JPH02310681A - Individual person authenticating device - Google Patents

Abstract

PURPOSE:To make it unnecessary to obtain position coordinate for separating each finger picture by providing a guide member to guide how to place fingers on the finger placing table of an input means in the case where an individual person is authenticated by using plural fingers. CONSTITUTION:The picture of the fingers Fa, Fb placed on the finger placing table 11 as interposing a partition plate 20 between is read by a camera 13, and after being converted into an electric signal, it is digitized by an A/D converter 14. This picture is stored temporarily in a picture memory 16 by a control part 15. The control part 15 separates it for every finger Fa, Fb in conformity to an output from a partition plate position detector 21, and by outputting each feature information to a dictionary part 19, the feature information for collation is registered. Besides, at the time of the collation, by comparison (positioning) between the obtained feature information and the feature information for collation corresponding to each finger registered in the dictionary part 19, the person in question is authenticated.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a personal authentication device that authenticates individuals using physical characteristics such as handprints and fingerprints.

(Prior Art) Recently, devices that use the distance between finger joints have been developed as devices for personal authentication.

This type of personal authentication device uses an addition signal (feature information) obtained by adding the densities of finger images in a direction perpendicular to the longitudinal direction of the finger. In other words, an addition signal obtained by adding the density of image information of a finger obtained from the difference in contact/non-contact between the finger and the finger rest on which the finger is placed in a direction perpendicular to the longitudinal direction of the finger; Comparison is made with addition signals to be verified that have been registered in advance in a dictionary, and verification is performed depending on whether the distances between the joints in each signal match.

Note that with this type of device, processing may be performed on a single finger, or processing may be performed on two or more fingers with an increased amount of information in order to further improve discrimination ability. is possible.

By the way, in the case of a device that performs personal authentication using multiple fingers, the images of multiple fingers are separated one by one, and the sum signal in the direction orthogonal to the longitudinal direction of the fingers is calculated for each finger image. , a proposal has been made to improve the reliability of verification by performing verification by comparing the added signal to be verified obtained in the same manner for each corresponding finger. In such a device that separates multiple finger images one by one and processes each finger, for example, an addition signal (in the y direction) is added to a finger image including multiple fingers in the longitudinal direction of the fingers. find the sum signal),
Taking advantage of the fact that peaks corresponding to each finger appear in this added signal, coordinate positions for separating each finger image are determined by determining the minimum value between the winter peaks.

However, when separating each finger image using the method described above, there are the following problems.

That is, it takes time to calculate the a and y direction addition signals and to search for the minimum value.

b. If the input image contains noise N as shown in Figure 8(a), the resulting y-direction addition signal is as shown in Figure 8(b).
The reproducibility of local minimum values as shown in is poor.

C9 When the finger is placed at an angle with respect to the y-axis, as shown in Figure 9(a), a clear minimum value appears in the y-direction addition signal, as shown in Figure 9(b). do not have.

In this way, with conventional personal authentication devices, the processing speed when separating finger images for multiple fingers into individual fingers is slow, and the processing accuracy is low, resulting in low verification reliability. It had become.

(Problem to be Solved by the Invention) As described above, in a conventional personal authentication device that separates multiple finger images one by one and processes each finger, finger images for multiple fingers are The processing speed when separating the images into individual fingers was slow, and the accuracy of the processing was low, resulting in low verification reliability.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a personal authentication device that can easily and quickly determine accurate coordinate positions for separating each finger image, and can significantly improve the reliability of verification. It is an object.

[Structure of the Invention] (Means for Solving the Problems) The personal authentication device of the present invention includes an input means for inputting image information of a finger, and a density of the image information of the finger inputted by the input means. a generating means for generating feature information by adding the above in a direction orthogonal to the longitudinal direction of the finger; and a matching means for matching the feature information generated by the generating means with previously registered feature information for a target to be matched. In the apparatus for authenticating individuals using a plurality of fingers, the finger rest of the input means is provided with a guide member for guiding how to place the fingers.

(Function) This invention enables the positions of multiple fingers to be defined by the above-described means, so each time,
It becomes possible to separate each finger image reliably and at high speed without requiring processing to obtain position coordinates for separating finger images.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 shows a personal authentication device of the present invention.

In FIG. 1, reference numeral 11 indicates, for example, two fingers Fa of the person to be authenticated when image information of the fingers is manually input.

It is a finger and a stand for placing the Fb, and is made of a transparent prism or the like. On this finger rest 11, there is a partition plate 20 as a guide member that partitions the fingers Fa and Fb.
is provided.

The image information of the fingers Fa and Fb placed on the finger rest 11 with the partition plate 20 in between is that the light from the light source 12 disposed below the finger rest 11 touches the finger rest 11. By being irradiated through the finger rest 11 and the fingers Fa, F
The camera 13 reads the reflected light image depending on the difference between contact and non-contact with b.

The image information of the fingers Fa and Fb inputted to the camera 13 is converted into electrical signals, and then A/D converted by the A/D converter 14 and digitized. This digitalized finger image is stored in the image memory 1 by the control unit 15.
6 is temporarily stored.

On the other hand, the IC card 17 stores an arbitrary finger (in this embodiment, finger
, Fb) are stored. This characteristic information is, for example, an addition signal obtained by adding pixel densities of the finger image in a direction orthogonal to the longitudinal direction of the finger. Above IC card 1
7 is a reading/writing device 1 controlled by the control section 15;
8, the characteristic information is written (at the time of registration) or read (during verification), and these ICs
The card 17 and the reading/writing device 18 constitute a dictionary section 19 in which characteristic information of each person to be authenticated is registered in advance.

The control unit 15 converts the finger image stored in the image memory 16 into 6 according to the output from the partition plate position detector 21.
In addition to separating each finger Fa and Fb, 6 fingers Fa and Fb
By obtaining the characteristic information of each and outputting it to the dictionary section 19, the characteristic information to be matched is registered. Also,
At the time of verification, the identity is verified by comparing (positioning) the feature information obtained as described above with the feature information for verification corresponding to the six fingers registered in the dictionary section 19.

The host machine 30 controls, for example, the opening and closing of a door and the sounding of an alarm buzzer in accordance with the verification result from the control section 15. Note that this host machine 30
The subject of control differs depending on the security system to which this personal authentication device is applied.

The partition plate position detector 21 reads the position of the partition plate 20 on the image. In this case, the partition plate 20
Since, in principle, there is no need to move once fixed, the position of the partition plate 20 corresponds to the coordinates on the image when fixed.

2 and 3 show the partition plate 20. FIG.

As shown in FIGS. 2(a) and 2(b), the partition plate 20 is provided perpendicular to the surface of the finger rest 11 that contacts the finger and parallel to the X axis on the image. Therefore, the person to be authenticated places the finger Fa on the finger rest 20.

When placing Fb, for example, use two fingers Fa.

By sandwiching the partition plate 20 between Fb, both fingers Fa,
Fb is placed substantially parallel to the partition plate 20.

Furthermore, the partition plate 20 is provided so as to be movable up and down, and is usually at a position raised by a small distance Ma that does not touch the surface of the finger rest 11 on which the fingers abut, as shown in FIG. 3(a). The installation is fixed. Therefore, when a finger image is input, the image of the partition plate 20 does not appear on the input image. Therefore, it is possible to stably detect only the finger image without degrading its quality.

Further, silicone rubber 20a is attached to the lowermost part of the temporary partition 20, for example, at both edges. This silicone rubber 20a is for improving the adhesion between the partition plate 20 and the finger rest 11 when the partition plate position detector 21 captures the position of the partition plate 20 as coordinates on an image. . In principle, the partition plate 20 does not need to be moved once it is fixed. Therefore, before being fixed at a position a minute distance Ha away from the finger rest 11,
A correspondence is established between the position of the partition plate 20 and the coordinates on the image. In this case, as shown in FIG. 3(b), the partition plate 2
Adjust the height of 0 to bring it into contact with the finger rest 11. Then, the silicone rubber 2 attached to the edge of the partition plate 20
Due to the influence of 0a, total reflection at that part is prevented and light is diffused. This allows the position detector 21 to detect the partition plate 20 as an image, as shown in FIG. 4(a).

When the image of the partition plate 20 is detected, the obtained partition plate 20
This X
In the direction addition signal, a minimum value appears at the y-coordinate position corresponding to the position of the partition plate 20, as shown in FIG. 4(b).

By finding the y-coordinate value y1 that gives this minimum value,
The positions at which the finger images are separated are obtained for each finger Fa and Fb.

In this way, when the separation position gi (y coordinate value y) of the image is determined, the partition plate 20 is fixed at a position a minute distance a away from the finger rest 11, and thereby the normal state in which personal authentication is possible is established. It is said that

Next, with reference to FIG. 5, the processing in the control section 15 will be explained in detail. Here, the index finger and middle finger of the right hand are used as fingers Fa and Fb, respectively, and it is assumed that these are determined by the will of the person to be authenticated, and in principle, only the person himself/herself can know the contents.

In FIG. 5(a), fa and fb are finger images, that is, read by the camera 13, and are read by the A/D converter 14.
Index finger Fa and middle finger F digitized by
b is each image pattern. These finger images fa,
For example, fb has a resolution of 10 lines/mm and is 512
It is assumed that the image is captured in an area of 512 pixels x 512 pixels.

First, there are two fingers Fa placed on the 1 effect stand 11.

When the image Fb is taken into the image memory 16, the density of each finger image fa and fb is determined in a direction perpendicular to the longitudinal direction of the finger according to the y coordinate value y determined by the partition plate position detector 21. By adding to y, an added signal in the X direction is obtained. That is, by determining the addition signal (X-direction pixel addition value) in the X-direction in the range from 0 to the y-coordinate value y among the pixels (512) in the X-direction, the index finger The feature information A(xa) for Fa is also 51 from the y coordinate value y1.
By determining the addition signal in the X direction in the range up to 1, feature information A(xb) for the middle finger Fb is generated, as shown in FIG.

These characteristic information A (xa) and A (xb) have a steep valley (minimum value) at the position of the horizontal wrinkle corresponding to the joint of each finger.
This is the parameter (individuality) for authenticating individuals.

Next, the operation of the personal authentication device configured as described above will be explained. In this device, there are two main processes: "registration processing" and "verification processing".

FIG. 6 is a flowchart for explaining the "registration process". For example, assume that the index finger Fa and middle finger Fb of the right hand of the person to be authenticated are placed on the finger rest 11 in the registration mode. Then, the image information based on the difference between contact and non-contact between the fingers Fa and Fb and the finger rest 11 is converted into an electric signal by the camera 13, and further converted into an electric signal by the A/D.
After being A/D converted and digitized by the converter 14, they are temporarily stored in the image memory 16 as finger images fa and fb (step 5TI).

The control unit 15 adds the densities of the finger images fa and fb in the direction y perpendicular to the longitudinal direction X of the finger. That is, the X-direction addition signal in the range from 0 to the y-coordinate value y of the pixel in the X direction and the X-direction addition signal in the range from the y-coordinate value yl to 511 are respectively extracted. As a result, as shown in FIG. 5(b), the feature information A(
xa) is generated (step 5T2), and the fifth
As shown in Figure (c), feature information A(x
b) is generated (step 5T3).

The feature information A(xa) as the X-direction addition signal for the index finger Fa and the feature information A(xb) as the X-direction addition signal for the middle finger Fb thus obtained are each subjected to appropriate band-pass filter processing. It is registered in the dictionary section 19 in a converted form that is easy to perform later collation processing (step 5T4). That is, the characteristic information A (xa) and A (xb) are sent to the IC card 17 of the person to be authenticated via the reading/writing device 18 under the control of the control unit 15.
are stored respectively.

The registration process is completed as described above, and the feature information for verification regarding the fingers Fa and Fb of the person to be authenticated is transferred to the dictionary unit 19.
It will be registered in .

FIG. 7 is a flowchart for explaining the "verification process". Here, first, two pieces of characteristic information to be collated that are registered in advance in the dictionary section 19 are respectively read out from the dictionary section 19 (step 5T11).

The subsequent input of a finger image (step 5T12) and the extraction of feature information A (xa) for the index finger Fa and feature information A (xb) for the middle finger Fb from this finger image (step 5T13.14) are performed based on the above-mentioned registration. It is obtained by processing similar to those in .

In this way, when the feature information A (xa) and A (xb) as X-direction addition signals for the two fingers Fa and Fb are generated, the feature information for comparison read out from the dictionary section 19 and the finger image are used. Generated feature information A (xa), A
(xb) are aligned for each corresponding finger to be compared (step 5T15).

Here, one of the two feature information read from the dictionary section 19 is set as Ad (i), and one of the feature information generated from the finger image, that is, the above feature information Ad
Let A(i) be the feature information corresponding to (i), and let N be the number of elements of each added signal.

The value of N here can be considered to be 512. Then, the addition result obtained by adding the squared error between the feature information Ad (i) and the information A (i+m) obtained by shifting the feature information A (i) by m over a certain range is defined as S (m). This addition result S (m) is expressed when m≧0 and mho. In this case, the addition result S (m) is information A obtained by shifting the feature information Ad (i) read from the dictionary section 19 and the feature information A (i) generated from the finger image by m.
This is a parameter indicating a negative political change with (i+m), and indicates that the smaller the addition result S (m) is, the higher the degree of coincidence is. Therefore, when m is varied within a certain range, alignment is achieved at M when the addition result S (m) is the smallest, that is, the addition result S (m) is the smallest.
The value of S (M) when becomes small is taken as the verification result.

The verification result S (M) obtained in this way is used to determine whether the person to be authenticated is the person in question (step 5T1).
6). In this case, determining whether it is the person himself or herself or another person,
The determination is made based on the magnitude relationship between the result of the above-mentioned comparison S (M) and a threshold value TH, which is determined experimentally, for example. That is, S (M)≦TH matchS (M)〉TH mismatch As a result of this judgment, if it is determined that there is a match, the other of the two feature information is compared by the same process as above. is performed (step 5T17).

Then, the result of the verification is similarly used to determine whether the person to be authenticated is the authenticated person (step 5T).
18). That is, if the result of the above matching is S' (M), and for example the experimentally determined threshold value serving as the criterion for judgment is TH', then S'(M)≦TH' Match S' (M
)>TH' InconsistencyThe two types of threshold values TH used here.

TH' may be the same value or may be different values depending on the purpose (importance of security).

Based on the result of the comparison between the index finger Fa and the middle finger Fb obtained as described above, it is finally determined whether the person to be authenticated is the authenticated person (step 5T19.21). In this case, whether the person is the person himself or not is determined based on, for example, both fingers Fa.

If the identity of the Fb user is confirmed (Step 5T19)
) only recognizes him/herself and uses both fingers Fa.

If the identity of the person on Fb is not confirmed (Step 5T2)
1) I try not to recognize the person. However, if the purpose is management that is not very important, in order to improve passability and give priority to smooth operation, it may be necessary to recognize the individual when the identity of the individual is verified on either finger. Also good. Further, a discriminant function may be used to determine whether the person is the person himself or herself.

As a result of this determination, if it is determined that the person is the person in question, a match signal is output to the host machine 30, and if it is determined to be someone else, a mismatch signal is output to the host machine 30 (step 5T20). Then, the host machine 30 performs processing according to the signals, such as opening a door in response to a matching signal, or sounding an alarm buzzer in response to a non-matching signal, etc., to the security system to which the device is applied. Control is performed accordingly.

As described above, by placing multiple fingers on the finger rest along the partition plate, it is possible to prevent the fingers from being placed in an inclined position, and also to prevent the fingers from being placed in an inclined position. By determining the correspondence with the position in advance, it is possible to separate images for each finger without performing any special processing when inputting finger images.

In other words, the coordinate positions for separating six fingers can be easily obtained without determining the y-direction addition signal from a plurality of finger images. Thereby, images of six fingers can be separated reliably and at high speed. Therefore, it is possible to significantly improve the reliability of verification.

In addition, in the above embodiment, the explanation was given using an example of a device that authenticates an individual using two fingers, but the invention is not limited to this, and can also be applied to a device that authenticates an individual using three or more fingers. Applicable.

It goes without saying that various other modifications can be made without departing from the gist of the invention.

[Effects of the Invention] As detailed above, according to the present invention, accurate coordinate positions for separating each finger image can be easily and quickly determined, and the reliability of matching is significantly improved. It is possible to provide a personal authentication device that can

[Brief explanation of the drawing]

Fig. 1 is a configuration diagram of a personal authentication device showing an embodiment of the present invention, Fig. 2 is a diagram showing the positional relationship between the partition plate and the finger rest, and Fig. 3 is a diagram showing the position of the partition plate with respect to the finger rest. A diagram shown to explain the difference, FIG. 4 is a diagram showing the relationship between the image of the partition plate and the y-direction addition signal, and FIG. 5 is a diagram shown to explain the operation using two fingers as an example. FIG. 6 is a flowchart shown to explain the flow of processing at the time of registration, FIG. 7 is a flowchart shown to explain the flow of processing at the time of verification, and FIGS. 8 and 9 are both conventional techniques. FIG. 2 is a diagram shown to explain the problem. DESCRIPTION OF SYMBOLS 11... Finger rest, 12... Light source, 13... Camera, 15... Control part, 19... Dictionary part, 20...
Partition plate (guide member), 21... Partition plate position detector, Fa... Index finger, Fb... Middle finger. Applicant's agent Patent attorney Takehiko Suzue Figure 1 (a) (b) Figure 2 (a) (b) 0 511 X No. 5
Figure 6 (a) Figure 8 (a) Figure 9 γhrIl & aggregate Kato value (b) (b)

Claims (2)

[Claims] (1) an input means for inputting image information of a finger; a generation means for generating feature information by adding the density of the image information of the finger inputted by the input means in a direction orthogonal to the longitudinal direction of the finger; A personal authentication device for authenticating an individual using a plurality of fingers, comprising a matching means for matching the characteristic information generated by the generating means with characteristic information to be verified that is registered in advance, A personal authentication device characterized in that a guide member for guiding how to place a finger is provided on a finger rest of the input means. (2) The personal authentication according to claim (1), wherein the guide member is composed of a partition plate and includes means for associating the position of the partition plate with the position on the input image. Device.