JP2003263640A - Personal identification device using biological information - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an inexpensive and highly reliable personal identification device capable of quickly and highly precisely realizing personal identification with much smaller data quantity. <P>SOLUTION: This personal identification device is provided with a means 20 for detecting a fingerprint pattern and vein pattern, a means 30 for extracting the feature part of the detected fingerprint pattern and vein pattern, and a means 18 for confirming the matching of the position of the feature part of the extracted fingerprint pattern with the position of the preliminarily registered corresponding feature part, and for confirming the matching of the position of the feature part of the extracted vein pattern with the position of the preliminarily registered corresponding feature part. The position of the feature point of the vein pattern is substantially associated with the position of the feature part of the fingerprint pattern. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a personal identification device for identifying an individual by using a fingerprint pattern and a vein pattern of a fingertip of a human body, and more specifically, a predetermined feature point of a fingerprint pattern and a predetermined vein pattern. The present invention relates to a personal identification device that can identify an individual quickly and with high accuracy by using the positional relationship with the feature points.

[0002]

2. Description of the Related Art In electronic commerce and the like, in order to improve the security of transactions, a personal identification device for collating biometric information peculiar to a human body with registered biometric information has been used in place of conventional password collation. There is. As such biometric information, for example, fingerprint feature points, pupil iris, blood vessels, etc. are used as described in JP 2001-273498 A and JP 7-21373 A. Such biometric information is information peculiar to an individual's human body and is hardly forged like a password, and greatly contributes to the security of transactions.

The personal identification device individually or independently collates two or more types of biometric information such as fingerprint feature points and pupil iris with the corresponding two or more types of corresponding biometric information registered in advance. By increasing the data amount of biometric information, it is possible to improve the identification accuracy and security.

[0004]

However, for such a personal identification device, not only improvement of identification accuracy and security but also quick response, that is, quick identification is important. This quick identification is desirable in any application, but requires a quick response, such as identification to unlock the vehicle door lock or identification to allow the vehicle engine to start. It is especially important in applications.
Specifically, when the personal identification device is applied to entrance / exit management of a security system, it is sufficient if the individual identification device can identify an individual in about 3 to 5 seconds, but it is applied to a door lock system or an engine starting system of an automobile. In some cases, identification should be completed within at least 1 second.

From this point of view, it is advantageous from the viewpoint of identification accuracy to use two or more types of biometric information or to use a large amount of biometric information (for example, 100 or more fingerprint pattern feature points). On the other hand, it is disadvantageous in that the amount of data to be processed at the time of matching is increased. On the other hand, it is possible to quickly process the increased data amount using a computer having a high processing capacity, but there is a problem that the cost increases. Also, the use of blood vessel bifurcation points and a limited number of fingerprint feature points contributes to speeding up identification to some extent, but from the viewpoint of identification accuracy, there is a certain limit to the amount of data that can be reduced. is there.

Therefore, an object of the present invention is to provide a low-cost and highly reliable personal identification device capable of realizing quick and highly accurate personal identification with a smaller amount of data.

[0007]

The object is to identify each individual by using the positional relationship between the characteristic points of the fingerprint pattern and the characteristic points of the vein pattern as described in claim 1. , Is achieved by a personal identification device using biometric information.

According to the above invention, since the positional relationship between the feature points of the fingerprint pattern and the feature points of the vein pattern is used,
Compared with the case where the characteristic points of the fingerprint pattern and the characteristic points of the vein pattern are compared with the corresponding characteristic points that are independently registered, it is possible to achieve highly accurate identification with a small amount of data. The positional relationship between the characteristic points of the fingerprint pattern and the characteristic points of the vein pattern is, for example, the positional relationship between the figure connecting the characteristic points of the fingerprint pattern and the figure connecting the characteristic points of the vein pattern, the distance or the overlapping area. Or
It may be the distance between the position of the characteristic point of the fingerprint pattern and the characteristic point of the vein pattern. Since these positional relationships are information peculiar to each individual and have the ability to identify each individual,
By collating the positional relationship with the registered corresponding positional relationship, it is possible to reduce the data amount of the biometric information to be used for achieving highly accurate identification. As a result, quick and highly accurate identification of an individual can be achieved.

The above-mentioned object is, as described in claim 2, means for confirming the consistency between the position of the detected feature point of the fingerprint pattern and the position of the corresponding feature point registered in advance, and the detected vein. A personal identification device, comprising: means for confirming the consistency between the positions of the characteristic points of the pattern and the positions of the corresponding characteristic points registered in advance, wherein the positions of the characteristic points of the vein pattern are It is achieved by a personal identification device characterized in that it is substantially associated with the position of the minutiae.

According to the above invention, since both the fingerprint pattern and the vein pattern are used as biometric information, highly reliable identification can be achieved. Moreover, since the positions of the characteristic points of the fingerprint pattern used for identification and the positions of the characteristic points of the vein pattern are substantially associated with each other, highly accurate identification can be performed with a smaller amount of data. That is, the positions of the characteristic points of the fingerprint pattern and the positions of the characteristic points of the vein pattern are respectively compared with the positions of the corresponding characteristic points registered in advance, so that the characteristic points of the fingerprint pattern and the characteristic points of the vein pattern are Since the positional relationships of the above are also verified at the same time, it is possible to achieve sufficiently accurate identification of the individual even if the number of the characteristic points of the fingerprint pattern and the vein pattern used for the verification are reduced. Becomes As a result, quick and highly accurate identification of an individual can be achieved.

In order to associate the positions of the characteristic points of the fingerprint pattern with the positions of the characteristic points of the vein pattern, for example, the positional relationship between the figure connecting the characteristic points of the fingerprint pattern and the figure connecting the characteristic points of the vein pattern. May be used, or the distance or direction between the characteristic point of the fingerprint pattern and the characteristic point of the vein pattern may be used. Alternatively, the position of the characteristic point of the vein pattern to be collated is substantially correlated with the position of the characteristic point of the fingerprint pattern by being corrected by the correction coefficient obtained from the collation result regarding the position of the characteristic point of the fingerprint pattern. Good. In this case, this correction coefficient plays a role of compensating for an error in the position of the fingertip to be placed, and also a role of substantially relating the position of the characteristic point of the vein pattern and the position of the characteristic point of the fingerprint pattern.
The target of the correction may be either the position of the detected feature point of the vein pattern or the position of the registered feature point of the vein pattern.

[0012] As for the above-mentioned object, as described in claim 3, the characteristic points of the fingerprint pattern and the characteristic points of the vein pattern which are registered in advance are collated with the detected characteristic points of the fingerprint pattern and the characteristic points of the vein pattern. By this means, a personal identification device for identifying each individual, including means for extracting the feature points of the vein pattern to be collated from the detected vein pattern, and the extraction position of the feature point of the vein pattern. Is specified by the collation result of the detected fingerprint pattern with the previously registered fingerprint pattern and the position information of the characteristic point of the previously registered vein pattern. To be done.

According to the above invention, since both the fingerprint pattern and the vein pattern are used as biometric information, highly reliable identification can be achieved. Furthermore, to confirm the presence of the characteristic points of the vein pattern at the extraction position specified from the positional relationship with the fingerprint pattern, the positional relationship between the characteristic points of the vein pattern and the fingerprint pattern is also confirmed. Therefore, even if the number of characteristic points of the fingerprint pattern and the characteristic points of the vein pattern used for the matching is reduced, it is possible to achieve sufficiently accurate identification of the individual. As a result, quick and highly accurate identification of an individual can be achieved.

The extraction position at which each characteristic point of the vein pattern is extracted is based on the displacement amount and the rotation angle of the detected fingerprint pattern with respect to the registered fingerprint pattern, and the position information of the characteristic point of the registered vein pattern. , May be specified. The extraction position specified in this way is one in which the error in the position of the fingertip to be placed is compensated. Further, the extraction position may be a region having a certain extent in consideration of a specific error of the extraction position. As a result, it is possible to quickly and surely extract the characteristic points of the vein pattern for collation, and even for a finger placed at an incorrect position, it is not necessary to reposition the finger, A quick identification can be realized.

Further, as described in claim 4, claim 1
In the personal identification device according to any one of items 1 to 3, if the characteristic points of the vein pattern are intersections of the vein pattern, highly accurate identification can be realized due to its uniqueness. In addition, the intersection of the vein pattern can be easily associated with the feature point of the fingerprint pattern, or the positional relationship with the feature point of the fingerprint pattern is clear, which can contribute to quick and highly accurate identification.

[0016]

1 is a block diagram showing an embodiment of a personal identification device 10 according to the present invention. The personal identification device 10 includes a biometric information detection device 20 that detects a vein pattern and a fingerprint pattern of a human body, and an image processing unit 3 that processes an image of the detected vein pattern and a fingerprint pattern.
0, a memory unit 16 that stores personal biometric information registered in advance, and a control unit 18 that executes various controls.

FIG. 2 schematically shows a configuration example for embodying the biological information detecting device 20. This biometric information detection device 20 includes vein information detection means 12 for detecting a vein pattern of a fingertip, fingerprint information detection means 14 for detecting a fingerprint pattern, and a detection table 22 on which a fingertip of an individual is placed. . Further, the biological information detecting device 20 may include a sensor unit 22a that detects that the fingertip is placed on the detection base 22.

The vein information detecting means 12 has an emission wavelength of 780.
The infrared light emitter 12a that generates infrared light of about nm to 1000 nm and the vein detector 12b that faces the infrared light emitter 12a with the detection table 22 interposed therebetween are included. The vein information detecting means 12 may use an incandescent lamp or a halogen lamp as a light source instead of the infrared light emitting body 12a. In such a case, the vein information detection means 12 has a visible light cut filter, and is configured such that only the infrared component of the white light that passes through the fingertip reaches the vein detector 12b.

The vein detector 12b is composed of a CCD (charge coupled device) and various lenses that form an image of infrared light transmitted through a fingertip on the CCD, and image-processes vein image data of a vein pattern. It is sent to the unit 30. The infrared emission wavelength of the infrared emitter 12a may be determined in consideration of the sensitivity of the CCD.

The fingerprint information detecting means 14 may be configured to optically read the fingerprint of the fingertip pressed on the detection base 22 using a CCD, or the fingertip detection base 22 using a capacitor array. It may be configured to detect the capacitance of each capacitor when mounted on the. The fingerprint image data detected by the fingerprint information detecting means 14 is
It is sent to the image processing unit 30.

Biological information detection device 2 having the above-mentioned configuration
With 0, both vein image data and fingerprint image data can be obtained at the time of registration and actual authentication. However, the configuration for detecting the vein pattern and the fingerprint pattern is known to those skilled in the art, and it should be understood that the configuration is not particularly limited to the above-described configuration and can be realized in various modes. For example, the infrared light emitter 12 described above
The a and the light source do not necessarily have to face the infrared light emitter 12a, and may be configured to illuminate the fingertip from the vein detector 12b side. Also, the fingerprint information detecting means 1
4 does not use the detection base 22 with which a fingertip comes into contact,
It may be configured to detect a fingerprint pattern in a contactless manner.

Next, the processing of the image processing unit 30 for processing the vein image data and the fingerprint image data thus obtained will be described.

FIG. 3 is a diagram schematically showing a vein pattern and a fingerprint pattern input to the image processing unit 30.

The image processing section 30 executes a process of extracting feature points of a fingerprint pattern from the input fingerprint image data. The characteristic point of this fingerprint pattern is the branch point a of the fingerprint.
The feature point of the fingerprint pattern to be extracted may be limited to a particular feature point existing near the center of the fingertip of the first joint from the viewpoint of speeding up recognition. The position information regarding the positions of the feature points extracted in this way is output as fingerprint feature data. The fingerprint characteristic data may further include information on a characteristic fingerprint pattern C formed on the substantially central portion of the fingertip of the first joint (hereinafter, referred to as “reference fingerprint pattern C”). The information about the reference fingerprint pattern C may be a set of position information of points forming the reference fingerprint pattern C. As will be described later, the information on the reference fingerprint pattern is used to predict the error of the position, rotation angle and inclination of the fingertip placed on the detection table 22.

Similarly, the image processing section 30 executes a process for extracting the characteristic points of the vein pattern from the input vein image data. The characteristic point of this vein pattern may be the intersection point d of the vein pattern or the change point e of the curved portion on the image showing the vein as shown in FIG. 3, and the characteristic point of the extracted vein pattern. From the viewpoint of speeding up recognition, may be limited to a specific intersection d existing near the center of the fingertip of the first joint. Information about the positions of the characteristic points of the vein pattern extracted in this way is output as vein characteristic data.

The fingerprint characteristic data and the vein characteristic data thus obtained are sent to the control unit 18. The control unit 18 stores these data obtained at the time of initial registration or update registration in the memory unit 16 as registered biometric data. At the time of matching, the control unit 18 responds to the input of fingerprint characteristic data and vein characteristic data from the image processing unit 30,
The consistency between these input data and the registered biometric data read from the memory unit 16 is confirmed.

FIG. 4 is a flow chart showing a first embodiment of the processing executed by the control unit 18 at the time of collation.

In step 100, the reference fingerprint pattern C of the input fingerprint characteristic data is compared with the reference fingerprint pattern C of the registered biometric data. Note that this processing is executed as primary authentication for high-accuracy and quick identification, and may be omitted. As a result of the collation, if the consistency can be confirmed, the process proceeds to step 102, and if the consistency cannot be confirmed, the process proceeds to step 110.

In step 102, the control unit 18 calculates the positional relationship of each feature point of the fingerprint pattern and vein pattern from the input fingerprint feature data and vein feature data, and proceeds to step 104. This positional relationship may be the relative position of each feature point of the fingerprint pattern and vein pattern with reference to the position of a particular feature point of the fingerprint pattern or vein pattern, or It may be the distance or the direction of the characteristic points of the pattern.

In step 104, the control unit 18 collates the positional relationship obtained in step 102 with the positional relationship relating to the registered biometric data corresponding to the positional relationship.
If the result of the collation is that the consistency can be confirmed, step 1
If the consistency cannot be confirmed in step 06, the process proceeds to step 108.

In step 108, the control unit 18 performs a predetermined process when the authentication is obtained, for example, by transmitting a signal instructing the actuator (not shown) to unlock the door, and the first operation is performed. Finish the example process. Further, in step 110, the control unit 18 performs a predetermined process to be executed when the authentication is not obtained, such as a process of generating an error signal, and ends the process of the first embodiment.

According to the first embodiment described above, not only the consistency of the fingerprint pattern but also the consistency of the positional relationship between the characteristic points of the vein pattern and the characteristic points of the fingerprint pattern are confirmed. A high matching result can be obtained. Also,
By using the positional relationship between the vein pattern feature points and the fingerprint pattern feature points, individual identification can be achieved with a very small amount of data compared to the case where both the fingerprint pattern and the vein pattern are collated independently. can do.
That is, information obtained from the positional relationship between the fingerprint pattern feature points and the vein pattern feature points is further added to the information obtained from the fingerprint pattern feature points and the vein pattern feature points, for highly accurate matching. The required number of characteristic points of the fingerprint pattern and the required characteristic points of the vein pattern can be significantly reduced. As a result, the personal identification device 10 according to the present invention can achieve reliable and quick personal identification.

By the way, the user detects the finger 2 at the time of verification.
The fingertips are placed on the detection table 2 in the same relationship as when they were registered.
It is not always placed on No. 2. In the above-described first embodiment, the fingerprint characteristic data and the vein characteristic data can be further corrected in consideration of the case where the fingertip is placed on the detection base 22 at an incorrect position. Next, such correction will be described with reference to FIG.

FIG. 5A is a front view showing the relationship between the detection table 22 of the biological information detecting device 20 and the fingertip.
5B is a top view of FIG. 5A, and FIG.
It is a side view of FIG. 5 (A). In FIG. 5A, the fingertip is
It is mounted on the detection table 22 with an inclination α with respect to the center line Z. Further, in FIG. 5B, the fingertip on the detection table 22 at the time of creating the registered biometric data is shown by a broken line. In FIG. 5B, the fingertip at the time of matching is placed at a rotation angle β with respect to the center line X passing through the reference feature point q, as shown by the solid line, and the reference feature point q ′ at the time of matching is placed. Is placed by being offset by (Δx, Δy) with respect to the reference feature point q of the registered biometric data. Further, in FIG. 5C, the fingertip is placed on the detection table 22 with an inclination γ with respect to the center line X.

These correction coefficients α, β, γ, Δx, Δy
May be derived based on the comparison result between the reference fingerprint pattern C of the registered biometric data and the input reference fingerprint pattern C. For example, the correction coefficients β, Δx, and Δy can be derived from the difference in the angle and position of the input reference fingerprint pattern C with respect to the reference fingerprint pattern C of the registered biometric data. The correction coefficients α and γ are the reference fingerprint pattern C of the registered biometric data of the input reference fingerprint pattern C.
It can be derived from the difference between the widths w1 and w2 (see FIG. 3) with respect to, or can be derived from the density difference between the left and right front and back fingerprint patterns. These correction factors are derived in step 100 in the first embodiment,
It may be applied when calculating the positional relationship between the fingerprint pattern and each feature point of the vein pattern (step 102) or when collating the positional relationship (step 104).

The second embodiment described below applies each of the feature points of the vein pattern by applying the above-mentioned correction coefficient without using the positional relationship between the fingerprint pattern and each feature point of the vein pattern. Can be associated with a fingerprint pattern.

FIG. 6 is a flow chart showing the second embodiment of the processing executed by the control unit 18 at the time of collation.

In step 200, the control unit 18 collates the reference fingerprint pattern C of the input fingerprint characteristic data with the reference fingerprint pattern C of the registered biometric data. As a result of the collation, if the consistency can be confirmed, step 202
Proceed to step 21 and if consistency is not confirmed, step 21
Go to 0.

In step 202, the control unit 18 compares the input reference fingerprint pattern C with the reference fingerprint pattern C of the registered biometric data, and corrects the correction coefficients Δx and Δ described above.
Calculate y, α, β, γ.

Next, at step 204, the control unit 18
Is the correction coefficient Δx, Δy calculated in step 202,
After using α, β, and γ as necessary to correct the position information of each feature point in the input fingerprint feature data, the corrected position information of each feature point is used for each feature point in the registered biometric data. Check if it matches the location information. However,
This correction method differs depending on the expression format of the position information.
For example, when the position information of each feature point in the fingerprint feature data is represented by relative coordinates based on the detected fingerprint pattern, the control unit 18 determines that the coordinates of each feature point of the input fingerprint feature data are It is only necessary to confirm whether or not the coordinates of the corresponding feature points of the registered biometric data match. On the other hand, when the position information is represented by relative coordinates (for example, pixel coordinates of each feature point) with reference to a position other than the fingerprint pattern, the control unit 18 sets the correction coefficient Δx to the coordinates of each feature point of the fingerprint feature data. , Δy, α, β, γ, and then it is confirmed whether the coordinates of the corrected feature points match the coordinates of the feature points of the registered biometric data. If the consistency is confirmed in this way, the process proceeds to step 206, and if the consistency is not confirmed, the process proceeds to step 210.

In step 206, the control unit 18 controls the correction coefficients Δx, Δy, α, β, γ calculated in step 202.
Is used as necessary to correct each feature point of the input vein feature data, and then check whether the coordinates of each feature point after correction match the coordinates of each feature point of the registered biometric data. . For example, when the position information of each feature point in the vein feature data is represented by the pixel coordinates of each feature point, the control unit 18 sets the coordinates of each feature point of the vein feature data to the correction coefficients Δx, Δy, α, After correction using β and γ, it is confirmed whether the coordinates of the corrected feature points match the coordinates of the corresponding feature points of the registered biometric data.

In step 208, the control unit 18 performs a predetermined process to be executed when the authentication is successful, such as a process of generating a signal instructing the unlocking of the door, and the second process.
The process of the embodiment is ended. Further, in step 210, the control unit 18 performs a predetermined process to be executed when the authentication is unsuccessful, such as a process of generating an error signal,
The process of the second embodiment ends.

According to the second embodiment described above, since the matching of both the fingerprint pattern and the vein pattern is confirmed, highly reliable personal identification can be achieved. Furthermore,
Since the position of the feature point of the vein pattern is corrected using the correction coefficient derived from the reference fingerprint pattern C, the position of each corrected feature point is substantially associated with the fingerprint pattern. As a result, the corrected position information of each feature point has information about the relationship with the fingerprint pattern together with the position information of each feature point itself, so that a reliable result can be produced with a small amount of data. . Further, by applying the correction coefficient, it is possible to perform verification even for a finger that is not placed accurately, so that it is not necessary to guide the finger to an accurate position or to relocate the finger, which is necessary for identification. The time can be shortened. As a result, the personal identification device 10 according to the present invention can achieve reliable and quick personal identification.

In the third embodiment, which will be described next, the collation is executed by utilizing the positional relationship between the figure formed by each characteristic point of the fingerprint pattern and the figure formed by each characteristic point of the vein pattern. .

FIG. 7 is a flow chart showing the third embodiment of the processing executed by the control unit 18 at the time of collation.

In step 300, the control unit 18 compares the reference fingerprint pattern C of the input fingerprint characteristic data with the reference fingerprint pattern C of the registered biometric data. As a result of the collation, if the consistency can be confirmed, step 302
If the consistency cannot be confirmed, the control unit 18
Performs processing similar to step 210 of the second embodiment.

In step 302, the control unit 18 compares the input reference fingerprint pattern C with the reference fingerprint pattern C of the registered biometric data and corrects the correction coefficients Δx, Δy, α,
Calculate β and γ.

Next, at step 304, the control unit 18
Compares the relationship between each feature point in the input fingerprint feature data and vein feature data with the relationship in the registered biometric data to determine the consistency of the vein feature data and fingerprint feature data with the registered biometric data. To do. When the consistency is confirmed, the control unit 18 performs the same processing as the above step 208 of the second embodiment, and when the consistency is not confirmed, the above step 21 of the second embodiment is performed.
The same process as 0 is performed. Here, the relationship compared in step 304 will be described in detail with reference to FIG.

Each drawing of FIG. 8 is a diagram showing the relationship between each characteristic point of the fingerprint characteristic data and each characteristic point of the vein characteristic data.
8A and 8B show a graphic 70 formed by connecting the respective feature points of the fingerprint feature data and a graphic 72 formed by connecting the respective feature points of the vein feature data. . In step 304, the control unit 18 may compare the shapes or areas of the figures 70 and 72 with the shapes or areas of the corresponding two figures in the registered biometric data, respectively. At this time, the control unit 18 may consider the correction coefficients α and γ together and execute the comparison.
For example, as the correction coefficients α and γ increase, the distortion of the shape of the graphic 72 related to the vein characteristic data progresses,
The area of the figure 72 also becomes smaller. Therefore, the control unit 1
8 can perform the comparison in consideration of the correction coefficients α and γ obtained in step 302.

In the third embodiment described above, the correction coefficient is taken into consideration in the matching confirmation step, but the correction coefficient may be applied in the fingerprint characteristic data generation step.
In such a case, in step 304, the control unit 18 determines the consistency by comparing the relationship between the feature points in the corrected fingerprint feature data and vein feature data with the relationship in the registered biometric data. Become.

According to the third embodiment described above, since the matching of both the fingerprint pattern and the vein pattern is confirmed, highly reliable personal identification can be achieved. Further, by applying the correction coefficient, it is possible to produce a reliable result with a small amount of data as in the second embodiment, and it is not necessary to guide the finger to an accurate position, so that authentication can be performed. The time required can be shortened.

Alternatively, as an alternative embodiment of the third embodiment, the control unit 18 does not use the correction coefficient, but the shape or area of the region where the shape of the graphic 70 and the shape of the graphic 72 overlap, or the graphic 70. Between the figure and the figure 72 (see FIG. 8).
The positional relationship between the graphic 70 and the graphic 72 such as (see (B)) or distance may be compared with the corresponding positional relationship in the registered biometric data. In such a case, the positional relationship between the characteristic points of the fingerprint pattern and the characteristic points of the vein pattern is compared, so that the amount of data used for identification can be reduced, as in the first embodiment. That is,
Even if the number of feature points forming the figures 70 and 72 (there are only four in the figure, but there are actually many), authentication with sufficiently high accuracy can be realized. However, in this alternative embodiment, it is possible for the control unit 18 to consider the correction factors α and γ when performing the comparison.

From the same point of view, as shown in FIG. 8C, the position of the characteristic point of the vein characteristic data in the graphic 70 relating to the fingerprint characteristic data or the fingerprint characteristic data in the graphic 72 relating to the vein characteristic data. It is also possible to use a positional relationship such as the position of a feature point. Alternatively, FIG.
As shown in (D), each feature point of the fingerprint feature data and the feature point of the vein feature data, such as the position, distance, angle, etc. of the feature point of the vein feature data adjacent to each feature point of the fingerprint feature data. You may use the positional relationship of. In such a case, the control unit 18 can consider the correction coefficients Δx, Δy, α, β, and γ.

According to the alternative embodiment of the above-mentioned third embodiment, since the consistency of both the fingerprint pattern and the vein pattern is confirmed, highly reliable personal identification can be achieved. Furthermore, since the positional relationship between the characteristic points of the fingerprint pattern and the characteristic points of the vein pattern is collated using a figure or the like,
As with the first embodiment, it is possible to achieve personal identification with a very small amount of data. As a result, the personal identification device 10 according to the present invention can achieve reliable and quick personal identification.

In the above-described three embodiments, the control unit 18 uses the characteristic points of the fingerprint pattern and vein pattern extracted by the image processing unit 30, but in the fourth embodiment described below. The control unit 18 specifies the extraction position from which the characteristic points of the vein pattern should be extracted, and the image processing unit 30
Then, it is confirmed whether or not the characteristic points of the fingerprint pattern and vein pattern actually exist at the specified extraction position.

FIG. 9 is a flow chart showing the processing executed by the control unit 18 according to the fourth embodiment during collation.
The image processing unit 30 of this embodiment outputs only the reference fingerprint pattern C to the control unit 18 as fingerprint feature data.

In step 400, the control unit 18 compares the reference fingerprint pattern C input as the fingerprint feature data with the reference fingerprint pattern C of the registered biometric data.
If the result of the collation is that the consistency can be confirmed, step 4
If the consistency cannot be confirmed in step 02, the process proceeds to step 410.

In step 402, the control unit 18 compares the input reference fingerprint pattern C with the reference fingerprint pattern C of the registered biometric data, and various correction coefficients Δx, Δy,
Calculate α, β and γ.

Next, at step 404, the control unit 18
Is the correction coefficient Δx, Δy calculated in step 202,
By appropriately using α, β, and γ, the image processing unit 30 specifies the extraction position to be searched in order to extract each feature point of the fingerprint pattern and vein pattern of the registered biometric data, and , A signal indicating the identified extraction position is transmitted.

In step 406, the image processing section 30
It is confirmed whether each feature point of the fingerprint pattern and the vein pattern actually exists at the extraction position. Note that the image processing unit 30 may check whether or not the feature point actually exists in a region having a certain degree of spread around the extraction position, taking into consideration the error of the specified extraction position. When the presence of all the feature points can be confirmed, the image processing unit 30 transmits a signal indicating that the authentication is possible to the control unit 18, and proceeds to step 408. When the image points cannot be confirmed, the image processing unit 30 authenticates the control unit 18. A signal indicating no is transmitted, and the process proceeds to step 410.

In step 408, the control section 18 performs a predetermined process to be executed when the authentication is obtained, and ends the process of the fourth embodiment. Further, in step 410, a predetermined process to be executed when the authentication is not obtained is performed, and the process of the fourth embodiment ends.

According to the fourth embodiment described above, since the matching of both the fingerprint pattern and the vein pattern is confirmed, highly reliable personal identification can be achieved. Furthermore,
Since the extraction positions of the characteristic points of the vein pattern will be specified based on the reference fingerprint pattern C, the characteristic points of the vein pattern will be substantially associated with the fingerprint pattern. Therefore, the feature point of the vein pattern used for authentication has information indicating the positional relationship between the feature point of the vein pattern and the fingerprint pattern in addition to the position information of the feature point itself, so that an enormous amount of data is required. A result as reliable as a match using can be derived. Also,
The image processing unit 30 confirms the existence of each feature point by
Since it is only necessary to process an area designated in advance, it is possible to eliminate the need to search each feature point in a wide range and reduce the processing load. Further, since the image processing unit 30 searches only a previously designated area and confirms the feature points, the possibility of extracting erroneous feature points is significantly reduced. Further, by applying the correction coefficient, it is possible to perform authentication even for a finger that is not placed accurately, so there is no need to request the user to guide the finger to an accurate position or place the finger again. It is possible to reduce the time required for authentication. As a result, the personal identification device 10 according to the present invention can achieve reliable and quick personal identification.

The "extracting means" described in claim 3
Is realized by the "image processing unit 30" of the embodiment described in the detailed description of the invention, and the "identification of extraction position" described in claim 3 is realized by the "control unit 18" described in the embodiment. To be done.

The extraction position of the characteristic point of the vein pattern is specified from the collation result of the detected fingerprint pattern with the previously registered fingerprint pattern and the position information of the characteristic point of the previously registered vein pattern. As described above, the preferred embodiments of the present invention have been described in detail, but the present invention is not limited to the above-described embodiments, and the above-described embodiments without departing from the scope of the present invention. Can be variously modified and replaced.

For example, in the above-described embodiment, various correction coefficients are derived using the reference fingerprint pattern C, but a plurality of characteristic points of a characteristic vein pattern or fingerprint pattern or a plurality of features of a vein pattern. Various correction factors can be derived using the points. Further, although some of the above-described embodiments are mentioned on the assumption that the position of the finger with respect to the detection base 22 varies to some extent,
The biological information detecting device 20 may have a tubular shape suitable for inserting a finger in order to reduce such variations, and may further include various positioning means for positioning the finger.

[0066]

Since the present invention is as described above, it has the following effects. According to the inventions of claims 1 to 3, since the positional relationship between the characteristic points of the fingerprint pattern and the characteristic points of the vein pattern is collated, the characteristic points of the fingerprint pattern and the characteristic points of the vein pattern are independently registered. Compared to matching with corresponding feature points,
Individuals can be quickly and accurately identified. Also,
According to the invention described in claim 4, since the singular point having the discriminating power of the vein pattern is used, it is possible to contribute to the reliability of the matching result.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing an embodiment of a personal identification device 10 according to the present invention.

FIG. 2 is a diagram schematically showing a configuration example for embodying a biological information detection device 20.

FIG. 3 is a diagram schematically showing a vein image and a fingerprint image input to the image processing unit 30.

FIG. 4 is a flowchart showing a process executed by the control unit 18 according to the first embodiment of the present invention.

FIG. 5 is a diagram for explaining a correction coefficient,
5A is a front view showing the relationship between the detection base 22 and a fingertip, FIG. 5B is a top view of FIG. 5A, and FIG. It is a side view of A).

FIG. 6 is a flowchart showing processing executed by the control unit 18 according to the second embodiment of the present invention.

FIG. 7 is a flowchart showing processing executed by the control unit 18 according to the third embodiment of the present invention.

FIG. 8A, FIG. 8B, FIG. 8C and FIG.
(D) is a diagram showing a relationship between each feature point of fingerprint feature data and each feature point of vein feature data.

FIG. 9 is a flowchart showing a process executed by the control unit 18 according to the fourth embodiment of the present invention.

[Explanation of symbols]

10 Personal identification device 12 Vein information detection means 12a infrared light emitter 14 Fingerprint information detection means 16 memory section 18 Control unit 20 biometric information detection device 22 Detection stand 22a sensor means 30 Image processing unit

Claims (4)

[Claims] 1. An individual identifying apparatus using biometric information, characterized in that each individual is identified by using a positional relationship between a characteristic point of a fingerprint pattern and a characteristic point of a vein pattern. 2. A means for confirming the consistency between the position of the detected feature point of the fingerprint pattern and the position of the corresponding previously registered feature point, and the position of the detected feature point of the vein pattern which is registered in advance. And a means for confirming the consistency with the position of the corresponding feature point, wherein the position of the feature point of the vein pattern is substantially associated with the position of the feature point of the fingerprint pattern. A personal identification device, characterized in that 3. Personal identification for identifying each individual by comparing the feature points of the fingerprint pattern and the vein pattern registered in advance with the detected feature points of the fingerprint pattern and vein pattern. An apparatus, comprising means for extracting the feature points of the vein pattern to be collated from the detected vein pattern, wherein the extraction positions of the feature points of the vein pattern are relative to the pre-registered fingerprint pattern. A personal identification device characterized by being specified from the collation result of the detected fingerprint pattern and the position information of the feature points of the vein pattern registered in advance. 4. The personal identification device according to claim 1, wherein the characteristic point of the vein pattern is an intersection of the vein patterns.