JPH11110541A - Individual discrimination device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an individual discrimination device which can discriminate an individual with only personal information without the nud for designating an ID number or deteriorating identification precision. SOLUTION: Plural pieces of identification data 7b and 7c are registered in an individual data base 7, in accordance with the ID number 7a. When a person to be discriminated inputs first individual information (forefinger, for example) from an input sensor 3 at collation, an entire identification data sequential collation means 9 collates it with entire identification data in the individual data base 7, extracts candidate data and stores it in a candidate table 10. When second individual information (second finger, for example) is inputted from the input sensor 3, a second individual data collation means 14 collates it with candidate data and retrieves matching data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for identifying an individual using physical characteristics of the individual such as a fingerprint.

[0002]

2. Description of the Related Art In order to prohibit unauthorized persons from entering a specific area in a building, a door which is opened and closed by an electric lock is provided at a passageway or an entrance to the specific area, and a passage to the specific area is provided near the door. In some cases, a personal identification device that determines whether or not a person is authorized is installed. In addition, a personal identification device may be installed in order to manage who can operate a terminal such as a personal computer and who should not operate the terminal. As this personal identification device, a device that detects a fingerprint to identify an individual is usually used.

FIG. 5 is an overall configuration diagram showing a conventional personal identification device. The personal identification device 1 includes a numeric keypad 2 and an input sensor 3 for inputting personal information such as a fingerprint.
A personal database (DB) 7 is provided inside the personal identification device 1, and the individual (ID) number 7 a of the discriminator input from the numeric keypad 2 and the input sensor 3 via the personal data registration means 31. The required number of registered personal identification data 7b are registered.

At the time of collation, the ID number of the discriminator is input using the ten keys 2 and the ID
It is checked whether the number is registered in the DB 7, and if it is registered, the ID number 7a is used as a search key from the DB 7 to extract the identification data 7b related thereto. Here, the discriminator's personal information captured from the input sensor 3 via the personal information input means 32 is input, and the identification data 7b extracted from the DB 7 is compared with the personal identification data comparing means 34.
Compare and collate with. As a result, it is determined whether the degree of coincidence is equal to or more than the predetermined degree of coincidence α. If the degree of coincidence is less than α, the NG lamp 17 is turned on as “No”.

[0005] As a general comparison / collation performance,
The rate at which another person's personal information is determined to be “correct” is called the false acceptance rate FA, and the degree of coincidence α is determined in order to guarantee its performance.

[0006]

In the above-described conventional personal identification device, the ID number of the discriminator is input using the ten keys 2 at the time of collation.
It is necessary to memorize the number, which is extremely troublesome. It is also conceivable to compare all the registered target identification data 7b without specifying the ID number. In this case, however, the false acceptance rate FA is reduced because of the independent matching of the number of registered records. It increases in proportion to the number of cases, and the basic performance of the device is greatly reduced.

If the operation of inputting the ID number using the numeric keypad 2 or the like is completely eliminated, a specific individual whose identification data is unstable (for example, a person whose fingerprint is thin in fingerprint identification) is identified only by the identification data 7b. However, there is a problem that the matching accuracy is low, and there is a possibility of erroneous recognition.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problem, and an individual discriminating apparatus capable of discriminating an individual without specifying an ID number and using only personal information without lowering the identification accuracy. The purpose is to provide.

[0009]

A personal identification device according to a first aspect of the present invention registers personal identification data in a plurality of types of personal databases corresponding to personal numbers, and when the first personal information is input, the personal identification data is registered. Is compared with all the identification data registered in the personal database to extract candidate data, and the candidate data is compared with the second personal information to search for matching data.

The personal identification device according to the second invention is a personal database for registering a plurality of types of personal identification data corresponding to personal numbers, and a first personal information input means for inputting first personal information of personal information. And second personal information input means for inputting second personal information, collating the first personal information with identification data registered in a personal database to extract candidate data, The matching data is searched by comparing the candidate data.

The personal identification device according to a third invention is the personal identification device according to the second invention, wherein personal identification data corresponding to a personal number is sequentially registered in a personal database.

A personal identification device according to a fourth invention is the personal identification device according to the second invention, wherein the inputted first personal information is sequentially collated with all the identification data registered in the personal database. It is.

The personal identification device according to a fifth aspect of the present invention is the personal identification device according to the second aspect, wherein candidate data falling within the matching condition is stored in the master collation table based on the identification data collation result.

The personal identification device according to a sixth aspect of the present invention is the personal identification device according to the second aspect of the present invention, wherein the target identification data in the personal database corresponding to the personal number input from the personal number input means is designated. Things.

[0015]

BEST MODE FOR CARRYING OUT THE INVENTION

Embodiment 1 FIG. 1 to 4 show one embodiment of the first to sixth aspects of the present invention. FIG. 1 is an overall configuration diagram, FIG. 2 is a block diagram, FIG. 3 is a registration operation flowchart, and FIG. 5 is an operation flowchart, in which the same reference numerals indicate the same parts.

In FIG. 1, 1 is an individual identification device, 2 is a numeric keypad, 3 is an input sensor for inputting personal information such as a fingerprint, for example, and 4 is ID data input means for inputting an individual (ID) number using the numeric keypad 5. Is first personal information input means for inputting the first identification data of the user (for example, fingerprint data of the index finger) by the input sensor 3, and 6 is also for inputting the second identification data of the user (for example, fingerprint data of the middle finger) This is a second personal information inputting means.

Reference numeral 7 denotes a personal database (DB) storing first identification data 7b and second identification data 7c corresponding to the user ID number 7a registered by the identification data continuous registration means 8, and 9 denotes first personal information. The data input from the input means 5 and the identification data 7b in the personal database 7,
All identification data sequential collation means 10 for comparing and collating all cases 7c is a candidate table for storing ID numbers, identification data and coincidences equal to or higher than a predetermined coincidence (described later) from the collation results.

Reference numeral 11 denotes candidate data extracting means for extracting the user's data from the candidate table 10 based on the collation result of the all identification data sequential collating means 9 and storing the data in the master collation data table 12; A matching target range designating means for determining a search range of the second personal data comparing means for comparing and matching with new identification data of a user input from the second personal information input means 6 within the search range , 15 are the second personal data matching means 14
A comparison result display / output means for outputting a comparison / comparison result by an OK lamp 16 indicating that the comparison result is "correct", an NG lamp 17 indicating "no", and an electric lock 18 for locking or unlocking a door or the like.
Issue a control command.

In FIG. 2, reference numeral 21 denotes an A / D converter which is connected to the input sensor 3 and converts an analog input into a digital output; and 22, an input / output circuit (I / D) connected to the numeric keypad 2.
O) and 23 are CPUs connected to the A / D converter 21 and the I / O 22 to perform digital calculations such as image processing using the memory 24, and 25 is for controlling the OK lamp 16, the NG lamp 17 and the electric lock 18 Digital output circuit (DO) 2
Reference numeral 6 denotes a connected I / O.

Next, the operation of this embodiment will be described with reference to FIGS. A Registration Operation (FIG. 3) This operation shows the contents of the identification data continuous registration means 8 of FIG. First, in step S1, the registrant's ID number is input using the numeric keypad 2. Next, in step S2, the first personal information of the registrant is fetched from the input sensor 3, and in step S3
And the data is stored in the personal database 7 with the ID number 7a.
Is stored in the area corresponding to the first identification data 7b.

Next, in step S4, the second personal information is similarly fetched from the input sensor 3, and in step S5, the data is stored in the area of the second identification data 7c at the location corresponding to the ID number 7a.

Collation Operation (FIG. 4) This operation is performed by sequentially comparing all the identification data in FIG. 1, the candidate data extracting means 11, the collation target range designating means 13, the second personal data collating means 14, and the collation result display / output. The contents of the means 15 will be described. First, in step S11, the first personal information of the discriminator is fetched from the input sensor 3. Next, in step S12, the input data fetched in step S11 is compared with respect to all the identification data (first and second identification data 7b, 7c) of the personal database 7, and it is determined whether or not each of them is equal to or more than a predetermined matching degree α.

If the degree of coincidence is not less than α, the ID number 7a of the personal database 7 and the first identification data 7
b, the second identification data 7c and the degree of coincidence α are stored in the candidate table 10, and the process proceeds to step S14. If the degree of coincidence is less than α, the process jumps to step S14. Step S1
In step 4, it is determined whether all the identification data in the personal database 7 have been compared. If all the data are compared, the process proceeds to step S15, and if not, the process returns to step S12.

In step S15, it is determined whether there is data in the candidate table 10, and if there is data, step S16
To the ID number 7a and the other identification data (e.g., the first identification data) corresponding to the ID number 7a in the descending order of the degree of coincidence α in the data stored in the candidate table 10.
If the identification data 7b is stored, the second identification data 7
c) is extracted from the personal database 7 and recorded in the master collation data table 12 as master collation data Mst-Data. If there is no data in the candidate table 10, the NG lamp 17 is turned on in step S21, and the process ends.

As a backup function, when an ID number is input by the ten keys 2 in step S22, the first identification data 7b corresponding to the ID number is extracted in step S23, and the master identification data Mst-Data is extracted as master collation data Mst-Data. It can be recorded in the collation data table 12.

Then, in step S17, the second personal information of the discriminator is fetched from the input sensor 3 again. Step S1
In step 8, the second personal information fetched in step S17 is compared with the master collation data Mst-Data previously recorded in step S16 or S23, and it is determined whether or not the degree of coincidence is equal to or greater than a predetermined degree of coincidence β. When the degree of coincidence is equal to or greater than β,
In step S24, the OK lamp 16 is turned on as collation OK,
In step S25, the electric lock 18 is controlled. If the degree of coincidence is less than β, the process proceeds to step S19, and it is determined whether all the data in the candidate table 10 have been compared.

If all the data have not been compared, the flow returns to step S16.
In step S20, the NG lamp 17 is turned on, and the process ends.

Here, the result of calculation regarding the discriminating state at the time of the above collation is shown. The false acceptance rate F in one-to-one comparison and collation between one mastery collation data and input data of this apparatus
A performance is Pa, the number of registered persons is n, Pa = 0.1%, n
= 10. When the first personal information is compared with all the identification data 7b and 7c in the personal database 7 in step S12,
Since the independent stochastic event is performed 2n times, the performance is FA (S12) = Pa × 2n = 2%.

The number m of candidates extracted in step S13 based on the collation result is a value obtained by adding the number of probabilities that even the correct person and another person will extract, and m = 1 + FA (S12) = 1 + Pa × 2n = 1. 02 people.

Therefore, in step S18, when the second personal information is compared with the identification data in the personal database 7 specified in the candidate table 10, the false acceptance rate FA is:
Since the number of independent stochastic events is performed m times corresponding to the number of candidates, FA (S18) = Pa × m = Pa × (1 + Pa × 2) =
pa + 2nPa ² = 0.102% ≒ Pa, which is equivalent to the performance of one-to-one collation of ordinary one master data and input data, and the false acceptance rate FA performance does not decrease.

Here, steps S1 to S5 are identification data continuous registration means 8, steps S11 to S18 are data collation means, and step S11 is first personal information input means 5.
In step S12, all the identification data sequential collation means 9
In step S16, the candidate data extracting means 11 is used. In step S17, the second personal information input means 6 is used.
8 constitutes the second personal data collating means 14, step S22 constitutes the ID data input means 4, and step S23 constitutes the collation target range designating means 13.

In this way, the first personal information is not entered at the time of collation, and the two types of identification data 7b, 7
c, the candidate data is extracted, and the candidate data is compared with the second personal information to search for matching data. Therefore, the matching accuracy is not reduced, and the input order of the personal information (for example, In the fingerprint collation, it is not necessary to specify the order of placing the fingers, and the collation operation can be simplified. Further, since the identification data included in the matching condition is extracted and stored as the master collation data, the second
The collation speed at the time of personal data collation is improved.

Also, for backup purposes, I
By inputting the D number, collation using the ID number and one type of identification data can be performed in the same manner as in the conventional device, and the collation accuracy is reduced even for a person whose identification data is unstable (for example, a person with a thin fingerprint). I will not do it.

Embodiment 2 As compared with the first embodiment, it is possible to increase the types of identification data and to perform multi-step collation, thereby further improving the false acceptance rate FA.

[0035]

As described above, in the first and second aspects of the present invention, personal identification data is registered in a plurality of types of personal databases corresponding to personal numbers, and when the first personal information is inputted, the personal information is registered. Since the candidate data is extracted by collating with all the identification data registered in the database, and the candidate data is collated with the second personal information to search for matching data, the discriminator stores the ID number. The collation operation can be simplified without necessity and without lowering the collation accuracy.

According to the third aspect of the present invention, the personal identification data corresponding to the personal number is sequentially registered in the personal database. Therefore, in addition to the effect that the collation operation can be simplified, the registration speed of the identification data is improved. can do.

In the fourth invention, the input first personal information is sequentially collated with all the identification data registered in the personal database, so that it is not necessary to specify the input order of the personal information. The collation operation can be simplified.

Further, in the fifth invention, candidate data falling within the matching condition are stored in the master collation table based on the identification data collation result, so that the collation operation can be simplified. (2) The collation speed at the time of personal data collation can be improved.

According to the sixth aspect of the present invention, the target identification data in the personal database corresponding to the input personal number is designated. As a result, even for a person whose identification data is unstable, it is possible to prevent the collation accuracy from lowering.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram showing a first embodiment of the present invention.

FIG. 2 is a block diagram showing the first embodiment of the present invention.

FIG. 3 is a registration operation flowchart showing the first embodiment of the present invention.

FIG. 4 is a flowchart illustrating a collation operation according to the first embodiment of the present invention;

FIG. 5 is an overall configuration diagram showing a conventional personal identification device.

[Explanation of symbols]

1 Personal identification device, 2 numeric keypad, 3 input sensor, 4
ID data input means, 5 first personal information input means, 6
Second personal information input means, 7 personal database, 7a
ID number, 7b first identification data, 7c second identification data, 8 identification data continuous registration means, 9 all identification data sequential collation means, 10 candidate table, 11 candidate data extraction means, 12 master collation data table, 1
3 collation range designation means, 14 second personal data collation means, 15 collation result display / output means, S1-S5 identification data continuous registration means, S11-S18 data collation means, S11 first personal information input means, S12 all identification Data sequential collating means, S16 the candidate data extracting means, S16
17 second personal information input means, S18 second personal data collation means, S22 ID number input means, S23 collation target range designation means.

Claims (6)

[Claims] 1. An apparatus for comparing input personal information with registered personal identification data to determine whether or not they match, registers the identification data in a plurality of types of personal databases corresponding to personal numbers. When the first personal information is input, the first personal information is collated with all the identification data registered in the personal database to extract candidate data, and the candidate data is collated with the second personal information to obtain a match data. An individual discriminating device comprising a data collating means for retrieving. 2. A personal database for registering a plurality of types of said identification data in correspondence with a personal number in an apparatus for comparing input personal information with registered personal identification data to determine whether or not they match. First personal information input means for inputting the first personal information of the personal information, and matching the input first personal information with identification data registered in the personal database to generate candidate data. Identification data matching means to be extracted; second personal information input means to input second personal information of the personal information; and matching data by collating the input second personal information with the extracted candidate data. And a second personal data collating means for searching for a personal data. 3. The personal identification device according to claim 2, further comprising identification data continuous registration means for sequentially registering personal identification data corresponding to the personal number in a personal database. 4. An identification data collating means, comprising:
3. The personal identification device according to claim 2, wherein the personal information is sequentially collated with all the identification data registered in the personal database. 5. A master collation data table is provided, and candidate data extraction means is provided for storing candidate data falling within the matching condition based on the collation result by the identification data collation means in the master collation data table. The personal identification device according to claim 2, wherein 6. A personal number input means for inputting a personal number, and a collation target range specifying means for specifying target identification data in a personal database corresponding to the personal number input from the personal number input means are provided. 3. The personal identification device according to claim 2, wherein: