authentication is determined based on the calculated sum of the false acceptance rates. Authentication processing means.

In the invention according to a second aspect, when the authentication processing means sets the reference value of the false acceptance rate to A1, n (n
Is an integer of 2 or more), and a value corresponding to a value obtained by multiplying the false acceptance rates Y1,..., Yn obtained with the input ecological characteristic information is obtained as a total value, and the total value is equal to or less than the reference value A1. Alternatively, it is characterized in that it is determined that the authentication is successful when it is less than.

According to a third aspect of the present invention, when the matching rate obtained by the authentication processing means with respect to the n-th ecological characteristic information is within a preset reference value of the matching rate, it is determined that the authentication is successful. It is characterized by determining.

According to a fourth aspect of the present invention, the input process of the ecological characteristic information, the previously registered ecological characteristic information, and the ecological characteristic information input in the input process are collated to determine a coincidence rate therebetween. A matching step, obtaining a matching rate for each of the two or more pieces of ecological feature information input in the inputting step and the matching step, and using the correspondence between the matching rate and the acceptance rate of others to determine the matching rate. And an authentication processing step of determining the success or failure of the authentication based on the total value of the obtained false acceptance rates.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a personal authentication system (personal authentication device) according to the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration of an embodiment of a personal authentication system according to the present invention. The personal authentication system shown in FIG. 1 includes a processing control device 1 including a CPU, a storage device, and the like and controlling each unit according to a predetermined program, an ecological feature input device 2 for inputting an ecological feature such as a fingerprint, an iris, and the like. An ecological feature information database 3 that registers a plurality of pieces of ecological feature information required for personal authentication,
An ecological feature matching device 4 for comparing the ecological feature input by the ecological feature input device 2 with the ecological feature information registered in the ecological feature information database 3 to obtain a matching rate; Other acceptance rate reference value file 5 that stores the relationship with the other acceptance rate, other acceptance rate total value file 6 that temporarily stores the other acceptance rate total value, and the matching match rate obtained by the ecological feature matching device 4 and the other person It comprises a comparison processing unit 7 for comparing the acceptance rate of other people by referring to the value in the acceptance rate reference value file 5 and the value in the total reception file 6 of others. .

In the personal authentication system of the present embodiment, when performing personal authentication, first, the ecological characteristic input device 2 is used.
Then, the ecological characteristic information is input by the ecological characteristic collating device 4, and the input ecological characteristic information is collated with the ecological characteristic information registered in the ecological characteristic information database 3. The matching rate between the statistical feature information is obtained. If the obtained matching rate reaches the set reference value, it is determined that the authentication is successful, and if it does not reach the reference value, it is determined that the authentication has failed. If the authentication fails, the user inputs further ecological characteristic information and performs collation processing. If the match rate reaches the set reference value, it is determined that the authentication is successful, and if the match rate does not reach the reference value, , A new reference value is set based on the previous matching rate, the current matching rate, and the false acceptance rate corresponding to each value, and the matching rate of the ecological characteristic information input this time or the false acceptance rate corresponding thereto is set. Based on the comparison operation, authentication success or authentication failure is determined. For example, when the ecological information is fingerprint information, the two pieces of ecological characteristic information are input by making the finger used at the first input different from the finger used at the next input. If it is still not possible to determine that the authentication has succeeded by inputting the two pieces of ecological characteristic information and collating the same, it is possible to continue the authentication processing by inputting other ecological characteristic information.

Here, the above-mentioned false acceptance rate will be described. When measuring the authentication accuracy in the personal authentication, as the value representing the authentication accuracy, there are an identification accuracy for recognizing the person as a registered person and an identification accuracy for recognizing the person as being distinguished from others. If an attempt is made to increase the identification accuracy, the ratio of recognizing a person as a person but not a person, that is, a rejection rate of the person, increases. On the other hand, when the identification accuracy is low, the false acceptance rate indicating the error rate for recognizing the stranger as the true person increases. Therefore, when evaluating the authentication accuracy, the false rejection rate and the false acceptance rate are important evaluation factors.

FIG. 2 shows an authentication measurement result of a certain personal authentication system. The authentication measurement result shown in FIG. 2 shows an example of the relationship between the matching rate at the time of matching, the false acceptance rate, and the false rejection rate. From FIG. 2, it is necessary to set the matching rate of the verified ecological characteristic information to 90% or more in order to achieve the false acceptance rate of 1% or less in consideration of security. However, in this case, the false rejection rate becomes as high as 27% or more, and the number of persons who are not authenticated increases. Therefore, in this embodiment, by performing a matching process using two or more pieces of ecological characteristic information, a total value of the false acceptance rate is obtained in a stepwise manner.
The authentication is determined based on the total value. That is, in the “stepwise personal authentication method using a plurality of matching coincidence rates” adopted in the present embodiment, when two or more pieces of ecological feature information are used, the coincidence rate of each information is not determined individually. Calculate the corresponding false acceptance rate from the coincidence rate of two or more ecological characteristic information, sum up the calculated false acceptance rate, and calculate the sum of the false acceptance rate. Is satisfied, the authentication is determined to be successful.

In the present embodiment, the relationship between the coincidence rate and the false acceptance rate as shown in FIG. 2 is extracted based on the measured values, stored as a false acceptance rate reference value file 5, and based on the value. Authentication processing. When the false acceptance rate is extracted, in the authentication processing system that actually performs personal authentication, the ecological features input by the ecological feature input device 2 and all the ecological features registered in the ecological feature information database 3 A table showing the relationship between the match rate of the individual and the false acceptance rate based on the match rate obtained at each matching, and register this as the false acceptance rate reference value file 5. I have to.

In the present embodiment, if the matching rate at the first matching is less than the reference rate, matching with another ecological feature is continuously performed. At that time, both match rates and the corresponding false accept rate are used, so even if the match rate for the second and subsequent verifications is reduced gradually, the false accept rate that should serve as a reference can be secured. it can.

Next, the operation of the personal authentication system shown in FIG. 1 will be described with reference to FIG. FIG. 3 is a flowchart showing the operation of the personal authentication system shown in FIG. 1, and corresponds to the configuration of a program executed by the processing control device 1 or the like.

First, prior to the execution of the authentication process, a correspondence value between the matching rate of the ecological feature information and the acceptance rate of others by an identity authentication system that performs matching based on ecological features, and a reference value ( Target value). This is stored in a predetermined storage area in the processing control device 1, for example.

Next, execution of the authentication process is started. First, the stored false acceptance rate reference value is acquired (step 30).
1) Display an ecological feature input request on the ecological feature input device 2 or another predetermined display device (step 30).
2). Here, it is determined whether or not the input of the first ecological feature has been performed (step 303). If the input of the first ecological feature has been performed, the first ecological feature is registered with the first ecological feature. The matching is performed based on the ecological characteristic information, and the matching rate is obtained (step 304). On the other hand, if the input of the first ecological feature is canceled (step 303),
The authentication result indicating the authentication failure is output (step 31).
2), end the process.

Next, the false accept rate corresponding to the matching rate obtained in step 3 is extracted based on the data stored in the false accept rate reference value file 5 (step 305). A comparison is made with the acceptance rate reference value (target value) (step 306). If this value is equal to or less than the reference value, authentication is successful with the first ecological feature, and an authentication result indicating that authentication is successful is output (step 312).
On the other hand, other than that, the authentication fails ("out of standard" in step 306).

In the case of an authentication failure, it is determined whether or not the total value of the false acceptance rate has already been set (step 3).
07), in this case, since it has not been set yet, the first
After storing the false acceptance rate corresponding to the matching rate determined by the ecological characteristics of the above in the area determined, that is, a predetermined area in the false acceptance rate totaling file 6 (step 311), the second
The matching rate is determined by the ecological characteristics of (steps 302 to 304).

The false acceptance rate corresponding to the matching rate obtained in step (305) is extracted (step 305). If this value is equal to or smaller than the reference value, the authentication is successful with the second ecological feature (step 306: “within reference”). -Step 312). If the false acceptance rate corresponding to the matching rate obtained in step S1 exceeds the reference value ("out-of-standard" in step 306), the false acceptance rate total value check (step 307) and the false accept rate total file 6 Acquisition of the total value of the false acceptance rate (step 308) and recounting of the false acceptance rate for totaling the current value (step 30)
9) and a comparison of the false acceptance rate reference value for comparing the total value with the reference value (step 310), and is stored in the area determined as the false acceptance rate determined for the second ecological feature. If the total value, which is the value obtained by summing the received false acceptance rate, is equal to or smaller than the reference value (“within the reference” in step 310), the authentication is successful with the second ecological feature (step 312). As described above, the success or failure of the authentication is determined by comparing the aggregated value of the false acceptance rate with the reference value. Therefore, the verification using the first ecological feature or the verification using the second ecological feature indicates that the authentication has failed. Anyone can authenticate in a short time while ensuring sufficient security. On the other hand, if the total value of the false acceptance rate exceeds the reference value, the authentication fails ("out of reference" in step 310).

In the case of an authentication failure, the total value of the false acceptance rate calculated in step 3 is stored in the determined area (step 31).
1), collation is continuously performed based on the third ecological feature to obtain a coincidence rate.

By repeating the above operations,
It is possible to accommodate more than two ecological features.

Next, the contents of the authentication process in each of the above processes will be described. In the following description, the variable A represents the target false acceptance rate and the variable B represents the coincidence rate of the ecological feature information obtained from the characteristics shown in FIG. 2 corresponding to the target false acceptance rate A. , Variables X and Y respectively represent the coincidence rate obtained from the input ecological characteristics and the corresponding false acceptance rate.

Set a target value in consideration of security. ……… Target acceptance rate: A1 (A1 <1)
And Further, it is assumed that the coincidence rate of the ecological characteristic information at the false acceptance rate = A1 is B1 (B1 <1).

Matching is performed according to a first ecological feature. ... Matching rate of the ecological feature information collated by the first ecological feature: X1 (X1 <1), and matching rate = X
1 is determined as Y1 (Y1 <1).

In the case where the false acceptance rate Y1 ≦ the false acceptance rate A1, which is the target, the authentication is successful with the first ecological feature, but the authentication fails otherwise. If the authentication fails, the matching is continuously performed according to the second ecological feature. Match rate of ecological characteristic information collated by the second ecological characteristic: X
2 (X2 <1), and it is assumed that the coincidence rate = the false acceptance rate at the time X2: Y2 (Y2 <1).

In order to achieve the target false acceptance rate: A1 in the authentication accuracy, it is necessary to achieve the target false acceptance rate: A2 = A1 / Y1 in the second ecological feature matching. That is, the matching rate of the ecological feature information collated by the second ecological feature: the accepting rate of others at X2: Y2
Satisfies the condition of Y2 ≦ A2. Therefore, Y2 ≦ A1 / Y1 to Y2 * Y1 ≦ A1 are the conditions for achieving successful authentication. In the above description, the comparison may be performed with a matching rate that satisfies the false acceptance rate reference value. In other words, in the case where the matching rate X1 of the ecological feature information ≧ the matching rate B1 of the ecological feature information, the authentication is determined to be successful in the first ecological feature, and the matching rate of the fingerprint information in the false acceptance rate: A2 : The coincidence rate between B2 and the ecological feature information collated by the second ecological feature: X2 is X2 ≧
The authentication may be successful if the condition of B2 is satisfied.

Next, the operation of the present embodiment will be specifically described using the measurement result of FIG. 2 as the authentication result by the fingerprint.

Set a target value in consideration of security. ... A target false acceptance rate: A1 = 0.01. In this case, the fingerprint information coincidence rate at the false acceptance rate = 0.01 is 0.9.

The collation with the first finger is performed. ............
Assume that the coincidence rate of the fingerprint information collated by the first finger is 0.8. In this case, from FIG. 2, the false acceptance rate at the coincidence rate = 0.8 is 0.1.

Another person acceptance rate of collation with the first finger:
Since 0.1> the target false acceptance rate of 0.01 is satisfied, the first finger fails in authentication. In the case of an authentication failure, collation is continuously performed with the second finger. Here, it is assumed that the matching rate of the fingerprint information collated by the second finger is 0.8. In this case, from FIG. 2, the false acceptance rate at the coincidence rate = 0.8 is 0.1.

In order to achieve the target false acceptance rate in the authentication accuracy, the false false acceptance rate targeted in the second ecological feature matching: A2 (0.1) = 0.01 / 0.1, Match rate of fingerprint information collated by the second finger: X
2 acceptance ratio: Y2 is Y2 ≦ A2 (0.1)
Is satisfied. Therefore, Y2 ≦ 0.01 /
From 0.1 to Y2 * 0.1 ≦ 0.01 is a condition for achieving authentication success. From FIG. 2, the coincidence rate of fingerprint information collated with the second finger: X2 = 0.8, the false acceptance rate: Y2 =
0.1, which satisfies the above-mentioned achievement condition, and thus the authentication is successful. In the above and in the above, since the comparison may be performed with a matching rate that satisfies the false acceptance rate reference value,
In, since the matching rate of the collation by the first finger: 0.8 <the matching rate of the target value: 0.9, the first finger is determined to have failed in the authentication, and the false acceptance rate in the fingerprint of A2 (0.1) Information matching rate: B2 (0.8) and the matching rate of the fingerprint information collated with the second finger: 0.8 is 0.8 ≧ B2
Since the condition of (0.8) is satisfied, the authentication may be successful.

As a value representing the indirect effect in this case, the personal rejection rate at this point is the personal rejection rate: 0.1 when the coincidence rate of the fingerprint information collated with the first finger = 0.8, The rejection rate at the coincidence rate of the fingerprint information collated by the second finger = 0.8 is 0.1, and the probability that the first and second fingers are not authenticated is 0.1 * 0.1 =
It becomes 0.01. The result is the target false acceptance rate:
This indicates that fingerprint authentication failure is significantly reduced while securing 0.01. Therefore, in the present embodiment, by using a plurality of fingers, it is possible to greatly reduce fingerprint authentication failures while securing a target false acceptance rate.

The embodiment of the present invention is not limited to the above-described embodiment. For example, each device in the system configuration shown in FIG. 1 may be integrated or separated, or each file may be integrated or further separated and held. And can be changed as appropriate. 1 can be composed of a computer and a program executed by the computer, and the program can be distributed via a computer-readable recording medium or a communication line.

[0036]

As described above, according to the present invention,
Even if the input condition of the ecological characteristic information to be authenticated deteriorates, it is possible to perform the authentication process more reliably and in a shorter time than before. For example, in the case of fingerprints, even if the condition of the finger to be compared is poor and all the registered fingers do not satisfy the matching rate that should be a reference, two or more fingerprints can be compared for reliable authentication in a short time. It is possible to do.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a personal authentication system according to the present invention.

FIG. 2 is a diagram showing an example of an authentication system measurement result for explaining the embodiment of FIG. 1;

FIG. 3 is a flowchart showing the operation of the personal authentication system of FIG. 1;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Processing control device 2 Ecological feature input device 3 Ecological feature information database 4 Ecological feature collation device 5 Others acceptance rate reference value file 6 Others acceptance rate total value file 7 Others acceptance rate comparison arithmetic processing unit

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Rumiko Matsuzaki, Inventor Ritsuko Matsuzaki 3-9-1-2 Nishi-Shinjuku, Shinjuku-ku, Tokyo East Japan Nippon Telegraph and Telephone Corporation (72) Inventor Kazunari Watanabe 3-chome, Nishishinjuku, Shinjuku-ku, Tokyo 19-2 Tohoku Nippon Telegraph and Telephone Corporation (72) Inventor Shigeo Shimada 3-chome, Nishishinjuku 3-chome, Shinjuku-ku, Tokyo (72) Inventor Tetsuya Iwasaki Shinjuku-ku, Tokyo 3-19-2 Nishi-Shinjuku Tohoku Nippon Telegraph and Telephone Co., Ltd. (72) Inventor Kazuki Hirata 3-19-2 Nishi-Shinjuku, Shinjuku-ku, Tokyo Tohoku Nippon Telegraph and Telephone Co., Ltd. F-term (reference) 4C038 FF01 FF05 VA07 VB04 5B043 AA09 BA01 GA01 GA13

Claims (4)

[Claims] An input unit for inputting ecological characteristic information; a collation unit for collating the ecological characteristic information registered in advance with the ecological characteristic information input by the input unit to obtain a coincidence rate between them; Means and matching means are used to determine a match rate for each of the input two or more ecological characteristic information, and the match rate between the two or more is determined using the correspondence between a preset match rate and a false acceptance rate. Find the false acceptance rate corresponding to each,
An authentication apparatus comprising: authentication processing means for determining whether authentication has succeeded or failed based on the calculated total value of the false acceptance rates. 2. The method according to claim 1, wherein the authentication processing unit determines that the reference value of the false acceptance rate is A1, and the false acceptance rate Y1 determined for n (n is an integer of 2 or more) input ecological characteristic information. ,
.., Yn are multiplied by each other to obtain a total value, and when the total value is equal to or less than the reference value A1, it is determined that the authentication is successful. Authentication device. 3. The method according to claim 2, wherein the authentication processing means determines that the authentication is successful when the matching rate obtained for the n-th ecological characteristic information is within a reference value of the matching rate set in advance. The personal authentication device according to claim 1 or 2, wherein 4. A process of inputting ecological characteristic information, collating the preregistered ecological characteristic information with the ecological characteristic information input in the inputting process, and obtaining a matching rate thereof, A match rate is obtained for each of the two or more pieces of ecological characteristic information input in the process and the matching process, and the correspondence rate between the two or more pieces is determined using a correspondence relationship between a preset match rate and a false acceptance rate. An authentication processing step of determining each of the false acceptance rates and determining a success or failure of the authentication based on a total value of the determined false acceptance rates.