JP6495153B2 - Identity determination system and identity determination method - Google Patents

Description

The present invention relates to identity determining system及 beauty identity determination method.

  For patients with hypertension and chronic diseases, daily, vital information such as blood pressure, blood glucose level, weight, and exercise information such as steps are measured and recorded, and in accordance with the results, review of life and medical intervention It is important to do. Therefore, a PHR (Personal Health Record) system for measuring and recording these vital information and exercise information has been proposed. In the PHR system, in order to prevent false vital information and exercise information from being registered by a third party, when the measured person registers the vital information and exercise information, the person himself / herself is authenticated by ID and password authentication, device authentication, etc. You may have confirmed that. Further, the PHR system may confirm the identity by biometric authentication, knowledge authentication, property authentication, or the like.

  However, these authentications are intended only to prevent the act of impersonation by a third party against the intentions of the principal, and the person himself / herself deliberately passes the information used for the authentication to others. It is easy for the person to stand up as a third party. In addition, it is possible to confirm the identity by performing biometric authentication, but in that case, if a third party's living body is used instead of the person's living body, impersonation can be prevented. Can not. For example, as shown in FIG. 11, after logging in by fingerprint authentication using Mr. A's fingerprint (see FIG. 11 (1)), Mr. B's blood pressure value is measured (see FIG. 11 (2)), and B An unauthorized act of registering the blood pressure value of Mr. A as the value of Mr. A (see FIG. 11 (3)) is permitted.

  Therefore, it is necessary to confirm that the person to be confirmed confirmed by the authentication is the same as the person to be measured whose blood pressure value is measured by the blood pressure device. For this purpose, for example, in Non-Patent Document 1, the identification device that performs authentication and the measurement device that performs measurement collate biometric information, respectively, so that the person who is authenticated by the identification device and the measurement device Techniques have been proposed for confirming that the measured subject is the same person. In this technology, for example, biometric authentication is performed based on biometric information, and the first pulse wave acquired based on the biometric information is acquired by a pulse wave sensor in the vicinity of the upper arm where the blood pressure is measured by the measurement device. When the first pulse wave and the second pulse wave have the same timing when compared with the second pulse wave, the person to be confirmed and the person to be measured are the same.

  In addition, there are a plurality of methods such as fingerprint authentication, vein authentication, face authentication, etc. as methods for performing identity verification, but contactless so as not to physically interfere with the first pulse wave measurement performed simultaneously. A method that can verify the identity is desirable. Therefore, a camera can be considered as a sensor that can perform authentication and measurement of the first pulse wave in a non-contact manner by a sensor of the same sensor. Based on the face image generated by the camera that captures the person to be confirmed, both authentication and measurement of the face pulse wave as the first pulse wave can be performed. Can be guaranteed to belong to the confirmed person. Many face authentications using a camera are known, and there are some attempts to extract facial pulse waves from a moving image including a face region (Non-Patent Documents 2 to 4).

  For example, Non-Patent Document 2 describes that a forehead color that is a part of a face detected by a face detection technique is measured and a facial pulse wave is calculated based on the color. The face pulse wave calculated in this way is accurate compared to the face pulse wave calculated based on other parts of the face, and when the face pulse wave is calculated by measuring the color of the entire face. In comparison, the processing load related to the calculation can be reduced.

  Further, Non-Patent Document 3 discloses a pulse wave signal between a plurality of parts by detecting a pulse wave using only a part of a region such as a jaw for a face region obtained from an image by a face detection technique. It is described that a pulse wave with high accuracy is calculated by compensating for the phase difference.

  Non-Patent Document 4 describes a method of calculating a pulse wave from a face represented in an image with high accuracy even when a person whose face image is captured is moving.

The 19th Annual Meeting of the Japan Society for Medical Informatics, Satoshi Nakamura, “Fundamental Study on Authenticity of Medical and Health Data” M. Lewandowska, J. Ruminski, T. Kocejko, and J. Nowak, "Measuring pulse rate with a webcam-A non-contact method for evaluating cardiac activity," Computer Science and Information Systems (FedCSIS), 2011 Federated Conference on, 2011, p.405_410 Atsuko Tsuji, Kenichi Iwanai, Sakae Saito, "Pulse Wave Measurement Method Using Multiple Face Images," Video Media Processing Symposium, 2013, p.114_115 M.-Z.Poh, D.J.McDuff, and R.W.Picard, "Non-contact, automated cardiac pulse measurements using video imaging and blind source separation," Opt.ExpresS1910, vol.18, no.10, p.10762_10774

  However, when measuring the face pulse wave, a phase difference occurs in the pulse wave for each part of the face, so that there may be a minute fluctuation in the face pulse wave. Thus, when fluctuations are included in the face pulse wave, there is a problem that it is not possible to accurately determine whether the person to be confirmed and the person to be measured are the same.

Accordingly, an object of the present invention made in view of the foregoing, the determination sameness judging system及 beauty identity that makes it possible to accurately determine whether the measured person and the verification party are the same It is to provide a method.

  In order to solve the above problems, an identity determination system according to the present invention is an identity determination system including a facial pulse wave measurement device, a blood pressure measurement device, and a user terminal that transmit and receive information to and from each other via a network. The face pulse wave measurement device includes a camera that captures a person to be confirmed and generates a moving image, and a face area that is an area in which the face of the person to be confirmed is represented from each of the still images constituting the moving image. A face area extracting unit for extracting, a face pulse wave calculating unit for calculating a face pulse wave for each small area based on pixel values of each of the small areas obtained by dividing the face area, and determining a peak time of the whole face pulse wave A blood pressure pulse wave measuring unit that measures a blood pressure pulse wave based on the blood pressure of the person to be measured and determines a peak time of the blood pressure pulse wave, and the user The terminal Characterized in that it comprises a verification processing unit, which performs the same determination of whether the or the object to be sure party and the person to be measured is identical with the click time and peak time of the blood pressure pulse wave.

  The identity determination method according to the present invention is an identity determination method executed by an identity determination system including a face pulse wave measurement device, a blood pressure measurement device, and a user terminal that transmit and receive information to and from each other via a network. In the region where the face of the person to be confirmed is represented from each of the still image constituting the moving image by the step of capturing a moving image by imaging the confirmed person by the camera and the facial pulse wave measuring device. Extracting a certain face area; calculating a face pulse wave based on pixel values of each of the small areas obtained by dividing the face area by the face pulse wave measuring apparatus; and Determining a peak time of the entire facial pulse wave, measuring a blood pressure pulse wave based on the blood pressure of the measurement subject by the blood pressure measurement device, and determining a peak time of the blood pressure pulse wave; The step of performing identity determination as to whether or not the person to be confirmed and the person to be measured are the same using the peak time of the face pulse wave and the peak time of the blood pressure pulse wave by the user terminal, It is characterized by including.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to determine correctly whether a to-be-measured person and a to-be-checked person are the same.

It is a functional block diagram of the identity determination system which concerns on this embodiment. It is a conceptual diagram for demonstrating a face area | region. It is a figure which shows the example of the phase difference of the pulse wave of each area | region with respect to the pulse wave of a reference | standard area | region. It is a conceptual diagram of the phase difference information showing each small area number which a phase difference memory | storage part memorize | stores, and the phase difference with a reference | standard area | region. It is a figure which shows the detail of the blood-pressure measurement part and data processing part which are shown in FIG. It is a conceptual diagram of the information which a PHR memory | storage part memorize | stores. It is a sequence diagram which shows the process of an identity determination system. It is a flowchart which shows the detail of the process in which a phase difference calculation part calculates a phase difference. It is a flowchart which shows the detail of the process in which a phase difference reflection part reflects a phase difference and determines the peak time of a face pulse wave. It is a flowchart which shows the detail of the process in which a collation part performs identity determination. It is a conceptual diagram which shows the unauthorized identity verification in the conventional PHR system.

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

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

  As shown in FIG. 1, the identity determination system 1 includes a facial pulse wave measurement device 200, a blood pressure measurement device 300, and a user terminal 400. Further, the identity determination system 1 may include a PHR server 500. The face pulse wave measuring device 200, the blood pressure measuring device 300, the user terminal 400, and the PHR server 500 are connected via a network and transmit / receive information to / from each other.

<Functional configuration of facial pulse wave measuring device>
The face pulse wave measuring apparatus 200 includes a communication unit 210, a camera 220, an identity confirmation processing unit 230, a face pulse wave measurement unit 240, and a phase difference storage unit 250. The face pulse wave measuring device 200 specifies the user ID of the person to be confirmed based on the face image of the person to be confirmed photographed by the camera 220 and measures the face pulse wave of the person to be confirmed.

  The communication unit 210 transmits information output by the identity verification processing unit 230, information measured by the facial pulse wave measurement unit 240, and the like to the user terminal 400 via the network.

  The camera 220 is built in the face pulse wave measuring apparatus 200 or attached integrally with the face pulse wave measuring apparatus 200, and photographs the person to be confirmed who is the user 2 of the face pulse wave measuring apparatus 200. In addition, the camera 220 continuously captures the person to be confirmed, thereby generating a moving image including a plurality of images, and outputs the moving image to the person confirmation processing unit 230 and the face pulse wave measurement unit 240. The camera 220 may be a color camera or an infrared camera.

  The identity verification processing unit 230 performs identity verification processing based on the image generated by the camera 220. The identity verification process is to specify a user ID for uniquely identifying the person to be confirmed based on the image. In addition, the identity verification processing unit 230 outputs an identified user ID and an identity confirmation result including an identity confirmation date and time that is the date and time when the user ID is identified.

  The face pulse wave measurement unit 240 measures the peak time of the pulse wave of the face of the person to be confirmed based on the image generated by the camera 220 when the person confirmation processing unit 230 performs the person confirmation processing.

  Here, the details of the face pulse wave measurement unit 240 will be described. The facial pulse wave measurement unit 240 includes a phase difference determination unit 241 and a phase difference reflection unit 242.

  The phase difference determining unit 241 includes a face area extracting unit 243, a face area dividing unit 244, a representative value calculating unit 245, and a phase difference calculating unit 246.

  The face area extraction unit 243 extracts a face area that is an area representing the face of the person to be confirmed as shown in FIG. 2A from each of the still images constituting the moving image generated by the camera 220. To do.

  When the face area extracting unit 243 extracts the face area of each still image, the face area dividing unit 244 divides the face area into smaller areas that are smaller areas. Specifically, the face area dividing unit 244 extracts the minimum value and the maximum value of the X coordinate and the minimum value and the maximum value of the Y coordinate of the face area in the still image. Then, the face area dividing unit 244 divides the extracted X coordinate between the minimum value and the maximum value, for example, into ten, and similarly divides the extracted Y coordinate between the minimum value and the maximum value, for example, into ten. To divide into 10 × 10 small areas. In the following description, the coordinates for identifying the position of each small area within the area are represented by (x, y), and 10x + y is the small area number. In the example shown in FIG. 2A, for example, the small area number of the small area indicated by coordinates (7, 5) is 75, and the small area number of the small area indicated by coordinates (8, 6) is 86.

  The representative value calculation unit 245 calculates the representative value of the pixels constituting each small area divided by the face area dividing unit 244. The representative value is, for example, a statistical value (for example, an addition average value, a weighted average value, a median value, a mode value, etc.) of pixel values of each small area divided into face areas.

  The representative value calculation unit 245 calculates the representative value of each small region for each still image constituting the moving image as shown in FIG. 2B, and the representative value in the order in which the still images are captured as shown in FIG. The face pulse wave represents the transition of the. FIG. 3A shows a reference pulse which is a predetermined small reference area, for example, a facial pulse wave at the center of the left cheek. FIG. 3B shows a facial pulse wave in a small area different from the reference area. There is a phase difference between the face pulse wave shown in FIG. 3A and the face pulse wave shown in FIG.

  The phase difference calculation unit 246 compares the face pulse wave of the reference region calculated by the representative value calculation unit 245 with the face pulse wave of a small region different from the reference region, and calculates the phase difference. In the example shown in FIG. 2, the region of the small region number 75 is set as the reference region, the face pulse waves of the small region numbers 0 to 74 and 76 to 99 are compared with the face pulse wave of the reference region, and the phase difference is calculated. calculate. The face pulse wave shown in FIG. 3 is that of the person to be confirmed whose pulse wave is about 60. In this case, about 5 peaks appear in the moving image for 6 seconds. The phase difference calculation unit 246 calculates the difference between the peak time of the face pulse wave of the reference region and the peak time of the face pulse wave of the corresponding small region for the five peaks, and The average value is calculated as a phase difference. In the following description, the peak time in the reference region is referred to as “reference peak time”.

  Note that, as shown in small area numbers 0, 80, 90, etc. in FIG. 2, face pulse waves are not calculated for small areas that do not represent part of the face. Therefore, the phase difference calculation unit 246 calculates a number or symbol indicating that the face pulse wave is not calculated in association with the small area number for a small area in which the face pulse wave is not calculated, that is, the representative value of the pixel value does not change. Etc. (“9999” in the example shown in FIG. 4). Thereafter, the face pulse wave cannot be calculated, that is, a small area where the representative value of the pixel value does not change is referred to as “non-target area”.

  The phase difference reflection unit 242 includes a facial pulse wave calculation unit 247, a correction unit 248, and a peak determination unit 249.

  When a facial pulse wave measurement start command for starting measurement of facial pulse waves is transmitted, the facial pulse wave calculation unit 247 performs division by the facial region division unit 244 based on the moving image generated by the camera 220. The face pulse wave is calculated for each small area.

  The correcting unit 248 calculates each peak time for one or more peaks included in each face pulse wave calculated by the face pulse wave calculating unit 247, and stores the small region of the small region in the phase difference storage unit 250. Correction is made for the phase difference stored in association with the number, and each corrected peak time is obtained.

  The peak determination unit 249 determines the peak time of the entire face pulse wave of the person to be confirmed based on the corrected peak time obtained by correcting the peak time of each small region. As described above, since the face pulse wave has a different phase for each small area, the peak time differs for each small area. Therefore, the correction unit 248 corrects the difference in peak time based on the difference in the position of the small region, and determines the peak time of the face pulse wave based on the corrected peak time. Similarly, the peak determination unit 249 determines each peak time of the entire face pulse wave for each corrected peak time of a plurality of peaks included in the small region.

  Specifically, the peak determination unit 249 compares the peak time of the reference area with the corrected peak time of each small area. At this time, the non-target region may be excluded from each small region to be compared. Then, the peak determination unit 249 determines whether or not the corrected peak time of a small region equal to or higher than a predetermined ratio (for example, 50%) among all the small regions to be compared is the same as the peak time of the reference region. Determine. When it is determined that the correction peak time of the small region equal to or greater than the predetermined ratio is the same as the peak time of the reference region, the correction peak time of the reference region is determined as the peak time of the entire face pulse wave.

  In addition, when it is determined that the peak times of the small regions of a predetermined ratio or more are not the same as the peak times of the reference region, the peak determination unit 249 performs histograms of the peak times of all the small regions except the reference region and the non-target region. And whether or not the same correction peak time is reached in a small region of a predetermined ratio or more is determined. When it is determined that the same correction peak time is reached in a small region of a predetermined ratio or more, the correction peak time is determined as the peak time of the face pulse wave of the person being confirmed. When it is determined that the same corrected peak time is not reached in a small region of a predetermined ratio or more, the peak time of the face pulse wave of the person to be confirmed is not determined, and the peak time related to the peak is determined by the identity determination described later. Decided not to be targeted.

  Moreover, the peak determination part 249 determines peak time similarly about the peak which appears in multiple times during a measurement.

  Further, the phase difference storage unit 250 stores the phase difference calculated by the phase difference calculation unit 246 in association with the small area number as shown in FIG.

<Functional configuration of blood pressure measurement device>
The blood pressure measurement device 300 includes a communication unit 310, a blood pressure measurement unit 320, a blood pressure pulse wave measurement unit 330, and a data processing unit 340. The blood pressure measurement device 300 measures a blood pressure value including systolic blood pressure and diastolic blood pressure of the measurement subject and a measurement value for each measurement type that is a pulse, and the blood pressure pulse wave of the measurement subject based on these measurement values. Measure.

  The communication unit 310 transmits the information measured by the blood pressure measurement unit 320 and the information measured by the blood pressure pulse wave measurement unit 330 to the user terminal 400 via the network.

  Further, when the pressure inside the cuff 321 included in the blood pressure measurement unit 320 (hereinafter referred to as “cuff pressure”) becomes higher than a threshold value, the communication unit 310 transmits a high pressure state notification indicating that to the user terminal 400. In addition, the communication unit 310 receives from the user terminal 400 a pulse wave measurement start command that is a command for starting measurement of a blood pressure pulse wave. In addition, the communication unit 310 controls the valve 324 to open to discharge air from the cuff 321 and to control to close the valve 324 and stop discharging air from the cuff 321. A valve close command is transmitted to the valve 324.

  As shown in FIG. 5, the blood pressure measurement unit 320 is realized by an automatic sphygmomanometer that includes a cuff 321, a pressure sensor 322, a pump 323, and a valve 324.

  The pressure sensor 322 detects the time change of the cuff pressure in which the pulse intensity is expressed as pressure data. The pump 323 raises the cuff pressure to a pressure sufficiently higher than the systolic blood pressure by sending air into the cuff 321. The valve 324 reduces the cuff pressure by discharging air from the cuff 321.

  Based on the pressure data detected by the pressure sensor 322, the blood pressure measurement unit 320 measures a blood pressure value including systolic blood pressure and diastolic blood pressure, and a pulse as measurement values. Specifically, the blood pressure measurement unit 320 sets the cuff pressure at the time when the pulse intensity related to the pressure data detected by the pressure sensor 322 first exceeds a predetermined value as the systolic blood pressure, and then again The cuff pressure at the time when the pulse intensity falls below a predetermined value is measured as the diastolic blood pressure.

  In addition, the blood pressure measurement unit 320 outputs a measurement result that is information including systolic blood pressure, diastolic blood pressure, blood pressure value, a measurement type indicating a pulse type, a measurement value that is a value for each measurement type, and a measurement date and time. To do.

  When the communication unit 310 receives a pulse wave measurement start command from the user terminal 400, the blood pressure pulse wave measurement unit 330 measures the blood pressure pulse wave based on the pressure data detected by the pressure sensor 322, and the blood pressure pulse wave peak time To decide.

  In addition, when measurement by the blood pressure measurement unit 320 is started, the data processing unit 340 transmits a pressure data inquiry for acquiring pressure data to the pressure sensor 322. In addition, the data processing unit 340 acquires the pressure data transmitted from the pressure sensor 322 based on the pressure data inquiry.

<Functional configuration of user terminal>
The user terminal 400 includes a communication unit 410 and a verification processing unit 420. The user terminal 400 collates the face pulse wave measured by the face pulse wave measuring apparatus 200 with the blood pressure pulse wave measured by the blood pressure measuring apparatus 300.

  The communication unit 410 receives the identification result and the peak time of the whole face pulse wave from the face pulse wave measuring apparatus 200 via the network. Further, the communication unit 410 receives the measurement result and the peak time of the blood pressure pulse wave from the blood pressure measurement device 300 via the network. In addition, the communication unit 410 transmits the identity verification result, the measurement result, and the verification result determined by the verification processing unit 420 to the PHR server 500.

  Based on the control by the data processing unit 340, the collation processing unit 420 uses the peak time of the whole face pulse wave and the peak time of the blood pressure pulse wave to determine whether the person to be confirmed and the person to be measured are the same. Perform sex determination. Specifically, the matching processing unit 420 calculates an identity probability P representing the probability that the person to be confirmed related to the face pulse wave received by the communication unit 410 and the person to be measured related to the blood pressure pulse wave are the same, It is determined whether or not the person to be confirmed and the person to be measured are the same. More specifically, the matching processing unit 420 determines whether or not the identity probability P exceeds a predetermined identity confirmation threshold value. If the identity probability P exceeds a predetermined identity confirmation threshold value, And the measurement subject are the same, and the collation result “OK” is output. Further, when the identity probability P is equal to or lower than the identity confirmation threshold, the matching processing unit 420 determines that the person to be confirmed and the person to be measured are not the same, and outputs a matching result “NG”.

For the identity probability P, for example, the larger one of the variance value DΔT of the time difference ΔT between the peak time of the facial pulse wave and the peak time of the blood pressure pulse wave and the reference variance value Db is set to D _L , The smaller one is expressed as Equation (1), where D _S is the one. Here, the reference variance value Db is obtained by learning the variance of the time difference ΔT when the person being measured and the person being confirmed are the same in the identity determination for a certain number of people, that is, the collation result is “OK”. Value. In addition, the reference variance value Db was measured in a state where it is ensured that the person to be confirmed and the person to be measured are the same, for example, by the presence of a service provider who provides the identity determination system 1 of the present embodiment. It is good also as a value which learned dispersion | distribution of time difference (DELTA) T of the peak time of a face pulse wave and a blood pressure pulse wave. The time difference ΔT may be a difference between the time when the facial pulse wave peaks and the time when the blood pressure pulse wave peaks.

P = D _S / D _L (1)

<Functional configuration of PHR server>
As shown in FIG. 1, the PHR server 500 includes a communication unit 510 and a PHR storage unit 520. The PHR server 500 receives the identity verification result, the measurement result, and the verification result from the user terminal 400, and the PHR storage unit 520 stores the information.

  The communication unit 510 receives the identity verification result, the measurement result, and the verification result from the user terminal 400 via the network.

  The PHR storage unit 520 stores the identity verification result, the verification result, and the measurement result received from the user terminal 400 by the communication unit 510. FIG. 6 is a conceptual diagram showing information stored in the PHR storage unit 520. The PHR storage unit 520 stores the user ID and the identity confirmation date / time included in the identity verification result, the measurement type, the measurement value, and the measurement date / time included in the measurement result in association with each other.

<Operation of identity determination system>
Then, with reference to FIG. 7, operation | movement of the identity determination system 1 which concerns on this embodiment is demonstrated.

  First, the identity verification processing unit 230 of the face pulse wave measuring apparatus 200 performs an identity verification process by face authentication (step S10), and the communication unit 210 transmits the identity verification result to the user terminal 400 (step S11). When the personal identification process is performed in step S10, the camera 220 captures the face of the person to be confirmed and starts generating a moving image (step S12).

  On the other hand, the data processing unit 340 of the blood pressure measurement device 300 starts blood pressure measurement (step S13). When blood pressure measurement starts, the pump 323 is turned on (step S14), and air is sent into the cuff 321. When blood pressure measurement is started, the data processing unit 340 transmits a pressure data inquiry for acquiring pressure data representing the cuff pressure to the pressure sensor 322 (step S14). In step S14. When the pressure data inquiry is transmitted from the data processing unit 340, the pressure sensor 322 measures the cuff pressure and transmits the pressure data to the data processing unit 340 as needed (steps S15-1, S15-2,...). .., S15-N). Further, the data processing unit 340 outputs a measurement result based on the pressure data received from the pressure sensor 322.

  When the pressure data is transmitted from the pressure sensor 322 in steps S15-1, 15-2,... 15-N, the data processing unit 340 determines whether the cuff pressure is higher than the cuff pressure threshold based on the pressure data. When the cuff pressure becomes higher than the cuff pressure threshold (step S16), the pump 323 is turned off (step S17) and stops feeding air into the cuff 321. Further, the valve 324 is turned on (step S18), and air is discharged from the cuff 321.

  In step S17, when the cuff pressure becomes higher than the cuff pressure threshold, the communication unit 310 transmits a high pressure state notification to the user terminal 400 (step S19).

  In addition, after the generation of the moving image in which the face is imaged is started in step S12, the phase difference determination unit 241 determines the phase difference of the face pulse wave of each small region with respect to the face pulse wave of the reference region for the moving image. (Step S20).

  Here, details of the process in which the phase difference determination unit 241 determines the phase will be described with reference to FIG. First, the face area extraction unit 243 extracts a face area from each still image constituting the moving image generated by the camera 220 (step S201). When the face area is extracted in step S201, the face area dividing unit 244 divides the extracted face area into small areas (step S202). When the face area is divided into small areas in step S202, the representative value calculation unit 245 calculates the representative value of the pixels constituting each small area and calculates the face pulse wave for each small area (step S203). ).

  When the facial pulse wave is calculated in step S203, the phase difference calculation unit 246 calculates the phase difference of the facial pulse wave in each small region with respect to the facial pulse wave in the reference region (step S204). Here, the calculation of the phase difference related to the facial pulse wave in the non-target region may not be performed. Further, the phase difference storage unit 250 stores the phase difference calculated by the phase difference calculation unit 246 in association with the small area number.

  Returning to FIG. 7, when the high pressure state notification is transmitted in step S19, the communication unit 410 of the user terminal 400 transmits a pulse wave measurement start command to the communication unit 210 of the facial pulse wave measurement device 200 (step S21). . In addition, the communication unit 410 transmits a pulse wave measurement start command to the communication unit 310 of the blood pressure measurement device 300 (step S22).

  When the pulse wave measurement start command is transmitted in step S21, the identity verification processing unit 230 of the face pulse wave measurement device 200 performs the identity verification process by face authentication (step S23), and transmits the identity verification result to the user terminal 400. (Step S24). The identity verification process may be performed a plurality of times at random timing after the identity verification process in step S10. As a result, it is possible to prevent the person to be authenticated from being replaced once the identity verification is performed, and the PHR server 500 can more reliably manage the measurement result of the person to be authenticated.

  Further, as described above, the data processing unit 340 has the cuff pressure higher than the cuff pressure threshold based on the pressure data transmitted from the pressure sensor 322 in steps S15-1, 15-2,..., 15-N. Whether the cuff pressure is lower than the cuff pressure threshold (step S25), the valve 234 is turned off (step S26), and the discharge of air from the cuff 321 is stopped. Further, when the cuff pressure becomes lower than the cuff pressure threshold value in step S25, the data processing unit 340 transmits a low pressure state notification to the user terminal 400 (step S27).

  When the low pressure state notification is transmitted in step S27, the communication unit 410 of the user terminal 400 transmits a pulse wave measurement end command to the face pulse wave measurement device 200 (step S28), and the pulse wave measurement end command is measured. It transmits to the apparatus 300 (step S29).

  When the pulse wave measurement end command is transmitted in step S28, the face pulse wave is generated based on the moving image generated by the camera 220 from the pulse wave measurement start command in step S21 to the pulse wave measurement end command in step S28. The measurement unit 240 performs processing for determining the peak time of the facial pulse wave by reflecting the phase difference stored in the phase difference storage unit 250 (step S30).

  Here, with reference to FIG. 9, the face pulse wave measurement unit 240 measures the face pulse wave and determines each peak time of one or more peaks included in the face pulse wave in detail. First, based on the moving image generated by the camera 220, the face pulse wave calculation unit 247 calculates the representative value of the pixels constituting the small region for each of the small regions divided by the face region dividing unit 244 (step S301). . When the representative value is calculated in step S301, the facial pulse wave calculation unit 247 calculates a reference peak time (step S302). When the reference peak time is calculated in step S302, the correction unit 248 corrects the peak time of each facial pulse wave calculated by the facial pulse wave calculation unit 247 to obtain a corrected peak time (step S303).

  It is determined whether or not 50% or more of the corrected peak time of the small area excluding the non-target area, which is corrected in step S303, is the same as the peak time of the reference area (step S303). S304). If it is determined in step S304 that the corrected peak time of the small region of 50% or more is the same as the peak time of the reference region, the peak time of the reference region is determined as the peak time of the facial pulse wave (step S305). If it is determined in step S305 that the correction peak time of the small area of 50% or more is not the same as the peak time of the reference area, the peak determination unit 249 determines that all the small areas except the reference area and the non-target area are A histogram of corrected peak times is generated (step S306).

  When a histogram is generated in step S306, it is determined whether or not the same correction peak time is taken in a small region of 50% or more (step S307). If it is determined in step S307 that the same correction peak time is reached in a small region of 50% or more, the correction peak time is determined as the peak time of the face pulse wave (step S308). When it is determined in step S307 that the same peak time is not reached in a small region of 50% or more, it is determined that the peak related to the peak time is not subject to identity determination described later (step S309).

  When the process of step S305, S308, or S309 is completed, it is determined whether or not there is another peak in the face pulse wave calculated by the face pulse wave calculation unit 247 (step S310). Returns to step S302 and repeats the process. If there is no other peak, the process for measuring the face pulse wave is terminated.

  Returning to FIG. 7, when the process of measuring the facial pulse wave including the reflection of the phase difference is completed in step S30, the communication unit 210 transmits each peak time of the entire facial pulse wave to the communication unit 410 (step S31). .

  On the other hand, when the communication unit 310 receives a pulse wave measurement start command from the communication unit 410 in step S29, the blood pressure pulse wave measurement unit 330 starts measuring the blood pressure pulse wave, and the communication unit 310 outputs the data by the data processing unit 340. The measured measurement result and the peak time of the blood pressure pulse wave measured by the blood pressure pulse wave measuring unit 330 are transmitted to the communication unit 410 (step S32).

  When the communication unit 410 receives the peak time of the face pulse wave transmitted from the communication unit 210 in step S32, and receives the measurement result and the peak time of the blood pressure pulse wave transmitted from the communication unit 310 in step S32, a matching process is performed. The unit 420 calculates the identity probability P based on the peak times of the facial pulse wave and the blood pressure pulse wave (step S33). Furthermore, the collation processing unit 420 performs identity determination based on the identity probability P calculated in step S26 (step S34).

  Here, the identity determination process will be described in detail with reference to FIG. First, the matching processing unit 420 determines whether or not the identity probability P calculated in step S33 exceeds the identity confirmation threshold (step S341). When it is determined that the identity probability P exceeds the identity confirmation threshold, the matching processing unit 420 outputs a matching result “OK” (step S342). When it is determined that the identity probability P is equal to or less than the identity confirmation threshold, the matching processing unit 420 outputs a matching result “NG” (step S343).

  Returning to FIG. 7, in this way, when the identity processing is completed by the collation processing unit 420 in step S <b> 34, the communication unit 410 transmits the measurement result, the identity confirmation result, and the collation result to the communication unit 510 (step). S35). Then, the PHR storage unit 520 stores the measurement result, the identity confirmation result, and the verification result received by the communication unit 510 (step S36).

  As described above, in the identity determination system 1 according to the present embodiment, based on the pixel value of each small area obtained by dividing the face area extracted from each still image constituting the moving image generated by the camera, the facial pulse Since the peak time of the wave is determined, it is possible to prevent the fine fluctuation due to the phase difference of the facial pulse wave for each facial part from occurring in the facial pulse wave. Thereby, it is possible to accurately determine whether or not the person to be confirmed and the person to be measured are the same.

  In the present embodiment, the representative value calculation unit 245 calculates the statistical value of the pixel value as a representative value for each of the plurality of skin color portions shown in the face area. However, it is assumed that hemoglobin absorbs among the pixel values G ( The statistical value of the (Green) value may be calculated as a representative value. Further, the representative value calculation unit 245 may calculate a luminance value or a statistical value of brightness as a representative value.

  In the present embodiment, the face area extraction unit 243 may extract the area of the face area and the positions of feature points such as eyes for each still image constituting the moving image generated by the camera 220. In this case, the face area extraction unit 243 determines that the face has moved when the area of the face area and the position of the feature point change, and based on the change in the area of the face area and the position of the feature point, the face area coordinates The face area dividing unit 244 may divide the corrected face area into small areas. Thereby, when the person to be confirmed is imaged to determine the peak time of the entire facial pulse wave, the person to be confirmed does not need to remain stationary and can be comfortably spent.

  In addition, depending on the environment in which the face image is captured, noise may be included in the pixel values constituting the image captured by the camera 220. Therefore, the phase difference calculation unit 246 compares the calculated phase difference of each small region with the phase difference of the adjacent small region, and if the value is reversed, the phase difference calculating unit 246 calculates the small difference related to the reversed phase difference. The area may be a non-target area. As a result, it is possible to prevent the phase difference reflection unit 242 from performing processing based on the phase difference calculated based on the pixel value including noise, and to accurately determine the peak time of the entire face pulse wave. It becomes.

  In the present embodiment, the phase difference calculation unit 246 may calculate the size of the face of the person to be confirmed from the distance between the camera 220 and the person to be confirmed and the focal length of the camera 220. In this case, the phase difference calculation unit 246 determines the range of the phase difference that cannot be generated based on the calculated face size and the standard blood flow velocity, and is calculated as described above. When the phase difference of the small area is within this range, the small area is set as a non-target area. Thereby, it is possible to prevent the phase difference reflection unit 242 from performing processing based on an erroneous calculation due to the influence of noise or the like, and it is possible to accurately determine the peak time of the entire face pulse wave. .

  In addition, although the person confirmation processing unit 230 of the face pulse wave measuring device 200 performs the person confirmation processing, the present invention is not limited to this, and an image representing a face is transmitted to an external device connected via a communication network, The device may perform identity verification processing based on the image.

  In the present embodiment, the face pulse wave measuring unit 240 of the face pulse wave measuring apparatus 200 calculates the phase difference and reflects the phase difference, but the present invention is not limited to this. For example, a moving image captured by the camera 220 of the facial pulse wave measuring apparatus 200 may be transmitted to the user terminal 400, and the user terminal 400 may calculate the phase difference and reflect the phase difference based on the moving image.

  In the present embodiment, the identity probability P is represented by the peak time of the entire face pulse wave, but may be represented by the time at which the entire face pulse wave rises.

  In this embodiment, the face of the person to be confirmed may be three-dimensionally represented by the camera 220 in three dimensions, and the surface of the face of the person to be confirmed represented in three dimensions may be divided into small regions.

  In the present embodiment, the PHR storage unit 520 stores the identity verification result, the verification result, and the measurement result received from the user terminal 400 by the communication unit 510, but this is not restrictive. For example, the PHR storage unit 520 stores the identity confirmation result and the measurement result when the collation result is “OK”, and does not store the identity confirmation result and the measurement result when the collation result is “NG”. Also good.

  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 Identity determination system 200 Face pulse wave measuring device 210 Communication part 220 Camera 230 Identity confirmation process part 240 Face pulse wave measurement part 241 Phase difference determination part 242 Phase difference reflection part 243 Face area extraction part 244 Face area division part 245 Representative value Calculation unit 246 Phase difference calculation unit 247 Face pulse wave calculation unit 248 Correction unit 249 Peak determination unit 250 Phase difference storage unit 300 Blood pressure measurement device 310 Communication unit 320 Blood pressure measurement unit 321 Cuff 322 Pressure sensor 323 Pump 324 Valve 330 Blood pressure pulse wave measurement Unit 340 data processing unit 400 user terminal 410 communication unit 420 collation processing unit 500 PHR server 510 communication unit 520 PHR storage unit

Claims (6)

An identity determination system comprising a facial pulse wave measurement device, a blood pressure measurement device, and a user terminal that transmit and receive information to and from each other via a network,
The facial pulse wave measuring device is
A camera that captures the person to be confirmed and generates a moving image;
A face area extraction unit that extracts a face area that is an area in which the face of the person to be confirmed is represented from each of the still images constituting the moving image;
A face pulse wave calculation unit that calculates a face pulse wave for each small area based on the pixel values of each of the small areas obtained by dividing the face area;
A peak determining unit for determining a peak time of the whole face pulse wave,
The blood pressure measurement device includes:
A blood pressure pulse wave measurement unit that measures a blood pressure pulse wave based on the blood pressure of the measurement subject and determines a peak time of the blood pressure pulse wave;
The user terminal is
A verification processing unit that performs identity determination as to whether or not the person to be confirmed and the person to be measured are the same using the peak time of the face pulse wave and the peak time of the blood pressure pulse wave;
An identity determination system comprising: The facial pulse wave measuring device is
A phase difference calculating unit that calculates a phase difference of a face pulse wave of a small region different from the reference region with respect to a face pulse wave of the reference region of the small region;
A correction unit that corrects the peak time of the facial pulse wave of each of the small regions calculated by the facial pulse wave calculation unit based on the phase difference;
The identity determination system according to claim 1, wherein the peak determination unit determines a peak time of the entire face pulse wave based on the peak time corrected by the correction unit. The peak determination unit determines whether a peak time of a predetermined ratio or more out of the peak times of each of the small areas is the same as the peak time of the reference area, and the peak time of the predetermined area or more is 3. The identity determination system according to claim 2 , wherein when the peak time is the same as the peak time of the reference region, the peak time of the reference region is determined as a peak time of the whole face pulse wave. The peak determination unit determines whether a peak time of a predetermined ratio or more out of the peak times of each of the small areas is the same as the peak time of the reference area, and the peak time of the predetermined area or more is If it is not the same as the peak time of the reference area, it is determined whether or not the face pulse waves of a small area of a predetermined ratio or more have the same peak time, and the face pulse waves of a small area of a predetermined ratio or more are the same The peak time is determined as the peak time of the entire face pulse wave, and it is determined that the face pulse waves in a small area of a predetermined ratio or more do not have the same peak time. The identity determination system according to claim 2 , wherein a peak time of the whole face pulse wave is not determined. A representative value calculating unit that calculates a representative value that is a statistical value of the pixel value of each of the small regions, and calculates a transition of the representative value as the facial pulse wave in the order in which the still image is captured,
5. The identity determination system according to claim 2, wherein the phase difference calculation unit calculates a phase difference of the face pulse wave of the small region with respect to the face pulse wave of the reference region. . An identity determination method executed by an identity determination system including a facial pulse wave measurement device, a blood pressure measurement device, and a user terminal that transmit and receive information to and from each other via a network,
Capturing a confirmed person with a camera and generating a moving image;
Extracting a face area that is an area in which the face of the person to be confirmed is represented from each of the still images constituting the moving image by the facial pulse wave measurement device;
Calculating a facial pulse wave based on the pixel values of each of the small regions obtained by dividing the facial region by the facial pulse wave measurement device;
Determining a peak time of the whole face pulse wave by the face pulse wave measuring device;
Measuring a blood pressure pulse wave based on the blood pressure of the person to be measured by the blood pressure measurement device, and determining a peak time of the blood pressure pulse wave;
Determining whether or not the person to be confirmed and the person to be measured are the same using the peak time of the face pulse wave and the peak time of the blood pressure pulse wave by the user terminal, and The identity determination method characterized by including.

Images