JP2005293139A - Personal checking system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a biometric checking system excellent in operability and/or convenience and capable of saving the legitimate person to be checked even in the case of one-to-N checking where checking must be done on the basis of a strict standard value so as not to degrade security. <P>SOLUTION: The biometric checking system includes a biometric information acquisition part 1 for acquiring the biometric information of a person to be checked; a storage part 8 for storing individual reference values used in the determination of registrants' biometric information and a common standard value used in the determination of biometric information on a stricter standard than the individual reference values; a checking part 6 for calculating the similarity between the pieces of biometric information stored in the biometric information acquiring part 1 and the storage part 8; a determining part 9 for determining that when the similarity calculated by the identifying part 6 meets any of the reference values, the person to be checked is the registrant; a personal checking information input part 2 for letting the person to be checked input his or her personal identification information; and an output part 10 for outputting the result of the determination. The determining part uses the common standard value in the determination; when the person to be checked is determined not to be the registrant, the individual reference value that corresponds to his or her personal identification information is used on condition that the personal identification information has been input. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a collation device, and more particularly to a personal collation device that authenticates an individual by sequentially collating input biometric information and a plurality of registered biometric information.

  Conventionally, there has been an apparatus for authenticating an individual using biometric information such as a fingerprint, a voice print, a palm print, and a retina. As a specific authentication method, a person to be verified inputs personal identification information such as an ID number from a numeric keypad, calls registered biometric information corresponding to the ID number, and then inputs biometric information between the input and biometric information. There is a method of collation (hereinafter referred to as “one-to-one collation”).

  Biometric information, unlike keys, magnetic cards, and IC cards, has the advantage that it is less likely to be lost or illegally used. Due to environmental conditions such as temperature and humidity at the time of capture, there were cases where stable input was not possible. Therefore, for example, reference values of different levels of high and low are set, and the matching method is changed depending on where the one-to-one matching result corresponds to different reference values (see Patent Document 1). .

  As another authentication method using biometric information, for example, a method of sequentially collating one input biometric information and a plurality of registered biometric information (number N) (hereinafter referred to as “one-to-N collation”). (See Patent Document 2).

  Compared to 1-to-1 verification, 1-to-N verification does not require identification of biometric information to be verified from a plurality of registered biological information, and the person to be verified can omit the numeric keypad input. There is an advantage of high operability and convenience. In recent years, with the increase in CPU speed and the like, many biometric authentication devices using 1-to-N matching have been adopted.

JP-A-8-115422 JP-A-4-177478

  However, the one-to-one collation in the prior art requires the input of personal identification information such as a personal identification number, which is not convenient. On the other hand, the 1-to-N collation does not require the input of personal identification information and is more convenient than the one-to-one collation. There was a problem that it was inferior in safety. Therefore, in order to improve safety in the 1-to-N collation, it is necessary to collate with a strict reference value. There is a problem that the rejection rate of the person is further increased than when the verification is performed.

  Therefore, an object of the present invention is to remedy a legitimate person to be verified even in the case of 1-to-N verification, in which verification must be performed with strict reference values so as not to reduce safety, and Another object of the present invention is to provide a biometric information collating apparatus that is excellent in operability and convenience.

  In view of the above-described object, an object of the present invention is a personal verification device that sequentially collates biometric information acquired from a person to be collated with a plurality of registered biometric information, and personally collates the biometric information of the person to be collated. A biometric information acquisition unit for acquiring biometric information of a plurality of registrants and individual identification values of the plurality of registrants based on input of individual identification information of the plurality of registrants; A common reference value used in common with registrant's biometric information and applied to the determination of individual biometric information with a stricter criterion than the individual reference value, and a storage unit that stores the common reference value A matching unit that calculates the degree of similarity between the biological information acquired and the biological information acquired by the biological information acquisition unit, and the similarity calculated by the verification unit matches a reference value applied to the determination Determination unit that determines that the person to be verified is a registrant A personal identification information input unit that is operated by a person to be collated to input personal identification information, and an output unit that outputs a determination result, wherein the determination unit applies the common reference value to the determination, Provided is a personal verification device that applies the individual reference value corresponding to the personal identification information on condition that the personal identification information is input when it is determined that the person to be verified is not a registrant. There is.

According to the present invention, in one-to-N matching, matching is performed with a strict common reference value. G. In this case, the person to be collated inputs personal identification information and collates with a more permissive individual reference value. As a result, for the majority of people who can perform normal collation, it is possible to rescue a collated person whose biometric information input is bad due to scratches or the like without impairing the convenience of 1-to-N collation. Further, since the individual reference value corresponds to the personal identification information, it is possible to perform collation with an appropriate individual reference value corresponding to the state of the person's biometric information. In this case, since a tolerable reference value is used after the personal identification information is input, the identity rejection rate can be kept low while maintaining safety, and both safety and convenience can be achieved.
Furthermore, since the information corresponding to the personal identification information is matched not only with the individual reference value but also with the stored biological information, only the corresponding biological information is collated with the corresponding individual reference value. Sexual compatibility can be further improved.
Further, when collation is performed using the individual reference value, re-collation is performed using the biometric information input at the time of the first 1-to-N collation, so that the convenience of the collated person is improved. Therefore, when viewed as a whole, convenience is remarkably improved and safety is ensured.

  Hereinafter, a personal verification apparatus according to the present invention using a fingerprint as biometric information will be described in detail with reference to FIGS.

FIG. 1 is a block diagram showing a basic configuration of a personal verification apparatus according to the present invention.
The apparatus includes a fingerprint input unit 1 that reads a fingerprint input by a person to be verified with a camera, a personal identification information input unit 2 having a numeric keypad for inputting personal identification information such as a personal identification number and an ID code, a verification result, and an operation Various guidance data such as procedures and information are displayed as guidance. For example, the display unit 3 such as an LCD, and external devices such as a display unit and an electric lock based on inputs from the fingerprint input unit 1 and the personal identification information input unit 2 And a control unit 4 for outputting a signal.

  Further, the control unit 4 extracts from the fingerprint input from the fingerprint input unit 1 a feature parameter extraction unit 5 that extracts a fingerprint feature parameter that can be identified as the principal, such as a ridge direction branch point of the fingerprint, and the feature parameter extraction unit 5 Input from the personal identification information input unit 2 and the collation unit 6 that calculates the similarity between the fingerprint feature parameters that are input and temporarily stored in the storage unit 8 described later and the fingerprint feature parameters stored in advance in the storage unit 8 described later Personal identification information collating unit 7 for searching whether the personal identification information such as the entered personal identification number or ID code is in storage unit 8 to be described later, the registered fingerprint characteristic parameters, common reference value, individual standard value, personal identification number , A storage unit 8 that stores an ID code, a determination unit 9 that compares the similarity calculated by the collation unit 6 with a reference value, and an external device such as an electric lock based on a determination result from the determination unit 9 Trust Composed of the output unit 10 for outputting.

FIG. 2 is a flowchart showing a basic operation flow example of the personal collation apparatus according to the present invention, and FIG. 3 is a diagram showing an example of a table stored in the storage unit 8 of the apparatus.
As shown in FIG. 3, the table of the storage unit 8 stores the ID code of the person to be verified, the personal identification number, the individual reference value having a tolerance value, and the feature parameter in association with each other, and further has a strict value. A common reference value is stored. However, in all examples described herein, the tolerance value has a value greater than the strict value.

  Further, since the password is intended to rescue the person to be verified who is difficult to read the fingerprint, it is not necessary to store it for all the persons to be verified. Individual reference values are given to each person to be verified in which a personal identification number is stored, but a more permissive value is given to those who are difficult to read a fingerprint. In addition, the common reference value used in the 1-to-N collation is generally set to a strict value in order to reduce the other-person tolerance rate.

In the example of FIG. 3, the ID code 01 is associated with the PIN number 9265, the individual reference value 5 and the feature parameter data 1 respectively, and the common reference value 2 is common to all ID codes. Is associated with. The ID code 02 is not associated with a personal identification number or an individual reference value.
Further, the common reference value 2 is set smaller than the individual reference values 5, 4, 3, and the like, but in this embodiment, the common reference value is stricter than the individual reference value. Means value.
Further, the individual reference value associated with ID code 01 is 5, and the individual reference value associated with ID code 03 is 4, but this is for those who have ID code 01. Is more difficult to read a fingerprint than a person having an ID code 03, and is intended to give a more permissive individual reference value (that is, a larger value) to the former.

Next, a basic operation flow of the personal verification device according to the present invention will be described with reference to FIG.
When the person to be verified places his / her finger on the fingerprint input unit 1, the fingerprint read here is sent to the feature parameter extraction unit 5, where the feature parameter is extracted and temporarily stored in the storage unit 8 (S01). ).

  Next, the similarity between the feature parameters input / temporarily stored in step S01 and the feature parameters (data 1 to data N) stored in advance in the storage unit 8 as shown in FIG. The collation unit 6 sequentially calculates by a known method such as an analysis method, and the calculated similarity is compared with a common reference value (see FIG. 3) that is a strict value (S02).

  As a result of the comparison, when the similarity matches the common reference value (in the embodiment of the present specification, the similarity is a value smaller than the common reference value), the collation O.D. K. (Yes in S03), an unlock signal is output from the output unit 10 to the electric lock, and the ID of the person to be verified is left in the history (S07). As a result of the comparison, if the similarity does not match the common reference value (in the embodiment of the present specification, the similarity is a value equal to or greater than the common reference value), the process proceeds to step S04. The above is the one-to-N matching process.

  In step S04, the person to be verified selects whether to repeat steps S01 to S03 a predetermined number of times or to input personal identification information such as an ID code and a personal identification number from the personal identification information input unit 2 (S04). When the person to be verified selects the input of personal identification information, the process proceeds to the following one-to-one verification process. That is, the feature parameter temporarily stored in the storage unit 8 in step S01 and the feature parameter in FIG. 3 of the person corresponding to the personal identification information input in step S04 (for example, the PIN code 9265 is input as the personal identification information) In this case, the similarity is calculated by the matching unit 6 using a known method such as a pattern matching method, a minutiar method, and a frequency analysis method, and the calculated similarity is input. The determination unit 9 determines that the individual reference value is a permissive value corresponding to the personal identification information (for example, if the personal identification number 9265 is input as the personal identification information, the individual reference value is 5). Compare (S05).

  As a result of the comparison, if the similarity matches the individual reference value in FIG. 3 (in the embodiment of this specification, the similarity is a value smaller than the individual reference value in FIG. K. (Yes in S06), an unlock signal is output from the output unit 10 to the electric lock, the ID of the person to be verified is left in the history (S07), and the verification process is terminated. As a result of the comparison, if the similarity does not match the individual reference value in FIG. 3 (in the embodiment of the present specification, the similarity is a value equal to or greater than the individual reference value in FIG. 3), the matching process is terminated.

4 to 7 are flowcharts showing specific operation flow examples of the personal collation apparatus according to the first embodiment of the present invention.
First, when the person to be verified places a finger on the fingerprint input unit 1 (S11), the display unit is turned on (S12), and the number of finger placements i is set to 1 (S13). Next, the fingerprint read by the fingerprint input unit 1 is sent to the feature parameter extraction unit 5, where the feature parameter is extracted and temporarily stored in the memory P in the storage unit 8 (S14).

  Next, in the loops S15 to S18, the similarity between the memory P and the feature parameters (data 1 to data N) stored in advance in the table of the storage unit 8 as shown in FIG. By a known method such as a method, the collation unit 6 sequentially calculates and temporarily stores it in the memory S in the storage unit 8 (S16), and determines the memory S and a common reference value (see FIG. 3) that is a strict value. 9 is compared (S17).

  As a result of the comparison, if the memory S conforms to the common reference value (in the embodiment of this specification, the memory S is a value smaller than the common reference value), the process leaves the loop and the collation O.D. K. The process proceeds to O (S46). If the loop is terminated without exiting the loop, the number of finger placements i is compared with a predetermined number of times (S19).

  As a result of the comparison, if the number of finger placements i is greater than or equal to the predetermined number, collation N.I. G. The process proceeds to process N (S48). As a result of the comparison, if the number of finger placements i is less than the predetermined number, “Please place your finger again or enter your PIN” is displayed on the display unit 3 (S20). The above is the one-to-N matching process.

  Subsequently, the process proceeds to the process after B (S21). It is determined whether or not there is another finger placement (S21). If there is a finger placement, a timer (not shown) is reset (S26), the value of the number of finger placement i is increased by 1 (S27), and again. In order to repeat the processing from step S14 onward, the process proceeds to A (S14). If there is no finger placement, it is determined whether or not the timer is activated (S22). If the timer is not activated, the timer is activated (S23).

  Next, it is determined whether or not personal identification information such as an ID code and a personal identification number is input from the person to be verified through the personal identification information input unit 2 (S24). If there is personal identification information input, the timer is reset. (S28), the process proceeds to step S30. If there is no personal identification information input, the timer is compared with a predetermined time (S25).

  As a result of the comparison, if the timer exceeds the predetermined time, the timer is reset (S29). G. The process proceeds to process N (S48). As a result of the comparison, when the timer is equal to or shorter than the predetermined time, the processing from step S21 is repeated. Through the loops S21 to S25, the finger placement or personal identification information input from the person to be verified is monitored within a predetermined time.

  In step S30 and subsequent steps, a one-to-one matching process is performed. First, in loops S30 to S32, the personal identification information matching unit 7 searches the table (FIG. 3) in the storage unit 8, and obtains the ID code j in FIG. 3 of the person corresponding to the personal identification information input in step S24. Identify. In the determination of step S31 in this loop, if the personal identification information input in step S24 matches the personal identification information of the jth person, the loop exits and sets the number of finger placements i to 1 (S33), The process proceeds to C and subsequent processes (S35). At this point, the value of j is determined. If the loop is terminated without exiting from this loop, the verification N.P. G. The process proceeds to process N (S48).

  Below C, first, the similarity between the memory P (see S14) and the characteristic parameter (data j) in FIG. 3 is calculated by the matching unit 6 by a known method such as a pattern matching method, a minutia method, or a frequency analysis method. Then, the data is temporarily stored in the memory S in the storage unit 8 (S35), and the determination unit 9 compares the memory S with the j-th individual reference value (see FIG. 3) which is a permissive value (S36).

  As a result of the comparison, when the memory S conforms to the individual reference value (in the embodiment of the present specification, the memory S is smaller than the individual reference value), the collation O.D. K. The process proceeds to O (S46). As a result of the comparison, when the memory S does not conform to the individual reference value (in the embodiment of this specification, the memory S is a value equal to or greater than the individual reference value), the number of finger placements i is compared with the predetermined number of times. (S37). As a result of the comparison, if the number of finger placements i is greater than or equal to the predetermined number, collation N.I. G. The process proceeds to process N (S48). As a result of the comparison, if the number of finger placements i is less than the predetermined number, “Please place your finger again” is displayed on the display unit 3 to the person to be verified (S38).

  Subsequently, it is determined whether or not there is another finger placement (S39). If there is a finger placement, a timer (not shown) is reset (S43), and the value of the number of finger placements i is increased by 1 (S44). ), The process proceeds to step S34. In step S34, the fingerprint read by the fingerprint input unit 1 is sent to the feature parameter extraction unit 5, where the feature parameter is extracted, which is temporarily stored in the memory P in the storage unit 8, and subsequently C or less. Repeat the process.

  If there is no finger placement in step S39, it is determined whether or not the timer is activated (S40). If the timer is not activated, the timer is activated (S41). Next, the timer is compared with a predetermined time (S42).

  As a result of the comparison, if the timer exceeds the predetermined time, the timer is reset (S45). G. The process proceeds to process N (S48). As a result of the comparison, when the timer is equal to or shorter than the predetermined time, the processing from step S39 is repeated. Through the loops S39 to S42, the finger placement again from the person to be verified is monitored within a predetermined time.

  Verification O. K. In the process O, an unlock signal is output from the output unit 10 to the electric lock (S46), the ID and time of the person to be verified are left in the history (S47), and the process proceeds to step S49. Verification N. G. In the process N, “cannot collate” is displayed on the display unit 3 (S48), and the process proceeds to step S49. In step S49, the display unit 3 is turned off, and the collation process is terminated.

8 and 9 show a second embodiment of the present invention.
In this embodiment, step S05 in the basic flowchart (FIG. 2) of the present invention is replaced with step S05 ′ in FIG. That is, in this step, the similarity between the feature parameters input / temporarily stored in step S01 and the feature parameters (data 1 to data N) stored in advance in the storage unit 8 as shown in FIG. Individual verification values of a person (jth person) corresponding to the calculated similarity and the input personal identification information are sequentially calculated by a matching unit 6 by a known method such as a minutia method or a frequency analysis method (see FIG. 9). ) Is compared by the determination unit 9.

This individual reference value is verified by a legitimate person to be verified. G. In general, it is set to be more forgiving than the common reference value. For this reason, the tolerance rate for others tends to increase. Therefore, it is desirable to give a personal identification number only to limited persons who are particularly difficult to read fingerprints and need special relief.
Unlike the first embodiment, the second embodiment consistently performs 1-to-N matching. Therefore, FIG. 9 does not show individual feature parameters corresponding to each person to be verified, and it is clarified that only the criterion value for a specific individual is switched.

10 and 11 show a third embodiment of the present invention.
In this embodiment, step S05 in the flowchart (FIG. 2) for explaining the first embodiment is replaced with step S05 ″ in FIG. That is, in this step, the feature parameter temporarily stored in the storage unit 8 in step S01 and one or more feature parameters in FIG. 11 of the person (jth person) corresponding to the personal identification information input in step S04. The similarity is sequentially calculated by the matching unit 6 by a known method such as a pattern matching method, a maneuver method, a frequency analysis method, and the like, and the individual reference value of the j th person that is the tolerance and the calculated similarity (FIG. 11). Are sequentially compared by the determination unit 9.

The third embodiment is different from the first embodiment in that a plurality of data can be used as the j-th feature parameter. This is intended to more actively relieve a person who cannot use a specific finger for collation due to injury or the like. For example, in the example of FIG. 11, when the right index finger of a person having the ID code 01 (registrant A) cannot be used for collation due to injury or the like, the middle right finger can be used.
Further, if all the fingers of the registrant are registered in correspondence with the personal identification information, the person to be verified can collate without being aware of which finger is registered, and the convenience is further improved.

  In each of the above embodiments, the fingerprint is used as the biometric information. However, the biometric information may be information such as a voice print, a palm print, a retina, and a face image that can identify an individual. Further, the personal identification information input in step S04 of FIG. 2 and step S24 of FIG. 5 may be input by a magnetic card, IC card or the like in addition to the input by the numeric keypad of the personal identification information input unit 2.

It is a block diagram which shows the structure of the personal collation apparatus which concerns on this invention. It is a flowchart which shows the example of a basic operation | movement flow of the apparatus. It is a figure which shows an example of the table memorize | stored in the memory | storage part of the same apparatus. It is a flowchart (1) which shows the example of a specific operation | movement flow of a 1st Example. It is a flowchart (2) which shows the example of a specific operation | movement flow of a 1st Example. It is a flowchart (3) which shows the example of a specific operation | movement flow of a 1st Example. It is a flowchart (4) which shows the example of a specific operation | movement flow of a 1st Example. It is a part of flowchart which shows the example of a basic operation | movement flow of the personal collation apparatus which concerns on a 2nd Example. It is a figure which shows an example of the table memorize | stored in the memory | storage part of the same apparatus. It is a part of flowchart which shows the example of a basic operation | movement flow of the personal collation apparatus which concerns on a 3rd Example. It is a figure which shows an example of the table memorize | stored in the memory | storage part of the same apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Fingerprint input part 2 ... Personal identification information input part 3 ... Display part 4 ... Control part 5 ... Feature parameter extraction part 6 ... Collation part 7 ... Personal identification information collation part 8 ... Memory | storage part 9 ... Determination part 10 ... Output part

Claims (5)

A personal verification device for sequentially collating biometric information acquired from a person to be collated with a plurality of registered biometric information,
A biometric information acquisition unit that acquires biometric information of the person to be verified;
Common to both the individual reference value applied to the determination of the individual biometric information of the plurality of registrants on the condition that the biometric information of the plurality of registrants and the personal identification information of the plurality of registrants are input, and the biometric information of the plurality of registrants A storage unit that stores a common reference value that is used and applied to the determination of individual biological information based on a criterion that is stricter than the individual reference value;
A collation unit that calculates the similarity between the biological information stored in the storage unit and the biological information acquired by the biological information acquisition unit;
A determination unit that determines that the person to be verified is a registrant when the similarity calculated by the verification unit matches a reference value applied to the determination;
A personal identification information input unit that is operated by a person to be verified and inputs personal identification information;
An output unit for outputting the determination result;
Have
The determination unit applies the common reference value to the determination, and when it is determined that the person to be verified is not a registrant, the individual reference corresponding to the personal identification information on condition that the personal identification information is input A personal verification device characterized by applying a value. The storage unit stores the personal identification information and the biological information in association with each other,
The collation unit collates only with biometric information corresponding to the personal identification information input when determining using the individual reference value,
The personal verification device according to claim 1. The storage unit temporarily stores the biological information acquired by the biological information acquisition unit,
The collation unit, when collating with the individual reference value, compares the biometric information registered using the biometric information temporarily stored in the storage unit,
The personal verification device according to claim 1 or 2.   The personal verification device according to claim 1, wherein the biometric information is a fingerprint. A personal verification method for sequentially verifying biometric information acquired from a person to be verified with a plurality of registered biometric information,
Common to both the individual reference value applied to the determination of the individual biometric information of the plurality of registrants on the condition that the biometric information of the plurality of registrants and the personal identification information of the plurality of registrants are input, and the biometric information of the plurality of registrants A storage step of storing a common reference value that is used and applied to the determination of individual biological information with a determination criterion that is stricter than the individual reference value;
A biometric information acquisition step of acquiring biometric information of the person to be verified;
A collation step of calculating the similarity between the biological information stored in the storage step and the biological information acquired in the biological information acquisition step;
A first determination step of determining that the person to be verified is a registrant when the similarity calculated in the verification step matches the common reference value;
A personal identification information input step in which the person to be verified operates to input personal identification information when the similarity calculated in the verification step does not match the common reference value;
A second determination step of determining that the person to be verified is a registrant when the similarity calculated in the verification step matches the individual reference value on condition that the personal identification information is input;
An output step of outputting a determination result when it is determined that the person to be verified is a registrant;
A personal verification method characterized by comprising:

Images