KR20230082978A - Biometric authentication apparatus and method - Google Patents

Abstract

The present invention relates to an apparatus and a method for biometric authentication. The apparatus comprises: an electrocardiogram measurement module which measures an electrocardiogram of a subject; a registration module which collects a plurality of electrocardiogram signals through an electrocardiogram measurement result of the electrocardiogram measurement module for the subject, determines a first expected electrocardiogram signal expected to represent the plurality of collected electrocardiogram signals, and determines a representative electrocardiogram signal functioning as reference data for biometric authentication of the subject and registers the representative electrocardiogram signal, by using a method of analyzing a correlation between the determined first expected electrocardiogram signal and the plurality of collected electrocardiogram signals, and a method of verifying the validity of the first expected electrocardiogram signal; and an authentication module which performs biometric authentication of the subject by using a method of analyzing a correlation between an actual electrocardiogram signal currently obtained from the electrocardiogram measurement result of the electrocardiogram measurement module and the representative electrocardiogram signal registered by the registration module. The present invention can improve the accuracy and reliability of biometric authentication.

Description

Biometric authentication device and method {BIOMETRIC AUTHENTICATION APPARATUS AND METHOD}

The present invention relates to a biometric authentication device and method, and more particularly, to a biometric authentication device and method for performing biometric authentication of a subject using an electrocardiogram.

Biometric authentication refers to confirming and authenticating an individual's identity through physical characteristics of an individual, and such biometric authentication is applied to a wide range of fields such as personal information protection, financial security, access control, and attendance management. Furthermore, biometric authentication is being applied to an electronic supervision system to supervise and control criminals such as sexual violence, stalking, or domestic violence, and its application field is expanding, and as part of the above electronic supervision system, There is an increasing number of people under surveillance who are subject to restrictions on going out at night, confining them to the inside.

The conventional monitoring system for the person who imposes the night out restriction order identifies the subject of surveillance through voiceprint recognition (voice recognition), so it is accompanied by problems of frequent errors in the authentication process of the subject and increased work burden of the probation officer. . For example, when the surveillance target's voice changes due to a cold, etc., accurate authentication of the surveillance target cannot be performed, and the lack of accuracy in voiceprint authentication causes additional work burden on the probation officer. There is an inefficiency of the employee inputting into the system the next day, and furthermore, there is a limit in which the supervision of the person subject to surveillance is impossible for the entire time when the outing restriction is applied.

Therefore, there is a demand for a surveillance target authentication method that can replace voiceprint recognition in monitoring surveillance targets subject to a nighttime outing restriction order.

The background art of the present invention is disclosed in Korean Patent Publication No. 10-2017-0045912 (published on April 28, 2017).

An object according to one aspect of the present invention is to replace the conventional voiceprint recognition method applied to a surveillance system for a person who has been ordered to limit going out at night to a biometric authentication method through an electrocardiogram to improve the monitoring efficiency and at the same time to improve the monitoring efficiency, An object of the present invention is to provide a biometric authentication device and method capable of improving the accuracy and reliability of biometric authentication by suggesting a specific mechanism of an authentication method.

A biometric authentication device according to an aspect of the present invention collects a plurality of electrocardiogram signals through an electrocardiogram measurement module for measuring an electrocardiogram of a subject, an electrocardiogram measurement result of the electrocardiogram measurement module for the subject, and the collected plurality of electrocardiogram signals. Determining a first expected ECG signal expected to represent a method of analyzing a correlation between the determined first expected ECG signal and the collected plurality of ECG signals, and verifying validity of the first expected ECG signal method, a registration module for determining and registering a representative ECG signal serving as reference data for biometric authentication of the subject, and an actual ECG signal currently obtained from the ECG measurement result of the ECG measurement module, and the registration module and an authentication module that performs biometric authentication of the subject by analyzing a correlation between representative electrocardiogram signals registered by the user.

In the present invention, the first expected electrocardiogram signal is an electrocardiogram signal corresponding to a median of the plurality of electrocardiogram signals.

In the present invention, the registration module collects N electrocardiogram signals as the plurality of electrocardiogram signals, determines an electrocardiogram signal corresponding to a median value as the first expected electrocardiogram signal, and Correlation is analyzed for each of the N electrocardiogram signals, and M electrocardiogram signals having a correlation coefficient calculated according to a result of the correlation analysis are less than a preset reference value are removed from among the N electrocardiogram signals, and then the electrocardiogram measurement module It is characterized in that the representative electrocardiogram signal is determined by using a method of collecting M electrocardiogram signals of a subject and verifying the validity of the first expected electrocardiogram signal through the collected M electrocardiogram signals (here, N and M is a natural number greater than or equal to 2, N > M).

In the present invention, the registration module determines an electrocardiogram signal corresponding to a median value of the collected M electrocardiogram signals as a second expected electrocardiogram signal, and a correlation between the first expected electrocardiogram signal and the second expected electrocardiogram signal. When the correlation coefficient calculated according to the analysis result is equal to or greater than a predetermined reference value, it is determined that the validity of the first expected ECG signal has been verified, and the first expected ECG signal is determined as the representative ECG signal.

In the present invention, each of the electrocardiogram signals constituting the N electrocardiogram signals and the M electrocardiogram signals includes K pieces of sampling data defined as valid data representing the electrocardiogram of the subject, the value of K and the reference value Is characterized by being determined interdependently (where K is a natural number of 2 or more).

In the present invention, the registration module registers a sub-ECG signal group together with the representative ECG signal as the reference data, and the sub-ECG signal group is i) correlated in the N ECG signals collected as the plurality of ECG signals. It is characterized by a set of (N-M) ECG signals remaining after the M ECG signals are removed according to the result of the relationship analysis, and ii) a set of M ECG signals applied to the validation of the first expected ECG signal.

In the present invention, the authentication module performs primary authentication for analyzing the correlation between the actual ECG signal and the representative ECG signal, and secondary authentication for analyzing the correlation between the actual ECG signal and the lower group of ECG signals. It is characterized in that biometric authentication of the subject is performed.

In the present invention, the authentication module determines that the primary authentication is successful when a correlation coefficient calculated according to a correlation analysis result between the measured electrocardiogram signal and the representative electrocardiogram signal is equal to or greater than a preset reference value, and Subsequently, the second authentication is performed, but when each correlation coefficient calculated from the correlation analysis result between the actual ECG signal and each ECG signal constituting the lower ECG signal group is equal to or greater than a preset reference value, the second authentication It is characterized in that it is determined that the biometric authentication of the subject is successful, and finally it is determined that the biometric authentication is successful.

In the present invention, the authentication module collects I actual electrocardiogram signals through the electrocardiogram measurement module when the first authentication fails or the second authentication fails, and the collected I actual electrocardiogram signals and When the correlation coefficient calculated according to the correlation analysis result between the representative electrocardiogram signals is less than a preset reference value, it is characterized in that it is finally determined that the subject's biometric authentication has failed (I is a natural number of 2 or more).

A biometric recognition method according to an aspect of the present invention collects a plurality of electrocardiogram signals through an electrocardiogram measurement result of an electrocardiogram measurement module for a subject, and obtains a first expected electrocardiogram signal expected to represent the collected plurality of electrocardiogram signals. and a method of analyzing a correlation between the determined first expected ECG signal and the collected plurality of ECG signals and a method of verifying the validity of the first expected ECG signal, for biometric authentication of the subject. A registration step of determining and registering a representative ECG signal serving as reference data, and a method of analyzing a correlation between an actual ECG signal currently obtained from the ECG measurement result of the ECG measurement module and the representative ECG signal registered in the registration step. characterized in that it comprises an authentication step of performing biometric authentication of the subject.

According to one aspect of the present invention, in a registration process of registering reference data for biometric authentication of a subject, a representative electrocardiogram signal capable of representing a plurality of electrocardiogram signals from a subject's electrocardiogram measurement result is a predetermined correlation analysis technique. It is possible to improve the accuracy of the reference data by acquiring it by applying the ECG signal, and the representative ECG signal and the plurality of ECG signals used in the reference data registration process are used in the biometric authentication process of the subject. Reliability of biometric authentication can be improved by performing biometric authentication of a target through an organic coupling method.

In addition, biometric authentication based on the electrocardiogram as described above is an authentication technology using behavioral characteristics inside a living body, and requires authentication technology by using it in combination with authentication technology using physical characteristics such as fingerprint, face, iris, and vein. can improve security and diversity.

1 is a block diagram illustrating a biometric authentication device according to an embodiment of the present invention.
2 to 4 are exemplary diagrams for explaining an operation of a registration module in a biometric authentication device according to an embodiment of the present invention.
5 is an exemplary diagram for explaining an operation of an authentication module in a biometric authentication device according to an embodiment of the present invention.
6 to 8 are flowcharts for explaining a biometric authentication method according to an embodiment of the present invention.

Hereinafter, an embodiment of a biometric authentication device and method according to the present invention will be described with reference to the accompanying drawings. In this process, the thickness of lines or the size of components shown in the drawings may be exaggerated for clarity and convenience of description. In addition, terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of a user or operator. Therefore, definitions of these terms will have to be made based on the content throughout this specification.

As will be described below, the biometric authentication device according to the present embodiment measures an electrocardiogram of a subject, registers reference data, operates to perform biometric authentication of the subject through the registered reference data, and performs the above operation. may be applied to the above-mentioned monitoring system (hereinafter referred to as monitoring system) for monitoring the person who imposes the night out limit order. Specifically, in order to monitor the location of a subject to be monitored, such as a person who imposes an order to limit going out at night, the surveillance system is a terminal attached to the subject to be monitored (e.g., worn on the subject's wrist and senses the electrocardiogram of the subject as biometric information). It can be configured to include a wearable watch (Wearable Watch)). Since the identity of the monitoring target needs to be authenticated in order to monitor the location of the monitoring target, the present embodiment employs a biometric authentication method through electrocardiogram measurement as the identity authentication method. Accordingly, the electrocardiogram measurement module to be described below may be implemented as an electrocardiogram sensor provided in a monitored terminal worn by a subject to be monitored, and the registration module and authentication module register reference data from the electrocardiogram measurement result of the subject through the electrocardiogram measurement module, It may be implemented as a structure included in a control server that performs biometric authentication of a subject.

The above description describes an embodiment to which the biometric authentication device of the present embodiment can be applied. After registering reference data by measuring the electrocardiogram of a subject, biometric authentication through the electrocardiogram measurement of the subject is performed based on the registered reference data. The biometric authentication device of the present embodiment can be applied to various authentication systems (eg, personal information protection system, financial security system, access control system, attendance management system, etc.) within the range of performing.

Hereinafter, specific operations of the biometric authentication device according to the present embodiment will be described.

1 is a block configuration diagram for explaining a biometric authentication device according to an embodiment of the present invention, and FIGS. 2 to 4 are examples for explaining the operation of a registration module in a biometric authentication device according to an embodiment of the present invention. FIG. 5 is an exemplary diagram for explaining an operation of an authentication module in a biometric authentication device according to an embodiment of the present invention.

Referring to FIG. 1 , a biometric authentication device according to an embodiment of the present invention may include an electrocardiogram measurement module 100, a registration module 200, and an authentication module 300.

The electrocardiogram measurement module 100 may be implemented as an electrocardiogram sensor that measures the electrocardiogram of the subject, and the electrocardiogram of the subject measured by the electrocardiogram measurement module 100 is obtained through a reference data registration process by the registration module 200 and an authentication module. It can be utilized in the subject's biometric authentication process by (300).

The registration module 200 collects a plurality of electrocardiogram signals through the electrocardiogram measurement result of the electrocardiogram measurement module 100 for the subject, determines and registers (stores) a representative electrocardiogram signal serving as reference data for biometric authentication of the subject. can do. 2 shows a series of processes of determining a representative ECG signal corresponding to reference data by the registration module 200 . As shown in FIG. 2, the registration module 200 collects a plurality of electrocardiogram signals through the electrocardiogram measurement result of the electrocardiogram measurement module 100 for the subject (process ①), and represents the plurality of collected electrocardiogram signals. A method of determining a first expected ECG signal expected to be the ECG signal (process ②) and analyzing a correlation between the determined first expected ECG signal and the collected plurality of ECG signals (processes ③ and ④), and the first expected ECG signal It is possible to operate to determine and register a representative electrocardiogram signal functioning as reference data for biometric authentication of a subject by using a method for verifying the validity of .

The operation of the registration module 200 will be described in detail for each process. First, in process ①, the registration module 200 can collect N electrocardiogram signals (N is a natural number equal to or greater than 2), and at this time, the N electrocardiogram signals are each K sampling data may be included (ie, K sampling data may be included per one ECG signal). In this embodiment, as shown in FIG. 3 , sampling data is defined as valid data (eg, a peak value of a measured value) representing an electrocardiogram of a subject in an electrocardiogram signal. As a method of acquiring K sampling data, a peak value having the maximum amplitude is detected in the electrocardiogram signal, and based on the detected peak value, a piece of valid data is sampled at the front end and b pieces of valid data are sampled at the rear end. K sampling data can be obtained (ie, K = 1 (peak value) + a + b). To implement the above operation, the sampling rate of the electrocardiogram measurement module 100 may be preset to a specific rate (eg, 250 Hz (250 effective data sampling per second)). As will be described later, the number K of sampling data is determined interdependently with a 'standard value' to be compared with the correlation analysis result. The registration module 200 may collect a total of N electrocardiogram signals for the subject by repeating the electrocardiogram signal collection operation including the K pieces of sampling data as described above N times.

In step ②, the registration module 200 may determine a first expected ECG signal that is expected to represent a plurality of ECG signals (N ECG signals), and the first expected ECG signal is the reference data of the registration module 200. As a result, it serves as a candidate for a representative electrocardiogram signal that is finally determined. In this case, as shown in FIG. 4 , the first expected ECG signal may correspond to an ECG signal corresponding to the median of a plurality of ECG signals (eg, the aforementioned peak value is the median value of the plurality of ECG signals). corresponding ECG signal). That is, considering that noise is reflected in the plurality of electrocardiogram signals collected in step ① and the dispersion may be large, when only the simple average value is determined as the first expected electrocardiogram signal without considering the trend and noise of these signals, the final Since the accuracy of the representative ECG signal determined by may be lowered, the registration module 200 corresponds to the median of N ECG signals in order to remove distortion in the process of determining the representative ECG signal and improve the accuracy. The ECG signal to be performed may be determined as the first expected ECG signal.

In processes ③ and ④, the registration module 200 may analyze a correlation between the first expected ECG signal determined in process ② and a plurality of ECG signals collected in process ①. Specifically, the registration module 200 analyzes the correlation between the first expected electrocardiogram signal and each of the N electrocardiogram signals (Pearson's correlation analysis may be applied as a correlation analysis technique) (step ③), N Among the ECG signals, an operation may be performed to remove M ECG signals whose correlation coefficient (eg, Pearson's correlation coefficient, r) calculated according to the correlation analysis result is less than a preset reference value (M is a natural number less than N) (④) procedure). As will be described later, the authentication module 300 performs biometric authentication of the subject by considering the sub-ECG signal group as well as the representative ECG signal, and the sub-ECG signal group is basically defined to include N ECG signals collected in step ①. Considering that, in order to secure the reliability of biometric authentication by increasing the reference of the sub-ECG signal group, it is necessary to remove outliers among the N ECG signals. Accordingly, the registration module 200 may operate to remove M ECG signals whose correlation coefficient calculated according to the correlation analysis result is less than a preset reference value from among the N ECG signals.

In steps ⑤ to ⑦, the registration module 200 may determine and register a representative ECG signal serving as reference data for biometric authentication of a subject by using a method for verifying the validity of the first expected ECG signal.

Specifically, first, the registration module 200 additionally collects M electrocardiogram signals for the subject through the electrocardiogram measurement module 100 as much as the number M removed in process ④ (process ⑤), and collects the M electrocardiograms collected The validity of the first expected electrocardiogram signal may be verified through the signal. In step ⑤ of additionally collecting M ECG signals, N ECG signals included in the sub-ECG signal group functioning as reference data are secured to secure their reference, and at the same time, the validity of the first expected ECG signal is verified. It functions as a process of securing an electrocardiogram signal for When verifying the validity of the first expected ECG signal, the registration module 200 determines an ECG signal corresponding to the median of the M ECG signals collected in step ⑤ as the second expected ECG signal (step ⑥). ), when the correlation coefficient (eg, Pearson's correlation coefficient, r) calculated according to the correlation analysis result between the first expected ECG signal and the second expected ECG signal is greater than or equal to a preset reference value, the validity of the first expected ECG signal is verified. The first expected electrocardiogram signal may be determined as a representative electrocardiogram signal (step ⑦). That is, if the correlation coefficient between the second expected ECG signal corresponding to the median of M additionally collected ECG signals for the same subject and the first expected ECG signal is equal to or greater than the reference value, the first expected ECG signal represents the ECG of the subject. Since it can be seen that this can be done, the registration module 200 can determine a representative ECG signal through the above operation.

If, in process ⑦, the correlation coefficient between the first expected ECG signal and the second expected ECG signal is less than the reference value, it may be performed again from process ⑤, and even though processes ⑤ to ⑦ are repeated a preset number of times, the first expected ECG signal and the second expected ECG signal If the correlation coefficient between the second expected ECG signals remains less than the reference value, the registration module 200 determines that there is no reference between the N ECG signals acquired in step ① and the first expected ECG signal determined in step ②, and You can do it again from the process.

Previously, the 'reference value' was defined as the reference value applied in process ④ (that is, the parameter compared with the correlation coefficient between the first expected ECG signal determined in process ② and the N ECG signals obtained in process ①), and the reference value applied in process ⑦. (That is, parameters compared with the correlation coefficient between the first expected ECG signal determined in step ② and the second expected ECG signal determined in step ⑥), and the above two reference values are dependent on the number K of the aforementioned sampling data Within the range determined by , it may be preset to have the same value or a different value. For example, when the reference value applied in process ④ is defined as the first reference value and the reference value applied in process ⑦ is defined as the second reference value, both the first and second reference values are preset to 0.95, or the first reference value Is set to 0.95 and the second reference value may be set to 0.99. That is, the first and second reference values do not need to be determined mutually, but since the dependence between the first and second reference values and the number K of sampling data is a factor determining the reliability of the correlation analysis, the number of sampling data It only needs to be subject to constraints determined interdependently with K. As an example, assuming that the number of sampling data K is 300 and the first reference value is set to 0.95, the correlation coefficient between the first expected ECG signal and N ECG signals calculated in steps ③ and ④ is the first reference value. If it is 0.95 or more, it has a correlation analysis reliability of p-value < 0.000001 or more. In addition, assuming the case where the number of sampling data K is 300 and the second reference value is set to 0.99, the correlation coefficient between the first expected ECG signal and the second expected ECG signal calculated in step ⑦ is equal to or greater than the second reference value of 0.99. If , it also has a correlation analysis reliability of p-value < 0.000001 or more. That is, in order to improve the reliability of the correlation analysis applied in the present embodiment, the number K of sampling data and the first and second reference values may be determined interdependently, and the specific values may be determined based on the designer's experimental results. .

The representative electrocardiogram signal determined through the above process is registered as reference data. At this time, the registration module 200 may register a sub-ECG signal group as reference data together with the representative ECG signal. In this embodiment, the sub-ECG signal groups are: i) the (N-M) ECG signals remaining after the M ECG signals are removed in the process ④ according to the correlation analysis result from the N ECG signals collected in the process ①, and ii) ) It is defined as a set of M electrocardiogram signals applied to the validation of the first expected electrocardiogram signal in steps ⑤ to ⑦. That is, the lower ECG signal group is defined as a group of ECG signals directly involved in determining the representative ECG signal.

After the reference data is registered by the registration module 200 as described above, the authentication module 300 basically transmits the actual electrocardiogram signal currently obtained from the electrocardiogram measurement result of the electrocardiogram measurement module 100 to the registration module 200. The subject's biometric authentication may be performed by analyzing the correlation between the representative electrocardiogram signals registered by the user. In this embodiment, the authentication module 300, as shown in FIG. 5, performs primary authentication (processes ① and ②) of analyzing the correlation between the measured ECG signal and the representative ECG signal, and between the measured ECG signal and the sub-ECG signal groups. Biometric authentication of the subject can be performed through secondary authentication that analyzes the correlation (Steps ③ and ④). The correlation analysis technique applied in the biometric authentication process by the authentication module 300, the composition of the electrocardiogram signal, and the mutual dependence between the number K of sampling data and the reference value will be applied in the same manner as described in the registration process by the registration module 200. can

The operation of the authentication module 300 will be described in detail for each process. First, in process ①, the authentication module 300 can acquire the electrocardiogram signal of the subject through the electrocardiogram measurement module 100 (it is actually measured in the biometric authentication process). The ECG signal compared with the reference data is defined as the actual ECG signal). The ECG signal obtained in step ① of the authentication process may include K pieces of sampling data like the ECG signal obtained in step ① of the registration process.

In step ②, the authentication module 300 calculates a correlation coefficient by analyzing the correlation between the actual electrocardiogram signal and the representative electrocardiogram signal registered by the registration module 200, and the calculated correlation coefficient is a preset reference value (e.g., 0.99) or higher, it can be determined that the primary authentication has succeeded.

When the first authentication is successful, the authentication module 300 may perform the second authentication following the first authentication (step ③). When each correlation coefficient calculated from the result of correlation analysis between signals is equal to or greater than a preset reference value (eg, 0.99), it is determined that the secondary authentication has succeeded, and the subject's biometric authentication is finally determined to be successful (Step ④). Following the correlation analysis between the measured ECG signal and the representative ECG signal, biometric authentication is performed in such a way that the sub-ECG signal group used to determine the representative ECG signal is also correlated with the measured ECG signal, thereby enabling biometric authentication of the subject. It can be done very accurately and reliably.

Meanwhile, as shown in FIG. 5 , the authentication module 300 collects I actual electrocardiogram signals through the electrocardiogram measurement module 100 when the primary authentication fails or the secondary authentication fails, and the collection The correlation between the measured I electrocardiogram signals and the representative electrocardiogram signal is analyzed (step ⑤), and if the correlation coefficient calculated according to the result of the correlation analysis is less than a preset reference value, it can be finally determined that the subject's biometric authentication has failed ( ⑥ process). If the correlation coefficient calculated in step ⑤ is higher than the standard value, step ③ corresponding to the second authentication may be performed again (correlation analysis between I actually measured ECG signals and sub-ECG signal groups), and second authentication according to step ③ And, if the process ⑤ according to the failure of the secondary authentication is repeatedly performed a predetermined number of times, it may be finally determined that the subject's biometric authentication has failed.

6 to 8 are flowcharts for explaining a biometric authentication method according to an embodiment of the present invention. The biometric authentication method according to the present embodiment will be described with reference to FIGS. 6 to 8 , and description will be focused on a time-sequential configuration, excluding duplicate descriptions from those described above.

First, the registration module 200 collects a plurality of electrocardiogram signals through the electrocardiogram measurement result of the electrocardiogram measurement module 100 for the subject, and determines a first expected electrocardiogram signal expected to represent the collected plurality of electrocardiogram signals. And, using a method of analyzing the correlation between the determined first expected ECG signal and a plurality of collected ECG signals and a method of verifying the validity of the first expected ECG signal, functioning as reference data for biometric authentication of the subject A representative electrocardiogram signal is determined and registered (S100).

Referring to step S100 in detail with reference to FIG. 7 , the registration module 200 collects N electrocardiogram signals as a plurality of electrocardiogram signals through the electrocardiogram measurement results of the electrocardiogram measurement module 100 for the subject (S101).

Subsequently, the registration module 200 determines a first expected ECG signal expected to represent the plurality of ECG signals collected in step S101 (S102). In step S102, the registration module 200 determines an ECG signal corresponding to a median value among the N ECG signals as a first expected ECG signal.

Next, the registration module 200 calculates a correlation coefficient by analyzing the correlation between the first expected ECG signal determined in step S102 and the N ECG signals collected in step S101 (S103).

Subsequently, the registration module 200 removes M ECG signals having correlation coefficients calculated in step S103 that are less than a preset reference value from among the N electrocardiogram signals collected in step S101 (S104).

Subsequently, the registration module 200 additionally collects M ECG signals from the subject through the ECG measurement module 100 (S105).

Next, the registration module 200 determines a second expected ECG signal expected to represent the M ECG signals collected in step S105 (S106). As in step S102, an ECG signal corresponding to a median value among the M ECG signals is determined as a second expected ECG signal.

Subsequently, the registration module 200 calculates a correlation coefficient by analyzing a correlation between the first expected ECG signal and the second expected ECG signal (S107).

Subsequently, when the correlation coefficient calculated in step S107 is equal to or greater than the preset reference value (S108), the registration module 200 determines that the validity of the first expected ECG signal has been verified and determines the first expected ECG signal as a representative ECG signal. Register (S109). On the other hand, if the correlation coefficient is less than the reference value in step S108, it is performed again from step S105, and if the state in which the correlation coefficient is less than the reference value in step S108 is maintained even though steps S105 to S108 are repeated a preset number of times, it is performed again from step S101. Also, in step S109, the registration module 200 registers the representative ECG signal and the sub-ECG signal group as reference data.

After the reference data is registered through step S100, the authentication module 300 analyzes the correlation between the actual electrocardiogram signal currently obtained from the electrocardiogram measurement result of the electrocardiogram measurement module 100 and the representative electrocardiogram signal registered in step S100. In this way, biometric authentication of the subject is performed (S200).

Referring to step S200 in detail with reference to FIG. 8 , the authentication module 300 obtains an actual electrocardiogram signal through the electrocardiogram measurement result of the electrocardiogram measurement module 100 for the subject (S201).

Next, the authentication module 300 calculates a correlation coefficient by analyzing the correlation between the actual ECG signal obtained in step S201 and the representative ECG signal registered in step S100 (S202).

Subsequently, when the correlation coefficient calculated in step S202 is equal to or greater than the predetermined reference value (S203), the authentication module 300 analyzes the correlation between the actually measured ECG signal and each ECG signal constituting the sub-ECG signal group, and calculates the correlation coefficient, respectively. It is calculated (S204).

Subsequently, the authentication module 300 finally determines that the biometric authentication of the subject has been successful (S206) when each correlation coefficient calculated in step S204 is equal to or greater than a preset reference value (S205).

On the other hand, if it is determined that the correlation coefficient is less than the reference value in step S203 or if it is determined that the correlation coefficient is less than the reference value in step S205, the authentication module 300 obtains I actual measured electrocardiograms for the subject through the electrocardiogram measurement module 100. Signals are collected (S207), a correlation coefficient is calculated by analyzing the correlation between the collected I measured electrocardiogram signals and the representative electrocardiogram signal (S208), and if the calculated correlation coefficient is less than a preset reference value (S209), the subject's biometrics It is finally determined that authentication has failed (S210). If, in step S209, the correlation coefficient is greater than or equal to the reference value, correlation analysis between I actually measured ECG signals and sub-ECG signal groups may be performed through steps S204 and S205, and it is determined that the correlation coefficient is less than the reference value in step S205, and steps S209 In the step, if it is determined that the correlation coefficient is less than the reference value is repeatedly performed a predetermined number of times, it is finally determined that the subject's biometric authentication has failed.

In this way, the present embodiment is a method of applying a predetermined correlation analysis technique to a representative ECG signal that can represent a plurality of ECG signals from the ECG measurement result of the subject in the registration process of registering reference data for biometric authentication of the subject. Acquisition can improve the accuracy of the reference data, and the organic combination method of the registration process and the authentication process can be used in the biometric authentication process of the subject, including the ECG signal used in the registration process of the reference data along with the representative ECG signal. Reliability of biometric authentication can be improved by performing biometric authentication of the subject through the process. In addition, by applying the biometric authentication process through electrocardiogram measurement as described above to a monitoring system for a person who has been ordered to limit going out at night, the monitoring efficiency of the person to be monitored can be improved.

The term “module” used in this specification may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as, for example, logic, logic block, component, or circuit. A module may be an integrally constructed component or a minimal unit of components or a portion thereof that performs one or more functions. For example, according to one embodiment, the module may be implemented in the form of an Application-Specific Integrated Circuit (ASIC). Further, implementations described herein may be implemented in, for example, a method or process, an apparatus, a software program, a data stream, or a signal. Even if discussed only in the context of a single form of implementation (eg, discussed only as a method), the implementation of features discussed may also be implemented in other forms (eg, an apparatus or program). The device may be implemented in suitable hardware, software and firmware. The method may be implemented in an apparatus such as a processor, which is generally referred to as a processing device including, for example, a computer, microprocessor, integrated circuit or programmable logic device or the like. Processors also include communication devices such as computers, cell phones, personal digital assistants ("PDAs") and other devices that facilitate communication of information between end-users.

The present invention has been described above with reference to the embodiments shown in the drawings, but this is only exemplary, and those skilled in the art can make various modifications and equivalent other embodiments. you will understand the point. Therefore, the technical protection scope of the present invention should be determined by the claims below.

100: electrocardiogram measurement module
200: registration module
300: authentication module

Claims (10)

Electrocardiogram measurement module for measuring the subject's electrocardiogram;
A plurality of electrocardiogram signals are collected through the electrocardiogram measurement result of the electrocardiogram measurement module for the subject, a first expected electrocardiogram signal expected to represent the collected plurality of electrocardiogram signals is determined, and the determined first expected electrocardiogram signal is determined. A representative ECG signal serving as reference data for biometric authentication of the subject is obtained by using a method of analyzing a correlation between a signal and a plurality of collected ECG signals and a method of verifying the validity of the first expected ECG signal. a registration module that determines and registers; and
an authentication module that performs biometric authentication of the subject by analyzing a correlation between an actual electrocardiogram signal currently obtained from the electrocardiogram measurement result of the electrocardiogram measurement module and a representative electrocardiogram signal registered by the registration module;
Biometric authentication device comprising a. According to claim 1,
The first expected electrocardiogram signal is an electrocardiogram signal corresponding to a median of the plurality of electrocardiogram signals.
According to claim 2,
The registration module collects N ECG signals as the plurality of ECG signals, determines an ECG signal corresponding to a median value as the first expected ECG signal, and determines the first expected ECG signal and the N ECG signals. Correlation is analyzed for each signal, and after removing M ECG signals whose correlation coefficient calculated according to the correlation analysis result is less than a preset reference value among the N ECG signals, the M ECG signals for the subject through the ECG measurement module and determining the representative electrocardiogram signal by using a method of collecting electrocardiogram signals and verifying the validity of the first expected electrocardiogram signal through the collected M electrocardiogram signals (where N and M is a natural number greater than or equal to 2, N > M).
According to claim 3,
The registration module determines an ECG signal corresponding to a median value of the collected M ECG signals as a second expected ECG signal, and according to a correlation analysis result between the first expected ECG signal and the second expected ECG signal When the calculated correlation coefficient is greater than or equal to a preset reference value, the biometric authentication device determines that the validity of the first expected ECG signal has been verified and determines the first expected ECG signal as the representative ECG signal.
According to claim 3,
Each of the electrocardiogram signals constituting the N electrocardiogram signals and the M electrocardiogram signals includes K pieces of sampling data defined as valid data representing the electrocardiogram of the subject, and the value of K and the reference value are mutually dependent. A biometric authentication device characterized in that it is determined (where K is a natural number of 2 or greater).
According to claim 3,
The registration module registers a sub-ECG signal group together with the representative ECG signal as the reference data, wherein the sub-ECG signal group corresponds to i) a correlation analysis result from the N ECG signals collected as the plurality of ECG signals. a set of (NM) ECG signals remaining after the M ECG signals are removed according to the method, and ii) a set of M ECG signals applied to verify validity of the first expected ECG signal.
According to claim 6,
The authentication module performs first authentication for analyzing the correlation between the measured ECG signal and the representative ECG signal, and second authentication for analyzing the correlation between the measured ECG signal and the lower group of ECG signals. A biometric authentication device characterized in that it performs authentication.
According to claim 7,
The authentication module determines that the first authentication is successful when a correlation coefficient calculated according to a result of correlation analysis between the measured electrocardiogram signal and the representative electrocardiogram signal is equal to or greater than a preset reference value, and subsequent to the first authentication, the second Secondary authentication is performed, but when each correlation coefficient calculated from the correlation analysis result between the actual ECG signal and each ECG signal constituting the sub-ECG signal group is equal to or greater than a preset reference value, it is determined that the secondary authentication is successful , The biometric authentication device characterized in that it is finally determined that the subject's biometric authentication has succeeded.
According to claim 8,
The authentication module, when the primary authentication fails or the secondary authentication fails, collects I actual electrocardiogram signals through the electrocardiogram measurement module, and collects the I actual electrocardiogram signals collected and the representative electrocardiogram signal. The biometric authentication device (where I is a natural number of 2 or more) finally determines that the subject's biometric authentication has failed if the correlation coefficient calculated according to the result of the correlation analysis is less than a preset reference value.
A plurality of electrocardiogram signals are collected through the electrocardiogram measurement result of the electrocardiogram measuring module for the subject, a first expected electrocardiogram signal expected to represent the collected plurality of electrocardiogram signals is determined, and the determined first expected electrocardiogram signal and Determining a representative ECG signal serving as reference data for biometric authentication of the subject using a method of analyzing the correlation between the collected plurality of ECG signals and a method of verifying the validity of the first expected ECG signal registration step of registering; and
an authentication step of performing biometric authentication of the subject by analyzing a correlation between a currently measured electrocardiogram signal obtained from an electrocardiogram measurement result of the electrocardiogram measurement module and a representative electrocardiogram signal registered in the registration step;
Biometric authentication method comprising a.

Images