KR102450128B1 - Apparatus for supporting biometrics using individual threshold and method for the same - Google Patents

Abstract

An embodiment of the present invention, the information acquisition unit for acquiring the user's input biometric information; a similarity determination unit for determining a degree of similarity between the registered biometric information corresponding to the user stored in a database and the input biometric information; an individual threshold management unit for setting individual thresholds corresponding to the user by using the similarity determination record; and a biometric authentication unit configured to perform biometric authentication of the user using the similarity and the individual threshold.

Description

Biometric recognition support device and method using individual thresholds

The present invention relates to a biometric recognition support apparatus and method using individual thresholds set for each user. In detail, the present invention relates to a biometric recognition support apparatus and method for setting an individual threshold in consideration of a biometric record for each user, and providing a suitable biometric function for each user by using the set individual threshold.

Currently, in order to use a biometric system such as fingerprint recognition, face recognition, iris recognition, and vein recognition, it is essential for a user to pre-enroll through input of a plurality of correct templates (gallery images). In the personal authentication process, a similarity score value calculated through matching between a template (probe image) input in real time for authentication together with an individual's identifier (ID) and a pre-registered template is set to a predetermined global threshold. ) and the authentication method is used. At this time, the limit is to obtain a similarity score from the test data of the similarity (genuine) between biometric information of the same person and the similarity (imposter) between biometric information of the same person by using a biometric test database in advance, and FMR (False Match Rate) ) and FNMR (False Non-Match Rate) distribution.

However, this method inevitably requires an unnatural and inconvenient user registration process that requires a lot of time, and also requires the same process to update registration information, so that it is difficult to update after user registration. In addition, since global thresholds are used, there is a problem in that authentication performance is sensitively changed according to thresholds when the deviation of similarity values is not large, and various types (sensor, resolution, age , pose, lighting, etc.) has a problem that the performance sensitively responds to the input template by the limit point.

Therefore, there is a need for an authentication method that can simplify the user registration process, use individual thresholds for each user, and update individual thresholds according to biometric records.

The above-mentioned background art is technical information possessed by the inventor for the derivation of the present invention or acquired in the process of derivation of the present invention, and cannot necessarily be said to be a known technique disclosed to the general public prior to the filing of the present invention.

Domestic Patent Publication No. 10-2017-0003348

SUMMARY OF THE INVENTION It is an object of the present invention to provide a biometric recognition support apparatus and method for performing biometric recognition using individual threshold values set separately for each user to perform biometric authentication.

Another object of the present invention is to provide a biometric support apparatus and method for managing registration information of a new user by performing a registration procedure for a new user in the same manner as the biometric identification procedure.

An embodiment of the present invention, the information acquisition unit for acquiring the user's input biometric information; a similarity determination unit for determining a degree of similarity between the registered biometric information corresponding to the user stored in a database and the input biometric information; an individual threshold management unit for setting individual thresholds corresponding to the user by using the similarity determination record; and a biometric authentication unit configured to perform biometric authentication of the user using the similarity and the individual threshold.

At this time, the individual threshold management unit considers at least one of the time-series change in similarity from the similarity determination record, the frequency of occurrence of outliers, the change in individual thresholds, the similarity average, and the similarity standard deviation to determine the individual threshold. can be set.

In this case, the method may further include a feature information extracting unit for extracting feature information from the user's input biometric information, and the similarity determining unit may determine a similarity between the registered biometric information and the input biometric information using the feature information.

In this case, the apparatus may further include a similarity filtering unit configured to verify the appropriateness of the similarity by using the similarity determination record to filter the similarity, and to determine whether an abnormal user is present according to the result of the similarity filtering.

In this case, the individual threshold management unit may not use the filtered similarity to set the individual threshold because the appropriateness verification has failed.

In this case, the apparatus may further include a registration information management unit for managing registration information for users whose registration has not been completed and users whose registrations have been completed, wherein the registration information may include the registered biometric information for the user and the individual limit points.

In this case, the registration information management unit may update the registration information stored in the database by using the input biometric information upon the user's request or when a preset update cycle arrives.

In this case, when the registration information management unit does not perform the biometric information registration process of the ID for the user, the input ID may be linked with the database and the input biometric information may be registered in the ID.

In this case, the registration information management unit may request that the ID registration procedure be performed using the input biometric information when the ID for the user has performed the biometric information registration process but the registration process has not yet been completed.

In this case, the similarity determining unit determines the similarity between the registered biometric information and the input biometric information when performing the ID registration procedure, and the similarity filtering unit determines the appropriateness of the similarity using the similarity determination record when performing the ID registration procedure. and the individual threshold management unit sets the individual thresholds using the similarity determination record when performing the ID registration procedure, and the registration information management unit verifies the appropriateness of the similarity by determining the similarity when performing the ID registration procedure The registration information may be updated according to a result and the individual threshold, and whether or not registration of the ID is completed may be set according to whether a registration condition is satisfied.

Another embodiment of the present invention comprises the steps of: obtaining user's input biometric information; determining a degree of similarity between registered biometric information corresponding to the user stored in a database and the input biometric information; setting individual thresholds corresponding to the user by using the similarity determination record; and performing biometric authentication for the user using the similarity and the individual threshold.

In this case, the step of setting the individual thresholds is performed in consideration of at least one of a time-series change in similarity from the similarity determination record, the frequency of occurrence of outliers, a change in individual thresholds, a similarity average, and a similarity standard deviation. Individual thresholds can be set.

In this case, the method may further include extracting feature information from the user's input biometric information, and determining the similarity may determine a similarity between the registered biometric information and the input biometric information using the feature information. .

In this case, filtering the similarity by verifying the appropriateness of the similarity using the similarity determination record; and performing the biometric authentication may further include determining whether the user is an abnormal user according to a result of the similarity filtering.

In this case, in the step of setting the individual threshold, the filtered similarity may not be used to set the individual threshold because the validity verification fails.

In this case, the method may further include managing registration information for a user whose registration has not been completed and a user whose registration has been completed, wherein the registration information may include the registered biometric information for the user and the individual limit points.

In this case, the managing of the registration information may include updating the registration information stored in the database by using the input biometric information upon request of the user or upon arrival of a preset update cycle.

In this case, in the step of managing the registration information, when the biometric information registration process of the ID for the user is not performed, the input ID may be linked with the database and the input biometric information may be registered in the ID.

In this case, in the step of managing the registration information, when the ID for the user has performed the biometric information registration process but the registration process is not yet completed, the ID registration process may be requested using the input biometric information.

In this case, the determining of the similarity determines the similarity between the registered biometric information and the input biometric information when performing the ID registration procedure, and the filtering of the similarity uses the similarity determination record when performing the ID registration procedure to verify the adequacy of the degree of similarity, and the step of setting the individual threshold sets the individual threshold using the similarity determination record when performing the ID registration procedure, and the step of managing the registration information includes the ID registration procedure The registration information may be updated according to the determination result of the similarity degree, the appropriateness verification result of the similarity degree, and the individual threshold points, and whether or not the registration of the ID is completed may be set according to whether the registration condition is satisfied.

The present invention can ensure reliable biometric performance for each of several users by using an individual threshold set separately for each user by using an apparatus and method for supporting biometrics using individual thresholds.

In addition, the present invention provides a biometric recognition support device and method using individual thresholds to manage registration information of new users by performing a registration procedure for a new user in the same manner as for a biometric identification procedure, which is inconvenient and saves a lot of time. The user's convenience can be improved by simplifying the required pre-registration process.

1 is a diagram showing the configuration of a biometric recognition system using individual limitations according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating an example of an apparatus for supporting biometrics using the individual thresholds shown in FIG. 1 .
3 is a conceptual diagram illustrating a biometric recognition support method using individual limitations according to an embodiment of the present invention.
4 is a flowchart illustrating a biometric recognition support method using individual thresholds according to an embodiment of the present invention.
FIG. 5 is an operation flowchart illustrating an example of a step of performing the user registration procedure shown in FIG. 4 .
6 is an operation flowchart illustrating an example of a step of performing the user authentication procedure shown in FIG. 4 .

Since the present invention can apply various transformations and can have various embodiments, specific embodiments are illustrated in the drawings and described in detail. Effects and features of the present invention, and a method for achieving them, will become apparent with reference to the embodiments described below in detail in conjunction with the drawings. Here, repeated descriptions, well-known functions that may unnecessarily obscure the gist of the present invention, and detailed descriptions of configurations will be omitted. The embodiments of the present invention are provided in order to more completely explain the present invention to those of ordinary skill in the art. Accordingly, the shapes and sizes of elements in the drawings may be exaggerated for clearer description.

However, the present invention is not limited to the embodiments disclosed below, but all or some of the embodiments may be selectively combined and implemented in various forms. In the following embodiments, terms such as first, second, etc. are used for the purpose of distinguishing one component from another, not in a limiting sense. Also, the singular expression includes the plural expression unless the context clearly dictates otherwise. In addition, terms such as include or have means that the features or components described in the specification are present, and do not preclude the possibility that one or more other features or components will be added.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, and when described with reference to the drawings, the same or corresponding components are given the same reference numerals, and the overlapping description thereof will be omitted. .

1 is a diagram showing the configuration of a biometric recognition system 100 using individual limitations according to an embodiment of the present invention.

Referring to FIG. 1 , in the biometric recognition system 100 using individual thresholds according to an embodiment of the present invention, the biometric recognition support device 110 using individual thresholds is interconnected with the biometric information acquisition device 120 .

The biometric recognition support apparatus 110 using individual thresholds according to an embodiment of the present invention obtains biometric information about a user and determines a similarity between registered biometric information corresponding to the user stored in a database and input biometric information. And, an individual threshold corresponding to the user is set using the similarity determination record, and biometric authentication for the user is performed using the similarity and the individual threshold.

Here, the biometric authentication may include various biometric authentication methods such as face recognition, iris recognition, fingerprint recognition, voiceprint recognition, and blood vessel recognition.

In this case, user ID information may be acquired as user identification information as well as biometric information about the user.

That is, the present invention does not determine whether or not to authenticate the pre-registered biometric information and the similarity (matching score) of the biometric information input in real time for authentication based on one global threshold, but time-series the record of determination of the similarity for each user. By setting and updating individual thresholds through systematic analysis, the recognition performance can be improved by applying appropriate biometric criteria for each user.

In an optional embodiment, the biometric recognition support device 110 using individual limitations according to an embodiment of the present invention obtains the user's biometric information and ID through the same procedure as the authentication process from the general user's point of view, If there is no registered information, the input biometric information is registered in the database. Thereafter, the similarity is determined through comparison with biometric information input in the authentication process for the same user, and the registration information can be updated by setting individual thresholds according to similarity filtering and analysis of the similarity determination record.

In other words, convenience and efficiency are increased by simplifying and automating the user pre-registration process essential in the biometric system to the same procedure as the authentication process, and setting and updating individual thresholds by analyzing time series data of similarity calculated for each authentication process. Thus, it is possible to provide a more efficient and high-performance biometric recognition function.

The biometric information acquisition device 120 is a device for acquiring biometric information to be used for user biometric recognition.

For example, the biometric information acquisition device 120 includes at least one black and white camera, a color camera, an RGB-D camera, an infrared camera, a constant voltage sensor, a pressure sensor, a touch sensor, a microphone, and an ultrasonic sensor to be used for biometric recognition. information can be obtained.

In this case, the biometric information acquisition device 120 may be implemented as a separate device that is distinct from the biometric recognition support device 110 using individual threshold points, but is implemented as a single component of the biometric recognition support device 110 using individual threshold points. it might be

FIG. 2 is a block diagram illustrating an example of the biometric recognition support apparatus 110 using the individual limit points shown in FIG. 1 .

Referring to FIG. 2 , the biometric recognition support apparatus 110 using individual thresholds according to an embodiment of the present invention includes a control unit 210 , a communication unit 215 , a memory 220 , an information acquisition unit 225 , and feature information. It includes an extraction unit 230 , a registration information management unit 235 , a similarity determination unit 240 , an individual threshold management unit 245 , a similarity filtering unit 250 , a biometric authentication unit 255 , a database 260 , and the like.

In detail, the control unit 210 controls the entire process of biometric recognition using individual thresholds as a kind of central processing unit. That is, the control unit 210 includes an information acquisition unit 225 , a feature information extraction unit 230 , a registration information management unit 235 , a similarity determination unit 240 , an individual threshold management unit 245 , a similarity filtering unit 250 , Various functions may be provided by controlling the biometric authentication unit 255 and the database 260 .

Here, the controller 210 may include all kinds of devices capable of processing data, such as a processor. Here, the 'processor' may refer to a data processing device embedded in hardware, for example, having a physically structured circuit to perform a function expressed as a code or an instruction included in a program. As an example of the data processing device embedded in the hardware as described above, a microprocessor, a central processing unit (CPU), a processor core, a multiprocessor, an application-specific integrated (ASIC) circuit) and a processing device such as a field programmable gate array (FPGA), but the scope of the present invention is not limited thereto.

The communication unit 215 provides a communication interface necessary to transmit a transmission/reception signal between the biometric recognition support device 110 and the biometric information acquisition device (see 120 of FIG. 1 ) using individual thresholds.

Here, the communication unit 215 may be a device including hardware and software necessary for transmitting and receiving signals such as control signals or data signals through wired/wireless connection with other network devices.

In this case, the communication unit 215 may receive the biometric authentication request from the biometric authentication request device (not shown) and transmit the biometric authentication request result.

The memory 220 performs a function of temporarily or permanently storing data processed by the controller 210 . Here, the memory 220 may include magnetic storage media or flash storage media, but the scope of the present invention is not limited thereto.

The information acquisition unit 225 acquires the user's biometric information.

In this case, the information acquisition unit 225 may acquire the user's biometric information acquired by the biometric information acquisition device (see 120 of FIG. 1 ).

In this case, the information acquisition unit 225 may acquire the ID as identification information together with the user's biometric information.

Here, the biometric information may include face information, voice information, blood vessel information, iris information, fingerprint information, and the like.

The feature information extraction unit 230 extracts feature information from the user's biometric information.

In this case, the feature information extraction unit 230 may extract feature information from the registered biometric information already stored in the database 260 and biometric information to be registered in the database 260 .

Here, the feature information may refer to selected information describing the biometric information used to determine the consistency between the biometric information.

In this case, the feature information extraction unit 230 may extract feature information from the input biometric information to generate an input template, and extract feature information from biometric information for registration or registered biometric information to generate a registration template.

The registration information management unit 235 manages registration information (or authentication data) for a user whose registration has not been completed and a user who has already been registered.

Here, the registration information may include at least one or more of the user's biometric information, the user's ID, the authority configuration, and the user's individual limit points.

In this case, the registration information management unit 235 may update the registration information corresponding to the user by using the biometric information input by the user when there is a user's request for the already registered user or when a preset update cycle arrives.

That is, the registration information management unit 235 may update the user's registration information every specific period (eg, one year), and accordingly, the user's physical change over time may be reflected in the registration information.

In this case, the registration information management unit 235 may link the ID for the user with the database and register the input biometric information in the ID if the ID for the user who is the target of biometric recognition is an ID that has not yet been registered. have. Here, when the ID is input together with the user's biometric information, the input ID may be linked with the corresponding user.

That is, a user whose registration has not been completed can input biometric information for biometric recognition like a registered user, and the registration information management unit 235 automatically or in advance personnel information or personal number for a user without an ID. The initial registration procedure can be performed by generating an ID through interlocking, etc. and notifying the user so that input biometric information can be registered. However, ID registration may not be completed only by generating an ID and registering input biometric information.

In this case, the registration information management unit 235 may set a preset global limit value as an initial individual limit value when the ID is initially linked and biometric information is registered.

In this case, the registration information management unit 235 may request that the ID registration procedure be performed using the input biometric information when the ID of the user who is the target of biometric recognition is linked but the registration procedure is not yet completed.

At this time, the ID registration procedure compares the registered biometric information with the input biometric information to determine the similarity similarly to the already registered user's biometric authentication process, verifies the adequacy of the similarity using the similarity determination record, and performs similarity filtering; This may mean setting individual thresholds using the similarity determination record, and updating registration information according to the similarity determination result, similarity adequacy verification result, and individual threshold. Furthermore, when the registration information is updated, it may include setting whether the registration of the corresponding ID is completed according to whether the registration condition is satisfied. Here, the registration condition may include conditions such as, for example, that the variation in the degree of similarity from the similarity determination record is less than or equal to a reference value, a predetermined number or more of registration information is accumulated, or the time of the most recently registered registration information is within a predetermined period.

That is, even for a user who has not been registered, biometric information to be used for automatically updating registered information by obtaining biometric information in the same manner as a registered user, determining the similarity, performing similarity filtering, and setting individual thresholds It is possible to update the registration information by determining whether or not it is recognized.

The similarity determining unit 240 determines the similarity between the input biometric information and the registered biometric information.

In this case, the similarity determining unit 240 may determine the similarity by comparing the characteristic information of the input biometric information with the characteristic information of the registered biometric information.

In this case, the similarity determining unit 240 may calculate a similarity score between the input biometric information and the registered biometric information by using a similarity comparison algorithm such as Hamming distance.

The individual threshold management unit 245 sets individual threshold values to be used for biometric recognition for each user. Here, the setting of the individual limit values includes updating of the individual limit values already set.

In this case, the individual threshold management unit 245 may set individual threshold values using the similarity determination record.

At this time, the individual threshold management unit 245 may set individual threshold values in consideration of the time-series change in similarity, the frequency of occurrence of outliers, the change in individual thresholds, the similarity average and standard deviation from the similarity determination record. have.

For example, when the similarity according to a specific user's biometrics shows a trend of continuously increasing, the individual thresholds may be adjusted upwards to be set.

The similarity filtering unit 250 filters the similarity by verifying the appropriateness of the determined similarity. That is, this is to determine whether the degree of similarity determined from the input biometric information corresponds to an outlier, and, if necessary, to exclude it.

In this case, the similarity filtering unit 250 may verify the adequacy of the determined similarity by using the similarity determination record.

In this case, the similarity filtering unit 250 may verify the appropriateness of the determined similarity using similarity factors and correlations, similarity change trends, similarity averages and standard deviations, individual thresholds, and the like.

In this case, when an abnormality is detected in the appropriateness of the similarity, the similarity filtering unit 250 may perform an abnormal user detection process for the corresponding user. If an abnormal user is detected, an action may be taken according to the abnormal user detection, which may include raising individual thresholds, denying access to the corresponding ID, or sending a message to an administrator. If it is determined that the user is not an abnormal user, a biometric authentication procedure may be performed through the biometric authentication unit 255 .

The biometric authentication unit 255 performs biometric authentication for the user.

In this case, the biometric authentication unit 255 may perform biometric authentication using the similarity determined from the user's input biometric information and the user's individual limit value.

In this case, the biometric authentication unit 255 may return an authentication result according to the biometric authentication.

In this case, the biometric authentication unit 255 may grant a preset authority to the corresponding user when authentication is successful.

The database 260 stores registration information required for biometric authentication.

In this case, the database 260 may store registered biometric information, user ID, individual limit values and access rights set for each user as registration information.

In this case, the database 260 may store feature information extracted from the user's registered biometric information.

In this case, the database 260 may store a similarity determination record (similarity determination history) for each user.

Here, there is no limit to the number of objects of registration information stored by the database 260 .

In this case, the database 260 may store a user biometric log, and the user biometric log may include an authentication attempt time, whether authentication is allowed, similarity, error information, and the like.

3 is a conceptual diagram illustrating a biometric recognition support method using individual limitations according to an embodiment of the present invention.

Referring to FIG. 3 , in the method for supporting biometrics using individual limitations according to an embodiment of the present invention, a new ID is first linked ( 301 ). This may mean setting a new ID by linking an ID and pre-prepared personal information based on a template and ID information that is input in real time or obtained for an ID in which biometric information is not registered, and registering biometric information.

Then, the initial individual threshold is set (302), and the parameters required for authentication, such as thresholds and templates, according to the renewal conditions of predefined authentication variables, in consideration of the characteristics of the biometric information used and the degree of security according to the system application field, etc. and data are set or adjusted (303). At this time, if the global threshold is set in the system, the initial individual threshold may be set as the global threshold based on this.

Then, the similarity between the registered biometric information (or the registration template) and the input biometric information (or the input template) is calculated ( 304 ). In this case, a score indicating the degree of similarity may be calculated between the registration template and the input template using a similarity comparison algorithm such as Hamming distance.

Then, similarity filtering is performed by verifying the adequacy of similarity through analysis of time series data such as the calculated similarity, change trend of similarity, individual thresholds, similarity mean, and standard deviation ( 305 ). Here, in order to perform similarity filtering, a similarity filtering variable necessary for similarity adequacy verification may be created or set.

Then, when an abnormality is detected in the appropriateness of the similarity according to the similarity filtering, whether the user is an abnormal user through high-level time series data analysis and usage time tracking, etc. is detected (306).

Then, user authentication is performed according to whether the calculated similarity and individual thresholds are satisfied ( 307 ). In this case, when the calculated similarity is greater than or equal to the individual threshold, it can be considered that the individual threshold is satisfied.

Then, the current individual threshold is updated according to the analysis of the similarity determination record, the similarity change trend, the outlier occurrence frequency, the individual threshold change trend, and the like ( 308 ).

In order to keep the data necessary for authentication, such as individual thresholds, registered biometric information, and registration template, up to date, the update cycle and update items are determined automatically or manually (309)

4 is a flowchart illustrating a biometric recognition support method using individual thresholds according to an embodiment of the present invention.

Referring to FIG. 4 , in the method for supporting biometrics using individual thresholds according to an embodiment of the present invention, a biometric support apparatus (refer to 110 of FIG. 1 ) using individual thresholds uses a biometric input biometric of a user to be subjected to biometric recognition. Obtain information and ID (S401). Here, the ID may mean an ID corresponding to the biometric information of the user registered in the database.

In this case, the input biometric information and ID may be obtained by the biometric information acquiring device (refer to 120 of FIG. 1 ).

In addition, in the biometric recognition support method using individual threshold points according to an embodiment of the present invention, the biometric recognition support device (refer to 110 in FIG. 1 ) using individual threshold points extracts feature information from input biometric information to generate an input template. (S403).

In this case, extraction of the feature information may be performed in the same manner as the registration template stored in the database.

In addition, in the method for supporting biometrics using individual thresholds according to an embodiment of the present invention, the biometrics support apparatus (refer to 110 in FIG. 1 ) using individual thresholds determines whether a user to be recognized is a registered user. do (S405). Here, the registered user may mean a user whose registration has been completed and registered.

As a result of the determination in step S405, if the user is not a registered user, a user registration procedure is performed (S407).

As a result of the determination in step S405, in the case of a registered user, a user authentication procedure is performed (S409).

FIG. 5 is an operation flowchart illustrating an example of a step S407 of performing the user registration procedure shown in FIG. 4 .

Referring to FIG. 5 , in the step S407 of performing the user registration procedure shown in FIG. 4 , the biometric recognition support device (refer to 110 of FIG. 1 ) using individual thresholds determines whether the ID of the corresponding user is an already linked ID. do (S501).

As a result of the determination in step S501, if the ID has not yet been linked, the input ID is linked with the database (S503).

In addition, in the step of performing the user registration procedure shown in FIG. 4 ( S407 ), the biometric recognition support device (see 110 in FIG. 1 ) using individual thresholds registers the input biometric information acquired in the newly linked ID ( S505 ). ).

In this case, the user's personal information may be linked together with the newly linked ID in advance.

In this case, it may indicate that the newly linked ID is not a correct ID.

As a result of the determination in step S501, if the ID is already linked, the similarity between the input template corresponding to the input biometric information and the registration template corresponding to the registered biometric information is determined (S507).

In addition, in the step S407 of performing the user registration procedure shown in FIG. 4 , the biometric recognition support device (refer to 110 in FIG. 1 ) using individual thresholds verifies the adequacy of the similarity determined using the similarity determination record to filter the similarity. to perform (S509). That is, it is determined whether the similarity determined through the similarity filtering is in an appropriate region.

In addition, in the step of performing the user registration procedure shown in FIG. 4 ( S407 ), the biometric recognition support device (see 110 in FIG. 1 ) using the individual threshold determines the similarity and the similarity determination record to the individual corresponding to the user. A threshold is set (S511).

In this case, only when it is determined that the similarity is in an appropriate region through similarity filtering, the corresponding similarity may be used to set an individual threshold.

In addition, in the step of performing the user registration procedure shown in FIG. 4 ( S407 ), the biometric recognition support device (refer to 110 of FIG. 1 ) using individual thresholds updates authentication data ( S513 ).

In this case, the authentication data may include registered biometric information, registration templates, similarity determination records, individual thresholds, and the like.

In addition, in the step (S407) of performing the user registration procedure shown in FIG. 4, the biometric recognition support device (refer to 110 in FIG. 1) using the individual threshold determines whether the ID registration condition is satisfied (S515), and the ID The ID registration completion setting is performed for the ID in which the registration condition is satisfied (S517).

In this case, the ID registration condition may include a condition that the number of registration templates or accumulated registration templates is greater than or equal to a certain number, the variation in the similarity is within a reference value, or the update time of the most recent registered biometric information is within a predetermined period.

Accordingly, a user who has not yet been registered can input biometric information in the same way as a registered user, and can easily and automatically register biometric information without performing a separate pre-registration procedure.

6 is an operation flowchart illustrating an example of a step S409 of performing the user authentication procedure shown in FIG. 4 .

Referring to FIG. 6 , in the step S409 of performing the user authentication procedure shown in FIG. 4 , the biometric recognition support device (refer to 110 in FIG. 1 ) using individual thresholds uses an input template corresponding to the input biometric information and a registered biometric information. A degree of similarity between registration templates corresponding to the information is determined (S601).

In addition, in the step of performing the user authentication procedure shown in FIG. 4 ( S409 ), the biometric recognition support device (refer to 110 in FIG. 1 ) using individual thresholds verifies the adequacy of the similarity determined using the similarity determination record to filter the similarity. to perform (S603). That is, it is determined whether the similarity determined through the similarity filtering is in an appropriate region.

In addition, in the step of performing the user authentication procedure shown in FIG. 4 (S409), the biometric recognition support device (refer to 110 in FIG. 1) using individual thresholds detects an abnormal user using the similarity filtering result (S605), It is determined whether the corresponding user is an abnormal user (S607).

For example, when an abnormality is detected in the similarity change from the similarity determination record, an abnormal user may be detected.

As a result of the determination in step S607, if the corresponding user is determined to be an abnormal user, a corresponding action is taken for the abnormal user (S609), and an authentication result indicating that the abnormal user has been accessed is returned (S617). Here, the countermeasures for the abnormal user may include blocking access of the corresponding user ID, raising individual limit values, non-setting of authority, and sending a message to an administrator.

As a result of the determination in step S607, if it is determined that the corresponding user is not an abnormal user, biometric authentication is performed (S611), and it is determined whether the authentication has been successful (S613).

In this case, biometric authentication may be performed by comparing the previously determined similarity with individual thresholds.

As a result of the determination in step S613, if authentication is not successful, an authentication result indicating that authentication has failed is returned (S617).

As a result of the determination in step S613, if the authentication is successful, the individual thresholds corresponding to the corresponding user are updated using the determined similarity and similarity determination records (S615), and the authentication result indicating that the authentication is successful is returned (S617).

In this case, the authentication result indicating that the authentication is successful may include a command for granting a preset authority.

If there is a user's update request or a preset update period has elapsed or arrives, the registration information (authentication data) stored in the database may be updated.

As described above, since biometric recognition is performed using individual thresholds for each registered user, high-accuracy recognition performance can be provided for each user. In addition, by updating individual thresholds using the results of repeated biometrics and updating registration information stored in the database, it is possible to effectively solve the problem of poor biometric recognition due to changes in the user's body.

The specific implementations described in the present invention are examples and do not limit the scope of the present invention in any way. For brevity of the specification, descriptions of conventional electronic components, control systems, software, and other functional aspects of the systems may be omitted. In addition, the connections or connecting members of the lines between the components shown in the drawings exemplify functional connections and/or physical or circuit connections, and in an actual device, various functional connections, physical connections that are replaceable or additional may be referred to as connections, or circuit connections. In addition, unless there is a specific reference such as “essential” or “importantly”, it may not be a necessary component for the application of the present invention.

Therefore, the spirit of the present invention should not be limited to the above-described embodiments, and the scope of the spirit of the present invention is not limited to the scope of the scope of the present invention. will be said to belong to

100: biometric system using individual thresholds
110: biometric recognition support device using individual thresholds
120: biometric information acquisition device
210: control unit 215: communication unit
220: memory 225: information acquisition unit
230: feature information extraction unit 235: registration information management unit
240: similarity determination unit 245: individual limit point management unit
250: similarity filtering unit 255: biometric authentication unit
260: database management unit

Claims (20)

an information acquisition unit for acquiring the user's input biometric information;
a similarity determination unit for determining a degree of similarity between the registered biometric information corresponding to the user stored in a database and the input biometric information;
Individual thresholds for setting individual thresholds corresponding to the user in consideration of at least one of the time-series change in similarity from the similarity determination record, the frequency of occurrence of outliers, the change in individual thresholds, the similarity average, and the similarity standard deviation Management; and
A biometric authentication unit that performs biometric authentication for the user by using the similarity and the individual threshold
including,
The individual limits are
A biometric authentication support device using individual thresholds, characterized in that it is set and updated to correspond to each user based on a time-series analysis of the similarity determination record for each user. delete The method according to claim 1,
A feature information extraction unit for extracting feature information from the user's input biometric information
further comprising,
The similarity determining unit
and determining a degree of similarity between the registered biometric information and the input biometric information by using the characteristic information. 4. The method of claim 3,
A similarity filtering unit for filtering the similarity by verifying the appropriateness of the similarity using the similarity determination record, and determining whether an abnormal user is present according to the result of the similarity filtering
Biometric authentication support device using individual thresholds, characterized in that it further comprises. 5. The method according to claim 4,
The individual threshold management unit
The apparatus for supporting biometric authentication using individual thresholds, characterized in that the similarity filtered due to failure of the appropriateness verification is not used to set the individual thresholds. 5. The method according to claim 4,
Registration information management unit that manages registration information for unregistered users and registered users
further comprising,
The registration information is
The biometric authentication support apparatus using individual thresholds, characterized in that it includes the registered biometric information for the user and the individual thresholds. 7. The method of claim 6,
The registration information management unit
and updating the registration information stored in the database by using the input biometric information when there is a request from the user or when a preset update cycle arrives. 7. The method of claim 6,
The registration information management unit
When the biometric information registration process of the ID for the user is not performed, the biometric authentication support apparatus using an individual limit point is characterized in that the input ID is linked with the database and the input biometric information is registered in the ID. 9. The method of claim 8,
The registration information management unit
When the ID for the user has performed the biometric information registration process but the registration process is not yet completed, the biometric authentication support device using individual thresholds, characterized in that it requests the ID registration process using the input biometric information . 10. The method of claim 9,
The similarity determining unit
determining a similarity between the registered biometric information and the input biometric information when performing the ID registration procedure;
The similarity filtering unit
verifying the adequacy of the similarity level using the similarity determination record when performing the ID registration procedure;
The individual threshold management unit
When performing the ID registration procedure, the individual threshold is set using the similarity determination record,
The registration information management unit
When the ID registration procedure is performed, the registration information is updated according to the determination result of the similarity degree, the adequacy verification result of the similarity degree, and the individual limit points, and whether the registration of the ID is completed is set according to whether the registration conditions are satisfied. , biometric authentication support devices using individual thresholds. obtaining user's input biometric information;
determining a similarity between the registered biometric information corresponding to the user stored in a database and the input biometric information;
setting an individual threshold corresponding to the user in consideration of at least one of time-series similarity change trend, outlier occurrence frequency, individual threshold change trend, similarity average, and similarity standard deviation from the similarity determination record; and
performing biometric authentication for the user using the similarity and the individual threshold
including,
The individual limits are
A biometric recognition support method using individual thresholds, characterized in that it is set and updated to correspond to each user based on a time-series analysis of the similarity determination record for each user. delete 12. The method of claim 11,
extracting feature information from the user's input biometric information
further comprising,
The step of determining the similarity
and determining a degree of similarity between the registered biometric information and the input biometric information by using the characteristic information. 14. The method of claim 13,
filtering the similarity by verifying appropriateness of the similarity using the similarity determination record; and
determining whether the user is an abnormal user according to the result of the similarity filtering;
Biometric recognition support method using individual thresholds, characterized in that it further comprises. 15. The method of claim 14,
The step of setting the individual threshold is
The biometric recognition support method using individual thresholds, characterized in that the filtered similarity level due to failure of the appropriateness verification is not used to set the individual thresholds. 15. The method of claim 14,
Steps to manage registration information for unregistered users and registered users
further comprising,
The registration information is
The biometric recognition support method using individual thresholds, characterized in that it includes the registered biometric information for the user and the individual thresholds. 17. The method of claim 16,
The step of managing the registration information is
The method for supporting biometrics using individual thresholds, characterized in that the registration information stored in the database is updated using the input biometric information when there is a request from the user or when a preset update cycle arrives. 17. The method of claim 16,
The step of managing the registration information is
If the biometric information registration process of the ID for the user is not performed, the input ID is linked with the database and the input biometric information is registered in the ID. 19. The method of claim 18,
The step of managing the registration information is
When the ID for the user has performed the biometric information registration process but the registration process has not yet been completed, the biometric recognition support method using individual thresholds, characterized in that the ID registration process is requested using the input biometric information . 20. The method of claim 19,
The step of determining the similarity
determining a similarity between the registered biometric information and the input biometric information when performing the ID registration procedure;
The step of filtering the similarity is
verifying the adequacy of the similarity level using the similarity determination record when performing the ID registration procedure;
The step of setting the individual threshold is
When performing the ID registration procedure, the individual threshold is set using the similarity determination record,
The step of managing the registration information is
When the ID registration procedure is performed, the registration information is updated according to a result of determining the degree of similarity, an adequacy verification result of the degree of similarity, and the individual threshold, and whether or not registration of the ID is completed is set according to whether a registration condition is satisfied , a biometric support method using individual thresholds.

Images