JP2017142580A - Attribution determination system, attribution determination apparatus, and attribution determination method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To speedily determine whether or not, health information measured from a measured person is health information of the measured person.SOLUTION: An attribution determination apparatus comprises: a communication section 410 which receives health information on a measured person measured by a measuring device 200, a first signal acquired by the measuring device 200, authentication information on an authenticated person extracted by an authentication device 300, and a second signal acquired by the authentication device 300; a biological authentication section 420 which authenticates an authenticating person based upon the authentication information; an offset time processing section 441 which controls the measuring device 200 to impart synchronizing pulses to the first signal and second signal before starting acquiring a part to be used for matching processing of the first signal, and synchronizes the first signal and second signal based upon the synchronizing pulses; and a matching processing section 443 which performs the matching processing to determine whether the health information is health information on the authenticated person based upon the first signal and second signal having been synchronized by the offset time processing part 441.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an attribute determination system, an attribute determination device, and an attribute determination method.

  In recent years, personal data generated by PCs, smartphones, IoT devices, and the like has increased. In particular, in the healthcare field, many lifelog services that acquire and store personal data on a daily basis are provided. In a system for operating such a service, a registrant registers personal data in the system through device authentication such as ID / password authentication.

  For example, Patent Document 1 describes a system using personal data. In the method described in Patent Document 1, a user authentication unit performs user authentication using a user ID, and a step count information acquisition unit generates personal data such as user step count information. Further, it is described that the service use restriction control unit restricts a service function that can be used by the user based on personal data such as the number of steps of the user related to the user ID.

  Thus, in a system using personal data, user authentication and personal data generation are usually performed separately. Since personal data is managed by the individual user, it is possible that the personal data of the user can be replaced with personal data of another person. In a system that uses personal data within the scope of personal responsibility, there is no disadvantage even if a user registers illegal data using personal data by another person.

  However, for example, in a service that grants a privilege based on personal data such as calculating insurance premiums according to personal data accumulated every day, if the user is allowed to register unauthorized data, the service is There is a risk of bankruptcy. Therefore, in order for both the individual user and the service provider to use personal data with peace of mind, it is important to ensure the attribution of personal data. In addition, it is also required that processing related to personal data attribution be performed in a realistic time and effort.

  Therefore, the measuring device measures the measurement device by incorporating sensors capable of acquiring biological signals in the authentication device and the measuring device, and collating the signals acquired by the sensor incorporated in the authentication device and the sensor incorporated in the measuring device, respectively. A system for determining whether a measured value belongs to the person specified by the authentication device is considered. As the above-described sensor, for example, a pulse wave sensor can be used. In this case, the pulse wave sensor of the authentication device and the pulse wave sensor of the measurement device respectively acquire a pulse wave as a signal from the person to be authenticated and the person to be measured, and collate the two pulse waves, thereby It is assumed that the measurement person is the same, that is, the measurement value measured from the measurement person by the measurement apparatus belongs to the authentication person specified by the authentication apparatus.

JP 2012-230503 A

  However, when comparing the two pulse waves acquired by the pulse wave sensor of the authentication device and the pulse wave sensor of the measurement device as described above, the time difference and the operation for instructing the authentication device and the measurement device to start the operation of the pulse wave sensor, respectively. Due to the difference in the response time of the sensor device from the start instruction to the actual start of pulse wave acquisition, and the propagation time difference due to the difference in the acquisition site of the pulse wave, the pulse wave due to the same heartbeat has a time difference Sometimes accompanied. Therefore, it is necessary to perform processing for synchronizing these two pulse waves in time, which takes time, and it takes a long time to determine whether or not the person to be measured and the person to be authenticated are the same. May be required.

  In order to solve the above-described problems, an attribute determination device according to the present invention is an attribute determination device that transmits and receives information to and from a measurement device and an authentication device via a communication network, and the attribute determination device measured by the measurement device. The health information of the measurer, the first signal that is the biometric information acquired by the measurement device, the authentication information of the person to be authenticated extracted by the authentication device, and the second signal that is the biometric information acquired by the authentication device Before starting the acquisition of the portion used for the verification process in the first signal, the biometric authentication unit that authenticates the authenticator based on the authentication information, and the measurement device An offset time for controlling the first signal and the second signal to be simultaneously applied with a synchronization pulse and synchronizing the first signal and the second signal based on the synchronization pulse. Whether the health information measured by the measurement device based on the first signal and the second signal synchronized by the processing unit and the offset time processing unit is the health information of the person to be authenticated. And a collation processing unit for performing collation processing for determining whether or not.

  An attribute determination method according to the present invention is an attribute determination method executed by an attribute determination apparatus that transmits and receives information to and from a measurement apparatus and an authentication apparatus via a communication network, and is measured by the measurement apparatus. The health information of the measurement subject, the first signal that is the biometric information acquired by the measurement device, the authentication information of the test subject extracted by the authentication device, and the biometric information acquired by the authentication device 2, before authenticating the authenticator based on the authentication information, and before the measurement device starts to acquire a portion of the first signal used for verification processing, Controlling to simultaneously apply a synchronization pulse to the first signal and the second signal, and synchronizing the first signal and the second signal based on the synchronization pulse; Performing a verification process for determining whether the health information is the health information of the person to be authenticated based on the first signal and the second signal that are synchronized. Features.

  An attribute determination system according to the present invention is an attribute determination system including a measurement device, an authentication device, and an attribute determination device that transmit and receive information to and from each other via a communication network. A measurement unit that measures health information including one or more of the biological information and exercise information of the measured person from the person, and a first signal that is a first signal that is the biological information of the measured person from the measured person The authentication device includes an authentication information extraction unit that extracts authentication information for extracting authentication information for authenticating the authenticated person from the authenticated person, and the authenticated person, A second signal acquisition unit that acquires a second signal that is biometric information of the person to be authenticated, and the belonging determination device includes the health information, the first signal, the authentication information, and the A communication unit for receiving the second signal; A biometric authentication unit that authenticates an authenticator based on the authentication information, and before the measurement device starts to acquire a portion of the first signal used for verification processing, the first signal and the second signal The control is performed so that a synchronization pulse is simultaneously applied to the signal, and an offset time processing unit that synchronizes the first signal and the second signal based on the synchronization pulse, and the synchronization performed by the offset time processing unit A verification processing unit that performs verification processing for determining whether the health information measured by the measurement unit is the health information of the person to be authenticated based on the first signal and the second signal. It is characterized by that.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to determine rapidly whether the health information measured from the to-be-measured person is the to-be-authenticated person's health information.

It is a functional block diagram of the attribute determination system which concerns on this embodiment. It is a figure which shows the example of the time difference of a pressure pulse wave and a volume pulse wave. It is a figure for demonstrating a pulse wave propagation time difference. It is a figure which shows the example of a pulse wave propagation time difference and a dispersion value. It is a sequence diagram which shows the process of the belonging determination system shown in FIG. It is a flowchart which shows the detail of the feature point collation process shown in FIG.

  Embodiments of the present invention will be described below with reference to the drawings.

<< Functional structure of attribute determination system >>
First, the overall configuration of the present embodiment will be described with reference to FIG.

  As shown in FIG. 1, the attribution determination system 1 includes a measurement device 200, an authentication device 300, and an attribution determination device 400. The measurement device 200, the authentication device 300, and the attribution determination device 400 are connected via a communication network and transmit / receive information to / from each other.

<Functional configuration of measuring device>
The measurement apparatus 200 includes a communication unit 210, a measurement unit 220, and a first signal acquisition unit 230.

  The communication unit 210 transmits the information measured by the measurement unit 220, the information acquired by the first signal acquisition unit 230, and the like to the attribution determination device 400 via the communication network.

  When the operation start instruction is input by the measurement subject, the measurement unit 220 measures the measurement target. The measurement target is arbitrary information belonging to an individual, and is, for example, health information including biological information of the measurement subject, exercise information that is a record of exercise performed by the measurement subject, and the like. Biometric information measurement targets include blood pressure, pulse, blood oxygen concentration, body temperature, bone conduction sound, body weight, heart sound, airway sound, and the like. Specific examples of the exercise information include the number of steps taken by the measurement subject. In the present embodiment, the health information is described as a blood pressure value. In this case, the measuring apparatus 200 is realized by a sphygmomanometer that measures blood pressure.

  Further, when the measurement unit 220 measures the health information of the measurement subject, the health information, the measurement ID received from the belonging determination device 400 via the authentication device 300, the measurement time when the measurement subject was measured, and the measurement target The measurement metadata including the type and the like is associated and output to the communication unit 210. The measurement ID is an identifier for uniquely identifying the measurement by the measurement unit 220.

  The 1st signal acquisition part 230 acquires the 1st signal which is living body information resulting from a to-be-measured person's body. The first signal is a characteristic that changes over time, and includes, for example, types such as pressure pulse wave, pulse, blood oxygen concentration, body temperature, bone conduction sound, body weight, heart sound, and airway sound. In the present embodiment, the first signal is described as a pressure pulse wave. In this case, the first signal acquisition unit 230 is realized by a pulse wave sensor.

  The pulse wave sensor that realizes the first signal acquisition unit 230 is configured such that the sphygmomanometer that realizes the measurement unit 220 cannot simultaneously acquire a pressure pulse wave from a person different from the person who measures the blood pressure. ing. For example, the pulse wave sensor and the sphygmomanometer may be configured as a single unit and configured to be worn only on the arm of the same person.

  In addition, when the first signal acquisition unit 230 acquires the first signal from the measurement subject, the first signal, the extraction time at which the first signal is extracted, the type of the first signal, and the signal are acquired. The measurement metadata including the body part name and the like is associated and output to the communication unit 210. The measurement metadata may not include the type. In that case, the attribute determination device 400 that has received the first signal determines the type from the characteristics of the first signal.

<Functional configuration of authentication device>
The authentication device 300 includes a communication unit 310, an authentication information extraction unit 320, and a second signal acquisition unit 330.

  The communication unit 310 transmits the information extracted by the authentication information extraction unit 320, the information acquired by the second signal acquisition unit 330, and the like to the attribution determination device 400 via the communication network.

  When the operation start instruction is input, the authentication information extraction unit 320 extracts authentication information for authenticating the person to be authenticated from the person to be authenticated. The authentication information is biometric information for use in biometric authentication, such as fingerprint information and vein information. Further, when the authentication information extraction unit 320 extracts the authentication information from the person to be authenticated, the authentication information extraction unit 320 outputs the extracted authentication information to the communication unit 310.

  The second signal acquisition unit 330 acquires a second signal that is biometric information resulting from the body of the person to be authenticated. The second signal is a characteristic that changes with time, and includes types such as pressure pulse wave, pulse, blood oxygen concentration, body temperature, bone conduction sound, body weight, heart sound, and airway sound. In the present embodiment, the second signal is described as a volume pulse wave. In this case, the second signal acquisition unit 330 is realized by a pulse wave sensor.

  In addition, the pulse wave sensor that realizes the second signal acquisition unit 330 is configured so that a volume pulse wave cannot be acquired simultaneously from a person different from the person to be authenticated that the authentication information extraction unit 320 authenticates. For example, the pulse wave sensor and the hardware that implements the authentication information extraction unit 320 may be configured as a single unit and may be configured to be worn only on the arm of the same person.

  In addition, when the second signal acquisition unit 330 acquires the second signal from the person to be authenticated, the second signal, the acquisition time when the second signal is acquired, the type of the second signal, and the second signal Are associated with authentication metadata including the body part name and the like acquired, and output to the communication unit 210. The authentication metadata may not include the type. In that case, the attribute determination device 400 that has received the second signal may determine the type of the second signal from the characteristics of the second signal.

<Functional configuration of attribute determination device>
The attribute determination device 400 includes a communication unit 410, a biometric authentication unit 420, an authentication information storage unit 430, and a verification unit 440.

  The communication unit 410 transmits / receives information to / from the communication unit 210 of the measurement apparatus 200 and the communication unit 310 of the authentication apparatus 300.

  The biometric authentication unit 420 performs biometric authentication processing using the authentication information extracted from the person to be authenticated by the authentication device 300 received by the communication unit 410. Specifically, the biometric authentication unit 420 searches the authentication information storage unit 430 based on the authentication information, and extracts one or more of the user ID and the user name corresponding to the authentication information, thereby authenticating the person to be authenticated. Is identified.

  The biometric authentication unit 420 generates a measurement ID when performing biometric authentication processing.

  The authentication information storage unit 430 stores information related to the user who is the person to be authenticated. The authentication information storage unit 430 records a user ID for uniquely identifying a user, a user name representing the user's name, and user authentication information in association with each other.

  The collation unit 440 includes an offset time processing unit 441, a feature point calculation processing unit 442, and a feature point collation processing unit (collation processing unit) 443. Based on the first signal and the second signal, the measurement unit 220 A verification process for determining whether the measured health information is the health information of the person to be authenticated is performed. When the offset time processing unit 441, the feature point calculation processing unit 442, and the feature point matching processing unit 443 perform processing, the matching unit 440 further includes a function unit for performing noise reduction or the like as necessary. Also good.

  The offset time processing unit 441 performs offset time processing based on the first signal and the second signal received by the communication unit 410.

  Here, the offset time processing will be described. The first signal and the second signal are generated due to the heartbeats of the person to be authenticated and the person to be measured, respectively, but it is assumed that the person to be authenticated and the person to be measured are the same person. However, due to some factors, a difference (detection time difference) may occur between the time when the first signal due to the same heartbeat is acquired and the time when the second signal is acquired.

  In the offset time processing, in order to remove the detection time difference, the first signal and the second signal generated due to the same heartbeat are represented at the same time (hereinafter referred to as “synchronize”). In addition, it is a process of relatively shifting the time change of the second signal with respect to the reference first signal.

  Here, the cause of the detection time difference will be described.

  The measurement device 200 that acquires the first signal, that is, the pressure pulse wave, receives the instruction to start the operation from the user or the like, and the authentication device 300 that acquires the second signal, that is, the volume pulse wave, receives the instruction to start the operation. There may be a pulse wave acquisition start time difference from the time. Further, the device reaction time from when the measurement apparatus 200 receives an operation start instruction until it starts to acquire a pressure pulse wave, and the device from when the authentication apparatus 300 receives an operation start instruction until it starts acquiring a volume pulse wave There may be a device reaction time difference between the reaction times.

Therefore, the offset time processing unit 441 controls the measurement device 200 and the authentication device 300 to simultaneously give intentional artifacts such as a synchronization pulse as electrical noise to the first and second signals, respectively. When the synchronization pulse is given, waveforms affected by the synchronization pulse appear in the first and second signals. The offset processing unit 441 generates a pulse wave based on the time t _AC when the waveform affected by the synchronization pulse appears in the first signal and the time t _FC when the waveform affected by the synchronization pulse appears in the second signal. The time difference including the acquisition start time difference and the device reaction time difference is defined as t _FC −t _AC .

Then, as shown in (a) and (b) of FIG. 2, the offset time processing unit 441 executes the second signal with respect to the first signal so as to represent the waveform affected by the synchronization pulse at the same time. Synchronization is achieved by shifting the signal by time t _FC -t _AC .

In addition, the body part of the measurement subject 200 whose pressure pulse wave is measured may be different from the body part of the subject person whose authentication device 300 measures the volume pulse wave. For example, the measuring apparatus 200 measures the pressure pulse wave with the upper arm of the person to be measured, and the authentication apparatus 300 measures the volume pulse wave with the wrist of the person to be authenticated. At this time, even if the person to be measured and the person to be authenticated are the same, there is a difference in the propagation time of the pulse wave because the measurement site is different, and the difference in the pulse wave acquisition start time and the device reaction time difference as described above. Even after synchronizing the included time difference, as shown in FIGS. 2C and 2D, there is a difference between the measurement time of the pressure pulse wave and the measurement time of the volume pulse wave due to the same heartbeat ( A pulse wave propagation time difference t _FA ) is generated. This pulse wave propagation time difference t _FA is set to a predetermined value determined in advance by experiments or the like.

Furthermore, the above-described pulse wave propagation time difference t _FA has an individual difference of about 100 milliseconds depending on age, sex, physique, health condition, etc., and a time difference that compensates for this individual difference, an error that may occur when detecting a feature point, etc. _Let ε _Ai be the error that includes.

Therefore, the time t _Ai when the pressure pulse wave feature point (referred to as “first feature point”) A _i appears is the volume pulse feature point (referred to as “second feature point”) caused by the same heartbeat. There is a relationship of the formula (1) with respect to the time t _Fj when F _j appears.

t _Ai = t _Fj + (t _FC −t _AC ) + t _FA + ε _Ai (1)

Therefore, the offset time processing unit 441 shifts the pressure pulse wave by time (t _FC −t _AC ) + t _FA + ε _Ai to synchronize the pressure pulse wave and the volume pulse wave, that is, the first signal and the second signal. Synchronize with the signal.

  The offset time processing unit 441 performs offset time processing based on the first and second signals acquired in the pressurization period in which the cuff pressure increases in the sphygmomanometer as the measuring apparatus 200. The collation process by the feature point collation processing unit 443, which will be described in detail later, is the first and second signals acquired during the decompression period when the cuff pressure is reduced and the pressure pulse wave that is the first signal can be stably acquired. Based on. Therefore, the offset time processing unit 441 can perform the offset time processing in the background based on a signal that is not used for the collation processing, and thus it is possible to reduce the time required for the entire attribution determination processing. .

The feature point calculation processing unit 442 includes a first feature point A _i that is a feature point of the first signal synchronized by the offset time processing unit 441 and a second feature point that is a feature point of the second signal. _Fj is detected. The first signal, that is, the characteristic point of the pressure pulse wave is a rising point corresponding to the start of the systole of the pressure pulse wave. Further, the second signal, that is, the feature point of the volume pulse wave is a peak that is a positive initial contraction wave of the acceleration pulse wave calculated by second-order differentiation of the time change of the volume pulse wave.

Since the pressure pulse wave changes periodically with respect to time, the first feature point A _i appears periodically in the time change of the pressure pulse wave. Therefore, the feature point calculation processing unit 442 detects the times t _A1 , t _A2 , t _A3 ,..., T _Ai ,.

Similarly, since the volume pulse wave periodically changes with respect to time, the second feature point F _j periodically appears in the time change of the volume pulse wave. Therefore, the feature point calculation processing unit 442 detects the times t _F1 , t _F2 , t _F3 ,..., T _Fj,.

  In this embodiment, the first signal is a pressure pulse wave, and the second signal is a volume pulse wave. For example, when both the first and second signals are pressure pulse waves or volume pulse waves, a feature point May be a point indicating characteristics in each period such as a rising point and a peak point of a time change of a pressure pulse wave or a volume pulse wave.

The feature point matching processing unit 443 includes the pressure pulse wave time series t _A1 , t _A2 , t _A3 ,..., T _Ai,. match _{F1, t F2, t F3,} ···, t Fj, whether the person to be authenticated is identical according to the subject and the volume pulse wave of the pressure pulse wave based on the ... Perform verification processing.

Specifically, the feature point matching processing unit 443 uses the number of first feature points A _i that appear within a predetermined time range in the pressure pulse wave time sequence in the decompression period (hereinafter, “first feature points”). And the number of second feature points F _j appearing in the same time range in the plethysmogram time sequence (hereinafter referred to as “second feature point number”) is greater than or equal to a predetermined value. Determine whether or not. When the difference between the first feature score and the second feature score is a predetermined value or more, the feature point matching processing unit 443 sets the matching result to “NG”, that is, the health information measured from the measurement subject is Suppose that it is not the health information of the person to be certified. When the difference between the number of first feature points and the number of second feature points is less than a predetermined value, the feature point matching processing unit 443 performs pair determination processing.

  Here, the pair determination process will be described in detail.

The feature point matching processing unit 443 uses the second feature point F _j (within the pair determination error threshold with reference to the time t _Ai when the first feature point A _i (i = 1, 2,... N) appears. When j = 1, 2,... m) appears, pair generation processing is performed in which a combination of the first feature point A _i and the second feature point F _j is paired.

Further, the feature point matching process section 443, for the total number of the first feature points A _i, as a pair percentage the ratio of the number of first feature points A _i which is to be the second feature point F _j paired calculate. And the feature point collation process part 443 determines whether a pair ratio is more than a predetermined pair determination threshold value. When the pair ratio is less than the pair determination threshold, the feature point matching processing unit 443 determines that the matching result is “NG”, that is, the health information measured from the person being measured is not the health information of the person to be authenticated, and the pair ratio Match determination processing is performed when is equal to or greater than the pair determination threshold.

  Here, the match determination process will be described in detail.

Time feature point matching process unit 443, feature point time difference, namely that the time t _Ai of the first feature points A _i has appeared, a second feature point F _j is the first feature points A _i and pair appeared It is determined whether or not the difference from t _Fj is equal to or greater than a predetermined match determination error threshold. The match determination threshold is a value smaller than the pair determination threshold.

When the feature point time difference is less than the match determination error threshold, the feature point matching processing unit 443 assumes that the first feature point A _i matches the second feature point F _j . If the feature point time difference is equal to or greater than the match determination error threshold, the feature point matching processing unit 443 assumes that the first feature point A _i and the second feature point F _j do not match.

Further, the feature point matching process section 443, for the total number of the first feature points A _i, match percentage the ratio of the number of first feature points A _i which is to be matched with the second feature point F _j Calculate as Then, the feature point matching processing unit 443 determines whether or not the match ratio is greater than or equal to a predetermined match determination threshold value. When the match ratio is less than the match determination threshold, the feature point matching processing unit 443 determines that the matching result is “NG”, that is, the health information measured from the measurement subject is not the health information of the person to be authenticated, and the match ratio is A variance value determination process is performed when the value is equal to or greater than the match determination threshold value.

  Here, the dispersion value determination process will be described.

  The feature point matching processing unit 443 calculates a variance value of the feature point time difference, and determines whether or not it is equal to or greater than a predetermined variance value threshold.

  With reference to the example of FIG. 3, the variance value of the feature point time difference will be described in detail. 3A and 3B show the pressure pulse wave (1 (1) of FIGS. 3A and 3B) which is the first signal subjected to the offset time processing as described above by the offset time processing unit 431. FIG. )) And the volume pulse wave as the second signal (see (2) in FIGS. 3A and 3B, respectively). In FIG. 3A, the difference in propagation time between the pressure pulse wave and the volume pulse wave due to the same heartbeat is expressed by the offset time processing so as to be very small. On the other hand, in FIG. 3B, even when the offset time processing is performed, the propagation time difference caused by the difference between the parts detecting the pressure pulse wave and the volume pulse wave is shown.

  FIG. 4 is a characteristic point time difference which is a difference between the characteristic point time of the pressure pulse wave and the characteristic point time of the volume pulse wave of each pair shown in (a) to (f) of FIGS. The variance of the feature point time difference is shown at the bottom. As shown in FIG. 4B, even if there is a time difference between the time when the pressure pulse wave feature point appears and the time when the volume pulse wave feature point appears, as in the example shown in FIG. If the variance value is less than the predetermined threshold value, this time difference is due to the difference between the parts detecting the pressure pulse wave and the volume pulse wave, and these characteristic points are considered to be caused by the same heartbeat.

  Therefore, the feature point matching processing unit 443 sets the matching result to “OK” if the variance value is less than the predetermined variance value threshold, and sets the matching result to “NG” if the variance value is equal to or greater than the variance value threshold.

<Operation of attribution system>
Then, with reference to FIG. 5, operation | movement of the belonging determination system 1 which concerns on this embodiment is demonstrated.

  First, the person to be measured wears a cuff of a sphygmomanometer that is the measuring apparatus 200 on the arm (step S11). Further, the person to be authenticated places a finger on the authentication device 300 (step S12).

  When the person to be authenticated puts a finger on the authentication apparatus 300 in step S12, the authentication apparatus 300 extracts authentication information from the finger that is a body part of the person to be authenticated (step S13). When the authentication information is extracted in step S13, the authentication device 300 transmits the acquired authentication information together with the authentication request to the attribution determination device 400 via the communication network (step S14).

  When the authentication information is transmitted in step S14, the biometric authentication unit 420 of the attribute determination device 400 receives the authentication information and performs an authentication process (step S15). When the authentication process is performed in step S15, the biometric authentication unit 420 generates a measurement ID (step S16). When the measurement ID is generated in step S16, the communication unit 410 transmits the measurement ID generated by the biometric authentication unit 420 to the authentication device 300 (step S17).

  When the measurement ID is transmitted to the authentication device 300 in step S17, the communication unit 310 of the authentication device 300 transmits the measurement ID to the measurement device 200 (step S18).

  Further, when the measurement ID is transmitted to the authentication device 300 in step S17, the second signal acquisition unit 330 of the authentication device 300 starts operating (step S19), thereby acquiring the volume pulse wave from the person to be authenticated. (Step S20).

  When the second signal acquisition unit 330 of the authentication device 300 starts operating in step S19, the measurement subject inputs an operation start instruction to the measuring device 200 (step S21). When a blood pressure measurement start instruction is input to the measurement device 200 in step S21, the measurement unit 220 of the measurement device 200 starts operation (step S22), measures the blood pressure value of the measurement subject, and acquires a volume pulse wave. (Step S23).

  Then, when the measurement of the blood pressure value by the sphygmomanometer that is the measurement unit 220 is completed, the first signal acquisition unit 230 ends the acquisition of the pressure pulse wave (step S24). In step S24, when the measurement unit 220 finishes measuring the blood pressure value and the first signal acquisition unit 230 finishes acquiring the pressure pulse wave, the communication unit 210 transmits measurement end information to the authentication device 300 (step S25). ). When the measurement end information is transmitted in step S25, the communication unit 310 of the authentication device 300 receives the measurement end information, and the second signal acquisition unit 330 ends the acquisition of the volume pulse wave (step S26).

  On the other hand, in step S24, when the measurement unit 220 finishes measuring the blood pressure value and the first signal acquisition unit 230 finishes acquiring the pressure pulse wave, the communication unit 210 uses the blood pressure value and the pressure pulse wave to determine the attribute determination device. It transmits to 400 (step S27). Similarly, when the second signal acquisition unit 330 finishes acquiring the volume pulse wave in step S26, the communication unit 310 transmits the volume pulse wave to the attribute determination device 400 (step S28).

  In step S27, when the blood pressure value and the pressure pulse wave are transmitted, the communication unit 410 of the belonging determination device 400 receives the blood pressure value and the pressure pulse wave. Similarly, when the volume pulse wave is transmitted in step S26, the communication unit 410 receives the volume pulse wave. Then, based on the pressure pulse wave and volume pulse wave received by the communication unit 410, the offset time processing unit 441 performs offset time processing (step S29).

  When the offset time process is performed in step S29, the feature point calculation processing unit 442 performs the feature point calculation process (step S30). When the feature point calculation process is performed in step S30, the feature point matching processing unit 443 performs the feature point matching process and outputs a matching result (step S31).

  Here, the feature point matching process in step S31 will be described in detail with reference to FIG.

  First, the feature point matching processing unit 443 performs the first feature point in the predetermined time range in the pressure pulse wave time sequence in the decompression period and the second feature point in the same time range in the volume pulse wave time sequence. It is determined whether or not the difference is greater than or equal to a predetermined value (step S311).

  If it is determined in step S311 that the difference between the first feature point number and the second feature point number is less than a predetermined number, the feature point matching processing unit 443 performs pair generation processing (step S312). When pair generation processing is performed in step S312, the feature point matching processing unit 443 calculates a pair ratio (step S313).

  When the pair ratio is calculated in step S313, the feature point matching processing unit 443 determines whether or not the pair ratio is equal to or greater than the pair determination threshold (step S314). If it is determined in step S314 that the pair ratio is greater than or equal to the pair determination threshold, the feature point matching processing unit 443 calculates a match ratio (step S315).

  When the match ratio is calculated in step S314, the feature point matching processing unit 443 determines whether or not the match ratio is equal to or greater than the match determination threshold (step S316). If it is determined in step S316 that the match ratio is greater than or equal to the match determination threshold, the feature point matching processing unit 443 calculates a variance value of the feature point time difference (step S317).

  When the variance value of the feature point time difference is calculated in step 317, the feature point matching processing unit 443 determines whether or not the variance value is equal to or greater than the variance value determination threshold (step S318). If it is determined in step S318 that the variance value is less than the variance value determination threshold, the feature point matching processing unit 443 sets the matching result to “OK”, that is, the health information measured from the measurement subject is the health of the authentication subject. It is assumed that the information is information (step S319).

  If it is determined in step S311 that the difference between the first feature point number and the second feature point number is greater than or equal to a predetermined value, if it is determined in step S314 that the pair ratio is less than the pair determination threshold value, step S316 is performed. When the match ratio is determined to be less than the match determination threshold value or when the variance value is determined to be greater than or equal to the variance value determination threshold value in step S318, the feature point matching processing unit 443 sets the matching result to “NG”. That is, it is assumed that the health information measured from the subject is not the health information of the subject (step S320).

  Returning to FIG. 5, when the feature point matching process is performed in step S31, the communication unit 410 transmits the matching result and the blood pressure value received in step S27 to the authentication device 300 (step S32). When the collation result and the blood pressure value are transmitted in step S31, the communication unit 410 of the authentication device 300 receives the collation result and the blood pressure value, and the display unit so that the person to be authenticated can visually recognize the received collation result and the blood pressure value. Or it displays on a display apparatus (step S33).

  As described above, in the attribution determination system 1 according to the present embodiment, the first signal and the second signal are synchronized with the portion of the first signal before the acquisition of the portion used for the collation processing is started. A pulse is given, and synchronization is established based on the synchronization pulse. Therefore, the offset time processing unit 441 can perform the offset time processing in the background based on the signal acquired in the pressurization period that is not used for the matching processing, thereby reducing the time required for the entire attribute determination processing. It can be shortened. Therefore, the user can feel the time required for acquiring the pressure pulse wave, and can complete the attribution determination process within the time equivalent to the time required for performing the normal measurement using the measuring apparatus 200. . Thereby, usability can be ensured.

  In addition, in the attribute determination system 1 of the present embodiment, the offset time processing unit 441 synchronizes based on a singular pulse generated by a synchronization pulse, so that, for example, two signals are used to perform temporal synchronization of pulse waves. There is no need to search for a synchronized point by comparing feature point time sequences with relatively shifted. For this reason, it is possible to reduce the time required for the processing for synchronizing the two signals, and consequently, it is possible to reduce the processing time for the attribute determination.

  In addition, since synchronization is performed based on the synchronization pulse in this way, calibration does not need to be performed at the time of manufacture, shipment, etc. of the attribute determination device 400, and the manufacturing process can be shortened.

In addition, in the attribute determination system 1 of the present embodiment, the two signals are synchronized based on the time difference predicted in advance by experiments or the like for the pulse wave acquisition start time difference, the device reaction time difference, and the pulse wave propagation time difference t _FA. . For this reason, the time lag between the two signals is not predicted, and synchronization can be achieved more quickly than when searching for a point to synchronize while shifting the time change of the two signals.

In the attribution determination system 1 of the present embodiment, the time t _Ai of the first feature points A _i has appeared, a second feature point F _j pairwise with the first feature points A _i appeared Collation processing is performed based on the variance value of the difference from time t _Fj . Therefore, whether or not the health information measured by the measurement unit 220 is the health information of the person to be authenticated even when the two signals are not synchronized due to an inappropriate time difference predicted for the pulse wave propagation time difference. Can be accurately determined.

  Further, in the attribute determination system 1 of the present embodiment, the collation unit 440 detects the pressure pulse wave time sequence and the volume pulse wave time sequence, and performs collation processing using the corresponding feature points. Specifically, feature points that appear within a predetermined range of time at which feature points appear are used as corresponding feature points. Therefore, even when one of the feature points cannot be detected due to the influence of noise or the like, the matching process can be performed using the corresponding feature points for the other feature points. For example, in the method of performing matching processing using the interval between feature points, accurate matching cannot be performed when one feature point cannot be detected due to the influence of noise or the like. The detection method according to the present embodiment can be compared more robustly than such a method.

  In the present embodiment, the communication unit 410 transmits the measurement ID generated by the biometric authentication unit 420 to the measurement device 200 via the authentication device 300, but this is not a limitation. For example, the communication unit 410 may transmit the measurement ID directly to the measurement apparatus 200.

  In the present embodiment, the measurement apparatus 200 starts measurement when an operation start instruction is input by the measurement subject, but this is not a limitation. For example, the measurement apparatus 200 may start measurement when the authentication apparatus 300 starts acquiring volume pulse waves.

  Further, in the present embodiment, a learning unit is provided that mechanically learns the propagation time difference due to the difference in body part for acquiring the first and second signals in association with conditions such as age, sex, physique, and health state, The offset time processing unit 441 may perform offset processing using the propagation time difference determined by the learning unit.

  In the present embodiment, the pair determination process, the match determination process, and the variance value determination process are performed. However, a matching result may be output based on one or more of these determination processes.

  In this embodiment, the match determination process is performed when the pair ratio is greater than or equal to the pair determination threshold, and the variance value determination process is performed when the match ratio is less than the match determination threshold. Regardless of the result, all of the pair determination process, the match determination process, and the variance value determination process may be performed, and the collation result may be output based on all these results.

  Although the above-described embodiments and examples have been described as representative examples, it will be apparent to those skilled in the art that many changes and substitutions can be made within the spirit and scope of the invention. Therefore, the present invention should not be construed as being limited by the above-described embodiments and examples, and various modifications and changes can be made without departing from the scope of the claims. For example, a plurality of constituent blocks described in the embodiments and examples can be combined into one, or one constituent block can be divided.

DESCRIPTION OF SYMBOLS 1 Attribution determination system 200 Measuring apparatus 210 Communication part 220 Measurement part 230 1st signal acquisition part 300 Authentication apparatus 310 Communication part 320 Authentication information extraction part 330 2nd signal acquisition part 400 Attitude determination apparatus 410 Communication part 420 Biometrics authentication Unit 430 authentication information storage unit 440 collation unit 441 offset time processing unit 442 feature point calculation processing unit 443 feature point collation processing unit (collation processing unit)

Claims (8)

An attribute determination device that transmits and receives information to and from a measurement device and an authentication device via a communication network,
The health information of the measurement subject measured by the measurement device, the first signal which is the biological information acquired by the measurement device, the authentication information of the verification subject extracted by the authentication device, and the authentication device A communication unit that receives the second signal, which is biometric information,
A biometric authentication unit that authenticates an authenticator based on the authentication information;
Before the measurement device starts to acquire a portion of the first signal used for verification processing, control is performed so that a synchronization pulse is simultaneously applied to the first signal and the second signal. An offset time processing unit that synchronizes the first signal and the second signal based on,
Based on the first signal and the second signal synchronized by the offset time processing unit, it is determined whether the health information measured by the measuring device is the health information of the person to be authenticated. A verification processing unit for performing verification processing;
An attribute determination device comprising: A feature point calculation processing unit that detects a first feature point that is a feature point of the first signal and a second feature point that is a feature point of the second signal, which are synchronized by the offset time processing unit. Further comprising
The collation processing unit performs collation processing based on the number of feature points of the first signal within a predetermined time and the number of feature points of the second signal within the predetermined time. The attribute determination device according to claim 1.   When the second feature point appears in the pair determination error threshold with reference to the time at which the first feature point appears among the first feature points, the matching processing unit The attribute determination apparatus according to claim 2, wherein a combination of the first feature point and a pair of the second feature points is used as a pair, and matching processing is performed based on a pair ratio with respect to the total number of the first feature points of the pair. .   The collation processing unit has a difference between a time when the first feature point appears and a time when the second feature point paired with the first feature point is equal to or greater than a predetermined match determination error threshold. 4. The attribute determination according to claim 3, wherein collation processing is performed based on a match ratio of a pair determined to be less than the match determination error threshold with respect to a total number of the pairs. apparatus.   The collation processing unit calculates a variance value of a difference between the time when the first feature point appears and the time when the second feature point paired with the first feature point appears, and the variance value 5. The attribute determination apparatus according to claim 3, wherein the collation process is performed based on the attribute.   The collation processing unit is configured to obtain health information measured by the measurement device when the pair ratio is equal to or greater than a pair determination threshold, the match ratio is equal to or greater than the match determination threshold, and the variance value is less than a variance value threshold. It is determined that the authentication target is health information, and the pair ratio is less than the pair determination threshold, the match ratio is less than the match determination threshold, or the variance value is greater than or equal to the variance value threshold. 6. The belonging determination apparatus according to claim 5, wherein in some cases, the health information measured by the measurement unit is determined not to be health information of the person to be authenticated. An attribute determination method executed by an attribute determination device that transmits and receives information to and from a measurement device and an authentication device via a communication network,
The health information of the measurement subject measured by the measurement device, the first signal which is the biological information acquired by the measurement device, the authentication information of the verification subject extracted by the authentication device, and the authentication device Receiving a second signal, which is biometric information,
Authenticating an authenticator based on the authentication information;
Before the measurement device starts to acquire a portion of the first signal used for verification processing, control is performed so that a synchronization pulse is simultaneously applied to the first signal and the second signal. Synchronizing the first signal and the second signal based on:
Performing a verification process for determining whether the health information is the health information of the person to be authenticated based on the first signal and the second signal that are synchronized;
A method for determining attribution, comprising: An attribute determination system comprising a measurement device, an authentication device, and an attribute determination device that transmit and receive information to and from each other via a communication network,
The measuring device is
A measuring unit that measures health information including one or more of the biological information and exercise information of the measured person from the measured person;
A first signal acquisition unit that acquires a first signal that is biological information of the measurement subject from the measurement subject,
The authentication device
An authentication information extraction unit for extracting authentication information for extracting authentication information for authenticating the person to be authenticated from the person to be authenticated;
A second signal acquisition unit that acquires a second signal that is biometric information of the person to be authenticated from the person to be authenticated;
The attribute determination device includes:
A communication unit that receives the health information, the first signal, the authentication information, and the second signal;
A biometric authentication unit that authenticates an authenticator based on the authentication information;
Before the measurement device starts to acquire a portion of the first signal used for verification processing, control is performed so that a synchronization pulse is simultaneously applied to the first signal and the second signal. An offset time processing unit that synchronizes the first signal and the second signal based on,
Based on the first signal and the second signal synchronized by the offset time processing unit, it is determined whether the health information measured by the measurement unit is the health information of the person to be authenticated. A verification processing unit for performing verification processing;
An attribute determination system characterized by comprising:

Images