JP2017162301A - Living body authentication device, living body authentication method, and living body authentication program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a living body authentication device capable of performing the guide of a living body corresponding to each of users and performing authentication, a living body authentication method, and a living body authentication program.SOLUTION: The living body authentication device includes a data base 20 which stores the living body features of a living body part and the shape of the living body part, a living body sensor which acquires the image of the living body part of the user, a three-dimensional shape calculation part 13 which calculates the shape of the living body part in the image acquired by the living body sensor, a guide control part 18 which presents guide information based on a relative comparison between the shape calculated by the three-dimensional shape calculation part 13 and the shape stored in the data base 20 to the user, and an authentication part 19 which collates the living body features stored in the data base 20 with the living body features in the image of the living body part of the user acquired by the living body sensor after presenting the guide information by the guide control part 18.SELECTED DRAWING: Figure 2

Description

  The present case relates to a biometric authentication device, a biometric authentication method, and a biometric authentication program.

  In biometric authentication, the posture of an authentication target part such as a hand is corrected with a guide when photographing registration data, but there are many cases where there is no assistance of the guide during authentication. For this reason, there is a difference in the shape of the part to be authenticated and the shape of the opening between the time of registration and the time of authentication, and the acceptance of the person is likely to occur (see, for example, Patent Document 1).

JP 2014-180435 A

R. Kimmel et al., "Global Shape from Shading", CVGIP: Image Understanding, pp.120-125, 1995 R. Zhang et al., "Shape from Shading: A Survey", IEEE PAMI (Transactions on Pattern Analysis and Machine Intelligence), Vol.21, No.8, pp.690-706, August 1999 E. Prados et al., "Shape from Shading: a well-posed problem?", INRIA, No.5297, pp.1-55, August 2004

  Therefore, it is conceivable to correct the biometric image according to the shape of the authentication target part, but there is a limit to the range of shapes that can be applied in the correction. It is effective to present a guidance message to the user so as to change the shape of the authentication target part itself, but even if a guidance message based on a uniform standard is presented, it may be difficult to follow the message due to individual differences is there. How to open the palm, that is, how much the palm can be opened varies from person to person. For this reason, it is not desirable to determine and guide the opening based on a uniform standard. For example, a message “please open the palm” is displayed to a person whose palm does not open to a certain extent, which is not desirable from the viewpoint of the user's psychological aspect and authentication accuracy.

  In one aspect, an object of the present invention is to provide a biometric authentication apparatus, a biometric authentication method, and a biometric authentication program that can perform authentication by guiding a biometric according to each user.

  In one aspect, the biometric authentication apparatus includes: a storage unit that stores biometric features of a biological part and the shape of the biological part; a biological sensor that acquires an image of the user's biological part; and an image acquired by the biological sensor. A calculation unit that calculates the shape of the living body part; a presentation unit that presents guidance information based on a relative comparison between the shape calculated by the calculation unit and the shape stored in the storage unit; A collation unit that collates the biometric feature stored in the storage unit with the biometric feature in the image of the user's biological part acquired by the biometric sensor after the guidance information is presented by the presentation unit.

  Authentication can be performed by guiding the living body according to each user.

(A) is a block diagram for demonstrating the hardware constitutions of the biometrics apparatus which concerns on Example 1, (b) is a schematic diagram of a biometric sensor. It is a block diagram of each function implement | achieved by execution of a biometrics authentication program. It is a flowchart which illustrates the detail of a registration process. It is a figure which illustrates a registration data set. (A) And (b) is a figure which illustrates the relationship between the state of a palm, and a curvature. It is a flowchart which illustrates the detail of an authentication process. (A) And (b) is a figure which illustrates the aspect ratio of a palm. (A) And (b) is a figure which illustrates the area of the palm.

  Hereinafter, embodiments will be described with reference to the drawings.

  FIG. 1A is a block diagram for explaining a hardware configuration of the biometric authentication device 100 according to the first embodiment. FIG. 1B is a schematic diagram of the biosensor 105 described later. With reference to FIG. 1A, the biometric authentication device 100 includes a CPU 101, a RAM 102, a storage device 103, a display device 104, a biometric sensor 105, a communication unit 106, an attribute information acquisition unit 107, and the like. Each of these devices is connected by a bus or the like.

  A CPU (Central Processing Unit) 101 is a central processing unit. The CPU 101 includes one or more cores. A RAM (Random Access Memory) 102 is a volatile memory that temporarily stores programs executed by the CPU 101, data processed by the CPU 101, and the like.

  The storage device 103 is a nonvolatile storage device. As the storage device 103, for example, a ROM (Read Only Memory), a solid state drive (SSD) such as a flash memory, a hard disk driven by a hard disk drive, or the like can be used. The biometric authentication program according to the present embodiment is stored in the storage device 103. The display device 104 is a liquid crystal display, an electroluminescence panel, or the like, and displays a result of each process described later.

  The biometric sensor 105 is a sensor that acquires user biometric information. In the present embodiment, the biometric sensor 105 acquires an image of the user's palm in a non-contact manner. As illustrated in FIG. 1B, the biosensor 105 is, for example, a CMOS (Complementary Metal Oxide Semiconductor) camera. For example, the biometric sensor 105 can acquire a palm print pattern, a palm shape, and the like based on visible light information. The biosensor 105 can also acquire a blood vessel pattern such as a vein pattern by using near infrared rays.

  The communication unit 106 is a connection interface to, for example, a LAN (Local Area Network). The attribute information acquisition unit 107 is an input device such as a keyboard and a mouse. For example, the attribute information acquisition unit 107 is a device for inputting an ID for identifying a user, a user name, a password, and the like.

  The biometric authentication program stored in the storage device 103 is expanded in the RAM 102 so as to be executable. The CPU 101 executes a biometric authentication program expanded in the RAM 102. Thereby, each process by the biometric authentication device 100 is executed. By executing the biometric authentication program, registration processing, authentication processing, and the like are executed.

  The registration process is a process of registering in the database biometric features obtained from the palm image acquired by the biometric sensor 105 in association with each user. In this embodiment, palm prints, vein patterns, palm shapes, etc. extracted from palm images are registered in the database as biometric features. The authentication process is a process of collating the biometric feature for collation acquired by the biometric sensor 105 with the registered biometric feature registered in the database. In this embodiment, as an example, if the similarity between the vein pattern acquired during the authentication process and the vein pattern registered in the database is equal to or greater than a threshold value, the user is the same person as the registered user. Determined. Details of the registration process and the authentication process will be described later.

  FIG. 2 is a block diagram of each function realized by executing the biometric authentication program. By executing the biometric authentication program, the overall management unit 11, storage unit 12, three-dimensional shape calculation unit 13, fitting unit 14, curvature calculation unit 15, feature extraction unit 16, registration unit 17, guidance control unit 18, authentication unit 19, Database 20 etc. are realized. In the example of FIG. 1 and FIG. 2, it is configured with a stand-alone terminal, but is not limited thereto. For example, the present embodiment can also be applied to a client server system. In this embodiment, an example of a stand-alone terminal will be described in order to simplify the description.

(registration process)
FIG. 3 is a flowchart illustrating details of the registration process. Hereinafter, the registration process will be described with reference to FIGS. 2 and 3. The overall management unit 11 controls each unit according to the flowchart of FIG. First, the attribute information acquisition unit 107 acquires user attribute information (step S1). Next, the biometric sensor 105 acquires, for registration, a palm image r of the user who holds his / her hand over the biometric sensor 105 (step S2). The user attribute information and the palm image r are temporarily stored in the storage unit 12.

  Next, the three-dimensional shape calculation unit 13 calculates the three-dimensional shape of the palm in the palm image r stored in the storage unit 12 (step S3). For the calculation of the three-dimensional shape, SFS (Shape From Shading) technology can be used. The SFS technique is a technique for recognizing a three-dimensional shape of a subject from a luminance distribution of a subject image taken by irradiating the subject with light. An example of calculation processing used in the SFS technique is disclosed in Non-Patent Documents 1 to 3, for example.

  Next, the fitting unit 14 fits the three-dimensional shape calculated in step S3 to a quadric surface (step S4). Next, the curvature calculator 15 calculates the curvature of the palm from the quadratic surface expression obtained by the fitting in step S4 (step S5). The curvature obtained here is a Gaussian curvature K. The Gaussian curvature K is a value defined at each point on the curved surface. The curvature calculation unit 15 acquires the average value of the obtained Gaussian curvature K as the curvature Kr at the time of registration processing. If the differential value for each point of the three-dimensional shape is obtained, the amount of calculation increases. On the other hand, by fitting each point of the three-dimensional shape to a quadric surface, the curvature of each point can be easily obtained using the parameters of the curved surface as in the following formulas (1) to (8). .

２次曲面上のガウス曲率Ｋは、下記式（２）に従って求めることができる。ただし、Ｅ〜Ｎは、下記式（３）〜（８）のように、フィッティング係数ａ〜ｆと、注目点のＸＹ座標値である（ｕ，ｖ）とから求まる。

First, the palm shape is fitted with a quadric surface of the following formula (1). (X, Y, Z) represents each coordinate of the palm. a to f represent fitting coefficients.

The Gaussian curvature K on the quadric surface can be obtained according to the following equation (2). However, E to N are obtained from the fitting coefficients a to f and the XY coordinate values (u, v) of the attention point as in the following formulas (3) to (8).

  Next, the feature extraction unit 16 extracts a biometric feature from the palm image r as a registered biometric feature (step S6). In this embodiment, a vein pattern is extracted as a registered biometric feature. The vein pattern includes vein end points, coordinates and directions of branch points, connection information of two branch points, and the like. Next, the registration unit 17 associates the attribute information acquired in Step S1, the curvature Kr calculated in Step S5, and the registered biometric feature extracted in Step S6, and registers them in the database 20 as a registration data set (Step S1). S7). The registration process is completed by the above process. FIG. 4 is a diagram illustrating a registered data set. A registered template corresponds to a registered biometric feature. In the example of FIG. 4, the registered biometric feature and the curvature Kr are registered in association with the ID of each user.

  Here, the relationship between the palm state and the curvature will be described. As illustrated in FIG. 5A, when the Gaussian curvature K is a positive value, the palm has a convex shape. That is, a positive sign of the Gaussian curvature K means that the palm is closed. Next, as illustrated in FIG. 5B, when the Gaussian curvature K is a negative value, the palm has a concave shape. That is, a negative sign of the Gaussian curvature K means that the palm is warped. By guiding the opening of the user's hand according to the difference between the curvature Kr and the curvature Ki obtained during the authentication process, the opening of the hand during the authentication process is made closer to the opening of the hand during the registration process. be able to.

(Authentication process)
First, FIG. 6 is a flowchart illustrating details of the authentication process. As illustrated in FIG. 6, the attribute information acquisition unit 107 acquires user attribute information (step S11). Next, the guidance control unit 18 acquires the curvature Kr of the registered data set corresponding to the attribute information acquired in step S11 (step S12). Next, the biometric sensor 105 acquires the user's palm image as the authentication image i (step S13). The authentication image i is temporarily stored in the storage unit 12.

  Next, the three-dimensional shape calculation unit 13 calculates the three-dimensional shape of the palm in the authentication image i stored in the storage unit 12. Next, the fitting unit 14 performs quadratic curved surface approximation on the calculated three-dimensional shape. Next, the curvature calculation unit 15 calculates the curvature Ki of the palm from the obtained quadric surface (step S14). Next, the guidance control unit 18 calculates a curvature difference ΔK = Ki−Kr as the degree of relative comparison between the curvature Kr and the curvature Ki (step S15). Next, the guidance control unit 18 determines whether or not ΔK exceeds the positive threshold value Th1 (step S16). When it is determined as “Yes” in Step S16, the guidance control unit 18 displays “Please open your palm” on the display device 104 (Step S17). Thereafter, the process is executed again from step S13.

  When it is determined as “No” in Step S16, the guidance control unit 18 determines whether ΔK is less than the negative threshold Th2 (Step S18). When it is determined as “Yes” in Step S18, the guidance control unit 18 displays “Please relax and open your hands” on the display device 104 (Step S19). Thereafter, the process is executed again from step S13. When it is determined as “No” in step S18, the feature extraction unit 16 extracts the biometric feature from the authentication image i as a biometric feature for matching (step S20).

  Next, the authentication unit 19 collates the biometric feature for collation with the registered biometric feature of the registration data set corresponding to the attribute information acquired in step S11, and displays the result on the display device 104 (step S21). . Specifically, the authentication unit 19 determines whether the similarity between the matching biometric feature and the registered biometric feature is equal to or greater than a threshold value, and causes the display device 104 to display the result.

  According to the present embodiment, since the guidance information is presented to the user based on the relative comparison between the curvature Kr and the curvature Ki, authentication can be performed after guiding the living body part according to each user. For example, by presenting guidance information that reduces the curvature difference ΔK, the curvature during the authentication process can be made closer to the curvature during the registration process. Thereby, individual differences in how to open the palm can be absorbed. For this reason, the user who is difficult to open the palm is not guided to open the palm more than necessary. As a result, the authentication accuracy is improved. Moreover, curvature calculation can be performed at high speed by approximating the three-dimensional shape data to a quadric surface.

  Note that it is determined whether or not the absolute value of the curvature difference ΔK is equal to or less than a threshold value by executing steps S16 and S18. This determination corresponds to a determination as to whether or not the degree of difference in relative comparison between the curvature Kr and the curvature Ki is equal to or less than a threshold value. In the above example, the curvature difference ΔK = Ki−Kr is used as the difference in the relative comparison between the curvature Kr and the curvature Ki, but is not limited thereto. For example, the ratio between the curvature Ki at the time of the authentication process and the curvature Kr at the time of the registration process may be used as the difference in relative comparison.

  An index other than the curvature may be used as the shape parameter of the living body part. For example, the aspect ratio of the palm (aspect ratio = length / width) may be used. For example, as illustrated in FIG. 7A, when the ratio of the length is large with respect to the ratio of the width (large aspect ratio), the hand is open. On the other hand, as illustrated in FIG. 7B, when the width ratio is large with respect to the length ratio (small aspect ratio), the hand is closed. Therefore, the first threshold value and the second threshold value (<first threshold value) are used to guide the user to close the hand when the aspect ratio exceeds the first threshold value, and to move the hand when the aspect ratio is lower than the second threshold value. Opening guidance may be performed. 7A and 7B are palm shapes that approximate an ellipse.

  The palm area may be used as a parameter of the shape of the living body part. For example, as illustrated in FIG. 8A, when the palm area is large, the hand is open. On the other hand, as illustrated in FIG. 8B, when the palm area is small, the hand is closed. Therefore, using the first threshold value and the second threshold value (<first threshold value), when the palm area exceeds the first threshold value, guidance for closing the hand is performed, and when the palm area is less than the second threshold value, You may be guided to open your hand. 8A and 8B are palm shapes that approximate an ellipse.

  In each of the above examples, the palm is used as the biometric part for authentication and the palm vein is used as the biometric feature. However, the present invention is not limited to this. For example, a finger may be used as a biometric part for authentication, and a finger vein may be used as a biometric feature. In this case, what is necessary is just to guide | induce the bending condition of a finger | toe.

  Further, in each of the above examples, 1: 1 authentication for specifying a registered biometric feature to be verified by acquiring attribute information at the time of authentication processing has been described, but the present invention is not limited thereto. The present embodiment can also be applied to 1: N authentication in which matching biometric features and a plurality of registered biometric features are collated without acquiring attribute information during authentication processing. In this case, since the attribute information is not acquired, the curvature Kr of a specific user cannot be used. Therefore, an average curvature <Kr> may be used for guidance. Here, the curvature <Kr> is an average value of the curvature of the user assumed by the biometric authentication device. In other words, by using the average curvature of a plurality of assumed users for guidance, the individuality of how to open the palm is reflected as much as possible. At this time, unlike the case of the 1: 1 authentication, it is impossible to reflect the opening method (curvature at the time of registration) of a specific individual, so the determination threshold values (Th1 and Th2 in FIG. 6) are set to 1: 1 authentication. It is desirable to set it more loosely.

  In each of the above examples, the database 20 functions as an example of a storage unit that stores the biological features of the living body part and the shape of the living body part. The biosensor 105 functions as an example of a biosensor that acquires an image of a user's living body part. The three-dimensional shape calculation unit 13 functions as an example of a calculation unit that calculates the shape of the biological part in the image acquired by the biological sensor. The guidance control unit 18 functions as an example of a presentation unit that presents guidance information based on a relative comparison between the shape calculated by the calculation unit and the shape stored in the storage unit to the user. The authentication unit 19 functions as an example of a collation unit that collates the biometric feature stored in the storage unit and the biometric feature in the image of the user's biometric part acquired by the biometric sensor after the guidance information is presented by the presentation unit.

  Although the embodiments of the present invention have been described in detail above, the present invention is not limited to such specific embodiments, and various modifications and changes can be made within the scope of the gist of the present invention described in the claims. It can be changed.

DESCRIPTION OF SYMBOLS 11 General management part 12 Memory | storage part 13 Three-dimensional shape calculation part 14 Fitting part 15 Curvature calculation part 16 Feature extraction part 17 Registration part 18 Guidance control part 19 Authentication part 20 Database 100 Biometric authentication apparatus 105 Biometric sensor

Claims (8)

A storage unit for storing the biological features of the biological part and the shape of the biological part;
A biosensor for acquiring an image of a user's biological site;
In the image acquired by the biological sensor, a calculation unit that calculates the shape of the biological part;
A presentation unit that presents guidance information to the user based on a relative comparison between the shape calculated by the calculation unit and the shape stored in the storage unit;
A collation unit that collates the biometric feature stored in the storage unit and the biometric feature in the image of the user's biometric part acquired by the biometric sensor after the guidance information is presented by the presentation unit. A biometric authentication device.   The collation unit includes the biometric feature stored in the storage unit and the biometric feature in the image of the biometric part of the user acquired by the biometric sensor when the degree of difference in the relative comparison is equal to or less than a threshold value. The biometric authentication device according to claim 1, wherein collation is performed.   The biometric authentication apparatus according to claim 1, wherein the calculation unit calculates a curvature as the shape of the biological part.   The biometric authentication apparatus according to claim 3, wherein the calculation unit obtains a curvature by fitting the biological part to a quadric surface.   The biometric authentication apparatus according to claim 1, wherein the biometric part is a palm.   The biometric authentication apparatus according to claim 1, wherein the shape stored in the storage unit is a shape obtained from a plurality of user shapes. The storage unit stores the biological features of the biological part and the shape of the biological part,
In the image of the user's biological part acquired by the biological sensor, the calculation unit calculates the shape of the biological part,
The presentation unit presents guidance information based on a relative comparison between the shape calculated by the calculation unit and the shape stored in the storage unit to the user,
The collation unit collates the biometric feature stored in the storage unit with the biometric feature in the image of the user's biological part acquired by the biometric sensor after the guidance information is presented by the presentation unit. Biometric authentication method. A process of storing the biological features of the biological part and the shape of the biological part;
In the user image acquired by the biological sensor, a process of calculating the shape of the biological part;
A process of presenting the user with guidance information based on a relative comparison between the calculated shape and the stored shape;
A biometric authentication program that executes a process of collating the stored biometric feature with a biometric feature in an image of the user's biometric part acquired by the biometric sensor after the guidance information is presented.

Images