JP7237917B2 - Biometric device - Google Patents

Description

The present invention relates to a biometric authentication device, a biometric authentication system, and a mobile terminal.

In recent years, the health care market is expected to expand due to the rise in consumer awareness of health (disease prevention) against the backdrop of advances in IT technology and an aging society. Under such circumstances, the spread of wearable terminals capable of constantly monitoring the state of health with high accuracy is expected.

In addition, with the spread of mobile terminals such as mobile phones and smart phones, Internet shopping, Internet banking, and electronic payment using them have become possible. Along with this, in order to prevent unauthorized use of mobile terminals, products that apply biometric authentication technology such as fingerprint, vein, and iris have also appeared. However, it has been found that these biometric authentication technologies can be spoofed using gelatin or radish, and security vulnerabilities have been pointed out.

Against this background, in recent years, an authentication technique using an electrocardiogram technique is appearing. Electrocardiogram technology has the same level of authentication accuracy as fingerprints and veins, but it also has the advantage of being capable of biometric detection (confirmation of life) and making spoofing difficult.

Electrocardiogram technology is also an effective indicator for detecting heart disease, one of the three major adult diseases, in healthcare applications, and is also an important indicator for analyzing stress and fatigue levels, which have been attracting increasing attention within companies in recent years. there is

In addition, there is an authentication technology that uses pulse waves as an authentication technology that has the same features as the electrocardiogram technology. From the point of view of healthcare, the pulse wave is useful as an indicator of arteriosclerosis as well as heart disease. However, the electrocardiogram is a more accurate and important indicator than the pulse wave in diagnosing heart disease. Other features are similar to ECG technology.

Although electrocardiogram technology varies from person to person, the accuracy of authentication varies due to factors such as physical condition. Therefore, a mechanism for maintaining a high authentication rate regardless of physical condition is required. Fingerprints and finger veins are basically authenticated by capturing one image, but an electrocardiogram is biometric information that changes periodically over time and takes time to authenticate, so it is necessary to avoid authentication failures as much as possible.

An information terminal device having an authentication function using an electrocardiogram has a technique described in Patent Document 1. In this publication, "an input unit, a physical condition detection unit for detecting the physical condition of a subject, a template holding unit for holding a plurality of templates, an operation input to the input unit, and based on the template, the subject and an authentication unit that authenticates the template, and the template differs according to the physical condition."

Further, there is a technique described in Patent Document 2 for a personal authentication method, personal authentication system, and biometric information measurement system using pulse waves. This publication states that "biological information that changes periodically is stored in association with environmental parameters such as physical conditions such as the amount of exercise or body temperature of the person to be identified, and is stored in association with the environmental parameters detected at the time of authentication. By performing authentication using the biometric information obtained from the user, it is possible to perform personal identification with high accuracy regardless of fluctuations in environmental parameters."

Japanese Patent Application Laid-Open No. 2013-239125 Japanese Patent Application Laid-Open No. 2007-213196

However, Japanese Patent Laid-Open No. 2002-200000 does not describe a template creation/registration method according to the physical condition. Also, there is no description of a physical condition detection method when registering a template.

Further, in Patent Document 2, another sensor (for example, an acceleration sensor) for measuring information other than the biological information to be measured is used to detect an environmental parameter that is a factor of physical condition fluctuation. Therefore, the number of parts in the device increases. In addition, templates are registered even when physical condition fluctuation factors detected by the sensor do not necessarily correlate with the biological information at that time, so the number of unnecessary templates increases. For this reason, there is a risk of increasing the false acceptance rate while improving the personal authentication rate.

In the present invention, a variation pattern indicating a change in physical condition is automatically extracted from biometric information previously measured for healthcare, and a template suitable for registration is selected and registered from the extracted variation pattern to improve authentication accuracy. With the goal.

The biometric authentication device of the present invention includes a biometric information measurement unit that measures biometric information of a user, a feature amount extraction unit that extracts a feature amount of biometric information, and a variation pattern of biometric information from the biometric information stored in a storage unit. a variation pattern extraction unit that extracts and obtains the feature amount; a template registration determination unit that determines whether the feature amount of the variation pattern of the biometric information extracted by the variation pattern extraction unit can be registered; the feature amount extraction unit; A template management unit that manages a template recording the feature amount of the biometric information extracted by the variation pattern extraction unit, and a template managed by the template management unit that matches the feature amount extracted by the feature amount extraction unit and , and an authentication unit that determines whether or not the user is authenticated.

The present invention is also understood as a biometric authentication system using the biometric authentication device and a mobile terminal.

According to the present invention, a variation pattern indicating a change in physical condition is automatically extracted from biological information measured in advance for healthcare purposes, and a template suitable for registration is selected from the extracted variation pattern and registered, thereby improving authentication accuracy. can do.

2 is a functional block diagram of the biometrics authentication system of the embodiment; FIG. 1 is a hardware configuration diagram of a biometrics authentication system according to an embodiment; FIG. FIG. 4 is a schematic diagram of processing up to creation of an electrocardiogram template; It is a general flow of electrocardiogram authentication. This is the type of the feature amount of the electrocardiogram. FIG. 10 is a screen IF for initial registration processing of the biometric authentication device of the embodiment; FIG. It is biometric measurement information (electrocardiogram) in the biometric information storage unit. This is the configuration of a template (electrocardiogram). This is the configuration of the template management table. 4 is a flow of initial registration processing of the biometric authentication device of the embodiment; It is a screen and a registration table for user setting information and the like. It is an example of a risk evaluation table. This is the configuration of the additional template maximum value management table. 4 is a biometric authentication/biometric information measurement flow of the biometric authentication device of the embodiment. It is an example of the screen IF at the time of electrocardiogram authentication of the biometric authentication device of the present embodiment. It is an example of the screen IF for performing electrocardiogram measurement according to a user instruction of the biometric authentication device of the present embodiment. It is a flow of variation pattern extraction processing. 10 is a processing flow for calculating the upper limit of the number of templates that can be added; 6 is a flow of template registration processing; It is an example of a template registration notification screen IF. It is a hardware configuration of a biometrics authentication device to which pulse wave technology is applied. It is a wristband-type biometric authentication device that uses electrocardiogram technology. It is a shirt-type biometric authentication device that uses electrocardiogram technology. It is a smartphone built-in authentication device that uses electrocardiogram technology. It is a ring type authentication device using electrocardiogram technology. This is an earphone-type authentication device that uses pulse waves. It is an ear clip type authentication device that uses pulse waves. This is an example of a finger guide for a smartphone built-in authentication device using pulse wave technology. FIG. 11 is another functional block diagram of the biometric authentication device (registration selection using attribute information). FIG. 10 is a hardware configuration diagram of a biometric authentication device (addition of an attribute data acquisition sensor); It is a configuration example of an attribute data type table. It is biometric information (electrocardiogram) in the biometric information storage unit. It is a variable pattern reference table. It is a variation pattern extraction result. It is a correlation coefficient calculation formula. A template management table. It is a flow of pattern type analysis. It is a configuration example of the biometric authentication system of the present embodiment. Attribute data type table (peripheral information added). It is a connection configuration of a biometric authentication device, an external sensor, a smartphone, and a server. 2 is a processing flow of an attribute information acquisition unit (selects attribute data to be acquired from a biological information measuring device or sensor); It is a flow of biometric information variation pattern extraction processing in a server on a smartphone/cloud. It is a processing flow of an attribute information acquisition part. This is the configuration of a biometric authentication system with a disease diagnosis function. This is the waveform feature quantity by pattern. It is a configuration of a disease diagnosis result management table. It is an example of an electrocardiogram change at the onset of angina pectoris. It is a configuration of a variation pattern management table (biometric authentication device side). It is a rescue cooperator registration screen IF. It is a fluctuation pattern management table (in the disease countermeasure study department). It is a processing sequence when the onset of myocardial infarction is detected. It is an example of information transmitted from the biometric authentication device when a disease is detected. Fig. 10 is an IF of a diagnosis result notification screen when an arrhythmia develops; It is an example of a medical service and insurance service collaboration application using the biometric authentication device and system of the present embodiment. This is an example of an annuity payment service using the biometric authentication device and system of this embodiment.

An embodiment will be described below with reference to the drawings.

First, an electrocardiogram will be briefly described with reference to FIG. An electrocardiogram is a record of the electrical activity of the heart as a waveform from the body surface, and expresses the movement of the heart. The electrocardiogram is shown with potential (mV) on the vertical axis and time (ms) on the horizontal axis. FIG. 5B shows the basic waveform of an electrocardiogram. Waveforms of an electrocardiogram include P wave (541), Q wave (542), R wave (543), S wave (544), T wave (545), and U wave (546). A QRS wave (547) is a combination of the Q wave (542), R wave (543), and S wave (544). Each wave has a height and a time width.

The basic waveform of the electrocardiogram in FIG. 5(B) shows one beat of the heart (from the beginning of the P wave to the beginning of the next P wave). Therefore, the result of the electrocardiogram measurement of the living body is a continuous repetition of the waveform as shown in FIG. 5(B). The straight line connecting the beginning of the P wave to the beginning of the next P wave is called the baseline (548). Waveforms that appear above the baseline are called positive waveforms, and waveforms that appear below the baseline are called negative waveforms. Since the R wave is the highest wave in the ECG waveform, the R wave is the reference point for ECG waveform analysis.

In electrocardiogram waveform analysis in biometric authentication, waveforms are analyzed with reference to these basic waveforms. The amplitude and duration of these waves may change due to a change in physical condition or the development of a disease. Also, some waves may disappear.
<Explanation of functional blocks of biometric authentication device (Fig. 1)>

FIG. 1 is a functional block diagram of the biometric authentication device of this embodiment.

The biometric authentication device (101) of this embodiment includes a biometric information measurement unit (102) that measures an electrocardiogram, a biometric information storage unit (103) that stores the measured electrocardiogram in association with the measurement time, and minutes of biological information is read, the waveform feature amount is extracted, the average value or median value is calculated, and the average value or median value of the waveform feature value in the normal, typical or standard range for the user is calculated. is stored in cooperation with the feature amount extraction unit (104, the case of obtaining the average value for the sake of simplification of explanation below) that generates a template (first template) of A biometric information variation pattern extraction unit (106, hereinafter referred to as In order to simplify the explanation, the case where the average value is obtained will be explained), template management for registering the average value of the feature values obtained by the feature value extraction unit (104) and the biometric information variation pattern extraction unit (106) as a template. a template registration determination unit (107) for determining whether or not to register the feature amount of the variation pattern of the extracted biometric information; an identity authentication unit (115) for authenticating an individual using the biometric information; , a communication unit (109) for communicating with an external device, and a display unit (110) for displaying various information to the user. In the following description, simply referring to a template includes the first template and the second template.

A biometric authentication device (101) has both an authentication function using an electrocardiogram and a function of measuring/recording the electrocardiogram as healthcare information. A function of measuring/recording an electrocardiogram as healthcare information is also realized by the biological information measurement unit (102) and the biological information storage unit (103). In addition, since the biometric authentication device (101) also serves as a healthcare application, even when biometric information is measured for the purpose of authentication, the biometric information is always recorded in the biometric information storage unit (103). In addition, when measuring biological information, the measurement time is also recorded so that it can be used for physical condition analysis in healthcare applications.
<Description of hardware configuration of biometric authentication device (Fig. 2)>

Next, the hardware configuration of the biometric authentication device (101) described in FIG. 1 will be described with reference to FIG.

A biometric authentication device (101) includes electrodes (positive and negative) (201) for measuring electrocardiogram potentials, an acquisition module (202) for acquiring an electrocardiogram signal from the potentials acquired by the electrodes, and a biometric authentication device (101). A central processing unit (CPU) (203) that controls all operations related to the functional blocks of this biometric authentication device (101), various data used in each functional block, and biometric information measurement results. Random Access Memory (RAM) (205) for temporarily storing the programs and various data read from the storage device (204) and used as cache or work memory, analog measured by electrodes Analog to Digital Converter (ADC) (206) that converts the potential signal into a digital signal, Timer (207) that is used for time measurement and timer setting when measuring biological information, Wired communication with external equipment / Wireless communication module (208), and an operating/display display (209) for user operation or for providing information to the user.

Electrode (201), Acquisition module (202), CPU (203), RAM (205), ADC (206), Timer (207), Storage device (204) are included in functional block (116) of biometric authentication/measurement unit. Correspondingly, the communication module (208) corresponds to the communication unit (109), and the operation/display display (209) corresponds to the operation unit (108) and the display unit (110).
<General flow of authentication processing (Figs. 3, 4, 5)>

Next, FIG. 4 and FIG. 5 are used to show a general flow of authentication processing using electrocardiogram technology. Authentication processing is roughly composed of a template registration flow and a verification flow. These processes are performed by the biometric authentication device (101). First, the template registration flow will be explained.

In the template registration flow (401), the biological information measurement unit (102) measures an electrocardiogram signal (402) and performs a quality check on the signal in preprocessing (403). Next, the biological information measurement unit (102) extracts an electrocardiogram waveform in electrocardiogram extraction (404). In electrocardiogram extraction (404), the QRS waveform (547) of the electrocardiogram is extracted, and the P wave (541) and T wave (545) are extracted. Next, the feature amount extraction unit (104) extracts feature amounts in the waveform in feature extraction (405). The main features extracted are intervals, amplitudes and angles in the waveform. The feature amount to be extracted will be described below with reference to the drawings.

FIG. 5(A) shows the intervals of the extracted waveforms. Intervals RP (501, interval between P-wave peak and R-wave peak), RP_L (502, interval between P-wave onset and R-wave peak), RT_R (503, R-wave peak and T-peak) are shown in the figure. RP_R (504, interval from P-wave end to R-wave peak), RT_L (505, interval from R-wave peak to T-wave beginning), PR_S (506), ST_S (507), PR_I (508), ST_I (509), RQ (510, interval from Q-wave peak to R-wave peak), RS (511, interval from R-wave peak to S-wave peak), In addition to QT (512), add RR_I, which indicates the R-wave interval.

FIG. 5B shows the types of waveform amplitudes to be extracted. There are four, RP_A (520), RQ_A (521), RS_A (522), RT_A (523) that indicate the amplitude between the peaks of the P, Q, S and T waves from the R-wave peak.

FIG. 5C shows the types of waveform angles to be extracted. The extracted angles include the angle between the P-wave, Q-wave, and R-wave (530, AN_Q), the angle between the Q-wave, R-wave, and S-wave (531, AN_R), and the R-wave- There is an angle between the S-wave and the T-wave (532, given as AN_S). In the feature extraction (405), the feature amount extraction unit (104) and the biometric information variation pattern extraction unit (106) extract the feature amount of the waveform, calculate the average value of the feature amounts, and the template management unit (105) ) registers the result as a template. The operation of the biometric information variation pattern extraction unit (106) will be described later with reference to FIG.

An outline of processing up to template registration will be described with reference to FIG. The feature amount extraction unit (104) detects (302) the R peak from the waveform (301) measured by the biological information measurement unit (102), and divides the waveform into individual waveforms (303). The feature amount extraction unit (104) aligns the time-divided waveforms (303) using the R wave or the like as a reference point (304). After that, the feature quantity extraction unit (104) and the biometric information variation pattern extraction unit (106) extract the feature quantity for each waveform, obtain the average value of the extracted feature quantity (305), and the template management unit (105) registers the result as a template (306, 406). The feature extraction (405) and biometric information variation pattern extraction unit (106) in FIG. 4 also perform the calculation of the average value of the feature amount for each waveform.

Next, an electrocardiogram verification flow will be described. In the verification flow, the biological information measurement unit (102) measures an electrocardiogram signal (407), and performs a quality check on the signal by preprocessing (408). Next, the feature quantity extraction unit (104) extracts an electrocardiogram waveform in electrocardiogram extraction (409). In electrocardiogram extraction, the QRS complex of the electrocardiogram is extracted, and the P and T waves are extracted. Next, the feature amount extraction unit (104) extracts feature amounts in the waveform in feature extraction (410). The main features extracted are intervals, amplitudes and angles in the waveform. The processing content of feature extraction (410) is the same as the content described in the template registration flow.

Generally, the number of waveforms used during template registration is greater than the number of waveforms used during verification. A template is created using many waveforms to give a more average value. On the other hand, in the verification flow, the number of waveforms used for verification is set as small as possible in order to shorten the verification time. In collation (411), the person authentication unit (115) collates the feature amount obtained by the feature extraction with the feature amount registered as the template, and outputs the authentication result. At the time of matching, if the difference from the template value is within the set threshold value, the authentication is OK. If it does not fall within the set threshold, it is regarded as authentication failure.
<Correspondence between the authentication flow of the present embodiment and the general authentication flow described above>

Here, in the biometric authentication device (101), the biometric information measurement unit (102) performs the flow from electrocardiogram signal measurement (402) to preprocessing (403) and electrocardiogram extraction (404) in FIG. Also, the flow from feature quantity extraction (405) to template registration (406) in the template registration flow explained in FIG. The flow shown in A) is executed by the feature quantity extraction unit (104) and the template registration determination unit (107), respectively.

Similarly, the portions corresponding to feature extraction (410) to matching (411) in the verification flow (412) of FIG. It is executed by the personal authentication unit (115). That is, the feature amount extraction unit (104) acquires the measured waveform data from the biological information storage unit (103), uses the waveform data as an input for the feature amount extraction (104), and extracts the feature amount in the feature extraction (104). demand. The personal authentication unit (115) compares the feature amount extracted by the feature amount extraction (104) with the feature amount (FIG. 8) of the template registered by the template management unit (105), and the difference between the feature amounts is the threshold value. If it is within the threshold, it becomes authentication OK, and if it exceeds the threshold, it becomes authentication failure.
<Data configuration managed by biometric information storage unit (Fig. 7)>

Next, the configuration of the biometric information management table (700) stored in the biometric information storage unit (103) will be described with reference to FIG. The biological information management table (700) is a table in which user's biological measurement information (e.g., electrocardiogram) is recorded, and the electrocardiogram waveform acquired by the biological information measurement unit (102) is associated with the measurement time at the time of measurement. Recorded.

FIG. 7 shows an example of the biometric information management table (700). The biological information management table (700) has an item for recording electrocardiogram measurement start time (701) and measurement end time (702) and an item for recording measurement data (703). As the measurement data, the potential is recorded, for example, every 40ms. The interval for recording potentials is not limited to 40 ms, as long as it is an interval required to record waveforms with sufficient precision for healthcare and authentication.
<Data Configuration Managed by Template Management Unit (FIGS. 8 and 9)>

Next, one or more templates (800) managed by the template management unit (105) and the template management table (900) will be described with reference to FIGS. 8 and 9. FIG. Since the biometric authentication device (101) also serves as a healthcare application, the template management unit (105) registers the template (FIG. 8) and manages the template together with the time when the electrocardiogram was measured as the original data of the template. Record in table (900).

FIG. 8 shows an example of a template (800) stored in the template management section (105). In the template (800), the average value of the feature amount is described for each feature (interval 801, amplitude 802, angle 803) in the waveform described in (A), (B), and (C) of FIG. be done. The feature variables (805) are the feature amounts described in (A) (interval feature), (B) (amplitude feature), and (C) (angle feature) of FIG. 5, respectively. The interval unit is "ms", the amplitude unit is "mV", and the angle unit is "°" (807).

Since the biometric device (101) is characterized by registering a plurality of templates, a number indicating an identifier (804) for identifying the template (800) is also recorded. Data relating to the registered template (800) is registered in a template management table (900), which will be described later.

FIG. 9 shows the configuration of the template management table (900). The template management table (900) contains a template number (901), which is the ID of the template (800), the time when the electrocardiogram waveform used to obtain the template (measurement start time 902 and measurement end time 903), It includes a usage frequency (904) that indicates the number of times it has been used.
<Data Configuration Managed by Template Registration Determining Unit (FIGS. 11, 12, and 13)>

Next, the risk evaluation table (1200) and the additional template maximum value management table (1300) managed by the template registration determination unit (107) and stored in the storage device (204) will be described with reference to FIGS. 12 and 13. do.

FIG. 12 shows an example of a risk assessment table (1200). The risk evaluation table (1200) is a table for evaluating the user's risk determined based on the electrocardiogram. The risk evaluation management table (1200) includes age (1201), height (1202), weight (1203), disease presence and type (1204), alcohol, tobacco, and caffeine intake status (1205), and drug intake (1205). 1206), exercise (1207), and sleep (1208).

Each item in the risk evaluation management table (1200) is evaluated based on information set by the user from a series of user information registration screens displayed via the display unit (110) shown in FIG. In a series of registration screens, basic user information (1101) indicating basic characteristics of the user such as age, height, weight, presence or absence of disease and type of disease, intake status of luxury goods such as alcohol and tobacco (1102), medicine (1103) and lifestyle information (1104) are input by the user through the operation unit (108).

A series of user information registration screens shown in FIG. 11 are displayed when the user starts using the device, prompting the user to input information (a series of initial setting procedures will be described later). Here, the results input by the user are shown in the user basic information (1101), consumption status of favorite items (1102), medicine intake status (1103), and lifestyle information (1104) displayed on the display screen. It is stored by the template registration determination unit (107) with the same configuration as each table described above.

Here, the evaluation method (1209) for each item in the risk evaluation management table (1200) will be described. The age item assesses age-related risk. Since it is generally known that arrhythmia increases with age, risk can be evaluated for each age. For example, because arrhythmias are known to occur more frequently in middle-aged and older people, 31 to 50 years were assigned a risk rating of +1, 51 to 70 years of age were assigned a risk rating of +2, and those aged 71 years and older were assigned a risk rating of +3 (1210).

As for height, the risk of heart disease tends to increase when the height is shorter than the average height, so when the height is 10 cm or more shorter than the average height, the risk is +1 (1211).

With regard to body weight, if the body weight is greater than the average, the burden on the heart is likely to be imposed.

The Presence and Type of Disease section allows the assessment of risk by examining diseases that affect the heart. Specifically, we examine the presence or absence of heart disease, stroke, kidney disease, diabetes, hypertension, hyperlipidemia, and anemia to assess risk. Heart disease risk was +3, stroke risk +2, kidney disease risk +2, diabetes +2, hypertension risk +2, hyperlipidemia risk +2, anemia risk +1 (1213).

In addition, since it is known that luxury goods also affect the heart, we investigated the intake of alcohol, tobacco, and caffeine to assess the risk. Specifically, a risk of +1 was assigned to those who consumed alcohol at least once a day, and a risk of +1 to a daily alcohol intake of 30cc or more (equivalent to 1 go of sake or a large bottle of beer) (1214). .

For cigarettes, the frequency of smoking was investigated, and if the frequency of smoking was 1 to 5 times a day, the risk was +1, and if it was 5 or more times a day, the risk was +2 (1215). For caffeine, if the frequency of caffeine intake was once or more a day, the risk was +1 (1216).

We also investigated the drug use status and evaluated the risk. As for medicines, those said to have an effect on the heart are investigated for each type, and the risk is evaluated (1217).

As for exercise, it is known that people who exercise less are more likely to develop ischemic heart disease, so if 20 minutes or less a day was 20 minutes or less, the risk was +1 (1218).

As for sleep, it is known that lack of sleep also affects the heart, so if the average sleep time was 5 hours or less, the risk was set to +1 (1219).

The evaluation using the above-mentioned risk evaluation table is performed separately for men and women because the average values of weight and height differ between men and women. In addition, the risk evaluation of each item is an example, and is not limited to this example.

Next, FIG. 13 shows an example of an additional template maximum value management table (1300) that records the maximum number of registered templates (800). This table stores the default value (1301) and the maximum value (1302) of the number of additional templates. The number of templates that can be added (Na) is set to 1 (1303) by default. Therefore, even a person whose risk evaluation result is 0 can newly register one template. In the example of FIG. 13, 5 is set as the maximum value (1304). The additional template maximum value management table (1300) is registered in the storage device (204) by the template registration determination unit (107).
<Mounting example of the biometric authentication device of the present embodiment (FIG. 22)>

Next, an implementation example of the biometric authentication device (101) will be described with reference to FIG. In this embodiment, the example of the biometrics authentication system in FIG. 22 will be used for the explanation of the subsequent flow and screens.
FIGS. 22A and 22B are implementation examples of a wristband-type wearable device. This wearable device has a main body (2201) and a wristband (2202). The main body has a display (2203) for user operation or displaying information to the user, and a biometric information measurement unit (102). An example of electrodes (one on the surface 2204 of the main body and one on the back 2205 in FIG. 22B), a button (2206) for turning on/off the power, and a It has an LED (2207).
<Initial Registration Flow of Biometric Authentication Device (Registration Flow A in FIG. 1, FIG. 10, FIG. 6)>

Next, the registration flow A of FIG. 1 for the biometric authentication device (101) will be described with reference to FIG. The description also uses an image of the display screen that is shown to the user using FIG.

The biometric device (101) of FIG. 22 accepts the power ON operation of the power button (2206) of the body (2201) (step 1001). As a result, the biometric authentication device (101) displays a screen for starting registration of the template (800) on the display (2203) ((A) in FIG. 6). The screen displays the time required for registration. For example, each time the biometric authentication device (101) executes this process, the processing time from the display of the screen to the completion of registration is measured by a timer (207) and stored in a memory (not shown). Then, the average value of the past processing times may be calculated and displayed. The time (rough estimated time) stored in advance in the storage device (204) before product shipment may be displayed.

Next, the biometric authentication device (101) displays a template registration screen and prompts the user to register the template (800) ((B) in FIG. 6). An image of the operation by the user is displayed on the screen together with a message prompting the registration. The biometric authentication device (101) detects that the user puts his or her hand on the electrode (2204) of the body, and starts measurement (step 1002).

During the measurement, the biometric authentication device (101) displays a screen indicating that the template is being registered, and displays the remaining measurement time ((C) in FIG. 6). The biometric authentication device (101) confirms whether the measurement has been performed for a predetermined time (30 seconds in FIG. 6A) (step 1003), and if the measurement has been performed for the predetermined time (step 1003; Yes), the measured electrocardiogram A feature amount is extracted from the waveform of , and an average value is obtained (step 1004).

Next, the biometric device (101) registers the result as an initial template (step 1005). When template registration is completed, the biometric device (101) displays a registration completion screen ((D) in FIG. 6).

Next, the biometric authentication device (101) displays an authentication screen and prompts the user to perform authentication in order to confirm whether or not the registered template can be used for authentication ((E in FIG. 6). )). On the other hand, when the biometrics authentication device (101) detects that the user touches the electrodes (2204) of the main body (2201) again, the biometrics authentication device (101) authenticates using the detection flow described above. (step 1006).

Next, the biometric authentication device (101) confirms whether the authentication was successful (step 1007). If authentication fails as a result of step 1007 (step 1007; No), the biometric authentication device (101) displays the result of authentication failure and prompts authentication again (after displaying (F) in FIG. 6, (E) is shown). When the authentication fails, the biometric authentication device (101) counts the number of failures and checks whether the number of authentication failures is N times or less (step 1011).

When the biometric authentication device (101) confirms that the number of failures is N times or less (step 1011; Yes), it performs authentication again (step 1006). When the biometric authentication device (101) confirms that the number of failures is greater than N (step 1011; No), it discards the registered template (step 1012) and terminates the process (1010). The allowable failure count N can be arbitrarily set on the biometric device side.

On the other hand, as a result of step 1007, if the authentication is successful (step 1007; Yes), the biometric authentication device (101) displays an authentication success screen ((G) in FIG. 6), and asks the user for basic information, etc. is displayed ((H) in FIG. 6), and a setting screen for basic user information, etc. is displayed ((I) in FIG. 6: part of the registration screen, basic user information (1101) in FIG. 11). ). The biometric authentication device (101) receives input of basic information and the like from the user through the operation unit (108) (step 1008).

The biometric authentication device (101) determines whether or not all items have been entered (step 1009), and when the registration of basic information, etc. by the user is completed (step 1009; Yes), the series of initial registration processing ends ( step 1010).

When the user completes the registration of basic information, etc., the biometric device (101) displays the result on the screen ((J) in FIG. 6).

When the biometric authentication device (101) determines that all the items have not been entered (step 1009; No), it returns to step 1006.
<Authentication Flow of Biometric Authentication Processing and Measurement of Biometric Information (Verification Flow in FIG. 1, FIG. 14)>

Next, the verification flow of FIG. 1 and the biometric information measurement method will be described with reference to FIG.

The user wears the biometric authentication device (101) of FIG. 22, and the biometric authentication device (101) accepts the power ON operation of the power button (2206) (step 1401).

The biometric authentication device (101) displays the screen shown in FIG. 15A to prompt authentication. The biometric authentication device (101) detects that the user has put his or her hand on the electrode (2204), and starts authentication (step 1402). The biometric authentication device (101) displays a screen notifying that authentication is in progress ((B) of FIG. 15, the remaining time required for authentication is also displayed).

The biometric authentication device (101) determines whether or not the personal authentication has succeeded (step 1403), and if the authentication fails (step 1403; No), the process ends. However, for example, authentication may be retried for the number of allowable times set in advance. The biometric authentication device (101) in FIG. 22 displays an authentication NG screen ((C) in FIG. 15) when authentication fails.

When the personal authentication is successful (step 1403; Yes), the biometric authentication device (101) starts measuring biometric information using the biometric information measuring unit (102) (step 1404). The biometric device (101) displays a measurement start screen ((E) in FIG. 15), and displays information such as measurement time during measurement ((F) in FIG. 15).

When the biometric authentication device (101) detects that the user has let go of the electrodes, the measurement process ends, the measured biometric information is stored (step 1405), and the process ends (step 1406). It is also possible to suspend the process without performing the measurement when the personal authentication is completed. In that case, the finger should be removed from the electrode immediately after the final authentication OK screen ((D) in FIG. 15) is displayed.
<Authentication and biometric information measurement when timer setting is used (Fig. 15)>

After performing the authentication flow shown in FIG. 14 and measuring biometric information, the biometric authentication device (101) periodically performs authentication and measurement using, for example, a timer (207).

The biometric authentication device (101) may request execution of authentication at regular time intervals according to the required security level. In the example of the biometric authentication device (101) in FIG. 22, when the preset time comes, a re-authentication screen is displayed to prompt the user to re-authenticate ((G) in FIG. 15).

Also, the biometric authentication device (101) may prompt the measurement of biometric information at intervals set by a timer. The time set by the timer may be set, for example, to a time period in which there is no measurement data in the biological information storage unit, or to a time when arrhythmia or disease is relatively likely to occur. In the example of the biometric authentication device (101) in FIG. 22, when the preset time in the morning comes, the measurement screen is displayed to prompt the user to take measurements ((H) in FIG. 15).
<Authentication and biometric information measurement by user instruction (Fig. 16)>

Next, an example in which the user instructs personal authentication and biometric information measurement using the biometric authentication device (101) of FIG. 22 will be described.

The biometric authentication device (101) can perform re-authentication and measure biometric information after performing personal authentication. The user authentication OK state may be maintained while the device is attached. For example, in the biometric authentication device (101) of FIG. 22, it may be determined that the device is being worn while the potential from the electrode (2205) on the back side of the body is being detected. The personal authentication OK state is discarded when the biometric authentication device (101) is removed or when the power is turned off, and the personal authentication OK state is maintained until then.

The biometric authentication device (101) can perform authentication and measure biometric information according to user instructions when performing re-authentication or measuring biometric information even when the authentication OK state continues while the device is worn. can be

FIG. 16A shows an example of the main screen when the authentication OK state continues (the user continues to wear the biometric authentication device after personal authentication). An authentication icon for personal authentication and a measurement icon for electrocardiogram measurement are displayed on the screen. In this state, when measuring biological information, the user can start measuring an electrocardiogram by touching the displayed measurement icon (1601) ((B) of FIG. 16). ). Similarly, personal authentication can be performed by touching the authentication icon.

<Variation pattern extraction processing flow of biometric information variation pattern extraction unit (registration flow B in FIG. 1, FIG. 17)>

Next, with reference to FIG. 17, the contents of the variation pattern extraction processing performed by the biometric information variation pattern extraction unit (106), which is the feature of this embodiment, will be described.

A biometric information variation pattern extraction unit (106) of a biometric authentication device (101) starts a variation pattern extraction process when biometric information for a preset time is stored in a biometric information storage unit (103). do. It has been found that the electrocardiogram may change gradually depending on the time of day during activity. Therefore, it is desirable to start the processing of the biometric information variation pattern extraction unit (106) after several hours of biometric information is accumulated in the biometric information storage unit (103). However, it does not limit the storage time of the biometric information, which serves as a guideline for starting.

The biometric information variation pattern extraction unit (106) reads the measurement start time (see FIG. 7) of the measurement data from the biometric information storage unit (103), and sets it to the variable Time (step 1701). Also, the value of the variation pattern continuation counter N is reset (set to 0) (step 1702).

Next, the biometric information variation pattern extraction unit (106) reads a waveform for a predetermined time (T) from the biometric information storage unit (103) and inputs it to the feature amount extraction unit (104) (step 1703). At this time, a predetermined time T is added to the variable Time (step 1704). Here, the number of waveforms to be read out from the biometric information storage unit (103) is set to a number that averages the feature amount to some extent. A waveform of at least several beats is required.

The feature amount extraction unit (104) and the biometric information variation pattern extraction unit (106) calculate the feature amount of the input waveform for a predetermined time period T, and obtain the average value (step 1705). The calculated average value is held until the processing of FIG. 17 is completed.

Next, the biometric information variation pattern extraction unit (106) compares the average value calculated in step 1705 with the template managed by the template management unit (105) (step 1706), It is determined whether or not matching is possible within the threshold range (step 1708).

If collation is possible in step 1708 (step 1708; Yes), the process returns to step 1702. If collation is not possible (step 1708; No), the biometric information variation pattern extraction unit (106) adds 1 to the counter N (step 1709). Next, the biometric information variation pattern extraction unit (106) determines whether N is equal to or greater than a preset predetermined value (WN) (step 1710).

The WN value is set so that it can be changed according to the characteristics of individual waveform variation patterns. For setting or changing the setting value, for example, the user basic information setting screen shown in (I) of FIG. If this value is large, a relatively long-lasting variation pattern will be extracted, and if this value is small, a short variation pattern will be extracted. Since the continuity of the variation pattern may differ from person to person, it is desirable that this value can be flexibly set.

If the biometric information variation pattern extraction unit (106) determines in step 1710 that it is equal to or greater than the predetermined value (WN) (step 1710; Yes), it adds all the waveform feature amounts extracted in step 1705, The result is divided by the value of counter N (1711). The result is used as the result of the biometric information variation pattern extraction unit (106) and as the input value of the template registration determination unit (107). The variable Time is also used as an input value for the template registration determination unit (107). The above result becomes the variation pattern of the user. A fluctuation pattern is a pattern indicating a change in the user's physical condition, and is a waveform pattern to the extent that the user cannot be correctly authenticated. In other words, it can be said that the waveform pattern is unusual for the user and is out of the typical or standard range. When such a waveform pattern appears, it can be determined that there is a possibility of arrhythmia or heart disease. By registering the waveform of , the accuracy of user authentication can be improved.

A method of extracting fluctuation patterns using templates has been described above. good too. That is, biometric information having a different amplitude spectrum obtained by frequency analysis of standard biometric information and the result of frequency analysis may be extracted as a variation pattern.
<Template Registration Processing Flow of Template Registration Determining Unit (FIGS. 18, 19, and 20)>

Next, the template registration determination process performed by the template registration determination unit (107), which is a feature of this embodiment, will be described.

A template registration determination unit (107) determines whether or not to register the variation pattern extracted by the biological information variation pattern extraction unit (106) as a template in the template management unit (105). At that time, the template registration determination unit (107) evaluates the risk of variation pattern occurrence for each user using the risk evaluation table (FIG. 12), and limits the number of templates that can be additionally registered based on the result.

A risk evaluation method executed by the template registration determination unit (107) will be described with reference to FIG.

First, in initial setting processing (step 1008 in FIG. 10) at the start of device use, when the biometric authentication device (101) receives initial settings from the user (using the screen in FIG. 11) (step 1801), the template registration determination unit ( 107) evaluates the risk for each item set by the user while referring to the risk evaluation table (1200) based on these values (step 1802) (step 1803). Then, the risk evaluation result of each item evaluated in step 1803 is added to calculate a total score (step 1804). The template registration determination unit (107) sets this total score as the upper limit (Na) of templates that can be added (step 1805).

Next, referring to FIG. 19, a template registration process in which the template registration determination section (107) registers the variation pattern extracted by the biometric information variation pattern extraction section (106) as an additional template will be described.

When the feature amount and the measurement time are input from the biological information variation pattern extraction unit (106), the template registration determination unit (107) determines whether the current total number of additional templates N is equal to or greater than the upper limit Na of addable templates. Determine (step 1901). If the template registration determination unit (107) determines that the value is not equal to or greater than Na (step 1901; No), the template registration determination unit (107) registers the input feature amount as a template in template management (step 1902), and also registers the measurement time of the template. and record (step 1905).

Since the measurement time input from the biometric information variation pattern extraction unit (106) is the measurement end time, the template registration determination unit (107) registers it in the template management table (900) as the measurement end time. The variables WN and T in the biometric information variation pattern extraction unit are referred to, and the time obtained by subtracting WN*T from the measurement end time is set as the measurement start time and registered in the template management table (900) of the template management unit.

On the other hand, when the template registration determination unit (107) determines in step 1901 that the current total number of additional templates N is equal to or greater than Na (step 1901; Yes), the template registration determination unit (107) refers to the template management table (900) and A template with the lowest frequency is selected (step 1903). Then, the template registration determining section (107) replaces the template with a new template (step 1904), and also records the measurement time of the new template (step 1905). The method of registering the measurement time in the template management table is the same as described above.

When the additional registration of the template is completed, the biometric authentication device (101) displays the result on the display unit (110) and notifies the user that the new template has been added.

FIG. 20 shows an example of a template registration notification screen for notifying the content of additional registration of a template after completion. On the notification screen, additional registration notification, registration date and time, total number of registered templates, maximum number of templates that can be registered, addition of templates allows authentication according to changes in physical condition from the beginning of use. , and information such as improved convenience is displayed.
<Hardware configuration when pulse wave is used (Fig. 21)>

Although the first embodiment has been described based on electrocardiogram technology, the same can be applied to pulse waves. In the case of a pulse wave, since the measurement principle is different, FIG. 21 shows a hardware configuration for implementing pulse wave authentication. In the case of a pulse wave, a volume pulse wave is measured and the measurement result is used for authentication. The plethysmogram is obtained by measuring the blood vessel volume using an LED (2101) and a photodiode (PD, 2102). LEDs and PDs are used instead of electrodes, and an acquisition module (202) acquires a pulse wave signal from the plethysmogram measured by the LEDs and PDs. Otherwise, the hardware configuration is the same as the hardware configuration when using an electrocardiogram.
<Appearance of biometric authentication device (Figs. 23 to 28)>

Although an implementation example of the biometric authentication device (101) is shown in FIG. 22, other implementation examples are shown in addition to the wristband type wearable biometric authentication device.

FIG. 23 shows an example of a shirt-shaped biometric authentication device (101). Cloth-type electrodes are sewn into the body of the shirt to adhere to the body. The cloth-type electrodes (2301, 2302) are arranged to be worn above and below the shirt body portion corresponding to a location near the heart. Wires are connected to the electrodes and connected to the main unit (2303). In the main unit (2303), the potential difference measured at the positive and negative electrodes is calculated, and the result is transmitted to the smart phone (2304) using the wireless device provided in the main unit (2303). .

The smart phone (2304) can perform personal authentication and biometric information measurement on the smart phone side based on potential difference information transmitted from the main body using a built-in wireless device. Functional blocks other than the biological information measurement unit (102) shown in FIG. 1 are configured on the smartphone side. Since both electrodes of the shirt-shaped biometric authentication device are always in close contact with the body, continuous personal authentication and electrocardiogram measurement are possible. Although the smart phone is used as an example in FIG. 23, various terminals having a wireless communication function can be used without being limited to this.

FIG. 24 shows an example in which the biometric authentication device (101) is implemented in a smart phone. Electrodes, which are the biological information measuring unit (102), are arranged on the front surface ((A) of FIG. 24) and the rear surface ((B) of FIG. 24) of the smartphone (2401 and 2402, respectively). The usage is shown in FIG. 24(C). The electrode (2402) on the back is placed on the palm of one hand and brought into contact, and the electrode (2401) arranged on the surface of the smartphone is brought into contact with the other hand. In the example of (C), the thumb is brought into contact with the electrode (2401), and the positional relationship of the fingers when the user browses with one hand (the thumb is closer to the user's hand than the other fingers) is considered. (2401) is placed below the browsing surface (user's hand side). In this way, if one of the electrodes is arranged on the same surface as the browsing surface operated by the user, it is possible to browse the smartphone even during electrocardiogram measurement. Although a smart phone is used as an example in FIG. 24, various terminals having a wireless communication function can be used without being limited to this. As described above, in this example, a biometric information authentication unit that authenticates a user using biometric information having periodic fluctuations, and a contact with a human body to detect the biometric information used for user authentication. In a smartphone provided with a biometric information measuring unit as a biometric information detection unit, the biometric information detection unit is provided on the browsing surface side, which is the operation surface side operated by the user, and the back surface side, which is the back surface of the operation surface. (The same applies to the configuration in FIG. 28).

FIG. 25 shows an example in which the biometric authentication device (101) is mounted in a ring type device. Two rings (2501, 2502), which are a first biometric authentication device and a second biometric authentication device that can be attached to and detached from the fingers of both hands of the user, form a set (paired). Both rings are equipped with electrodes (2503, 2504) as biological information measuring units. Measured potential information is transmitted from a wireless device provided in one ring. This makes it possible to obtain the potential difference between the left and right hands within the body of the ring (2505, 2506) on the transmitting side. For example, when the main body (2505) receives the potential information measured by the main body (2506), it calculates the potential difference between the received information and the potential information measured from the electrodes (2503) provided on the ring (2501). .

Functions other than the above are the same as those of the wristband type biometric authentication device of FIG. 22 (however, functions other than the communication function may be integrated into one ring). The advantage of the ring-type biometric authentication device is that it is possible to continuously perform personal authentication and electrocardiogram measurement while the ring is worn on the left and right hands. FIG. 25 also shows an example of wearing a ring (in this example, one ring was worn on each of the left and right fingers to measure an electrocardiogram).

26, 27, and 28 show implementation examples when the pulse wave technology is employed. Pulse wave (volume pulse wave) is measured by LED and PD. For pulse wave measurement, all you have to do is touch the LED and PD to a part of your body. However, since it is likely to be affected by noise due to external light, etc., it is desirable to measure in a place that is relatively close to the body.

FIG. 26 shows an example in which a biometric authentication device using a pulse wave is mounted on an earphone. An earphone (2601) incorporates an LED (2602) and a PD (2603). An earphone (2601) is connected to the body (2604). The pulse wave signal measured by the earphone (2601) is processed by the main body (2604). The main body (2604) has a power button (2605), a display (2606), and a status display LED (2607) as shown in FIG. In FIG. 26, the earphone (2601) and the main body (2604) are connected with a wired cable, but they may be connected wirelessly.

FIG. 27 shows an example in which a biometric authentication device using a pulse wave is mounted on an ear clip. An ear clip (2701) incorporates an LED (2702) and a PD (2703). Ear clip (2701) is connected to body (2704). A pulse wave signal measured by the ear clip (2701) is processed by the body (2704). The main body (2704) has a power button (2705), a display (2706), and a status display LED (2707) as shown in FIG. In FIG. 27, the ear clip main body is connected with a wired cable, but it may be connected wirelessly.

FIG. 28 shows an example in which a biometric authentication device using a pulse wave is mounted on a smart phone. LED (2801) and PD (2802) are mounted on the back of the smartphone. Pulse wave measurement using the LED (2801) is susceptible to external light, so it is necessary to keep the LED (2801) and PD (2802) in close contact with the body (fingers) as much as possible. For this reason, a guide (2803) for fixing the finger was installed where the LED (2801) and PD (2802) were installed on the back of the main body. That is, on the back of the main body, LEDs (2801) and PDs (2802) are placed at positions corresponding to guides (for example, finger-shaped grooves; in FIG. 28, grooves in the shape of the index finger, middle finger, and ring finger) for holding the smartphone. is provided. This guide allows the middle finger to come into close contact with the LED (2801) and PD (2802) portions. A right-handed guide and a left-handed guide are required according to the user's dominant hand. The guide may be of a detachable type, such as a cover case, instead of being provided on the back surface of the main body.

As described above, in the biometric authentication device of the present embodiment, the biometric information variation pattern extraction unit (106) greatly fluctuates the waveform from the biometric information measured for healthcare in order to prevent the authentication rate from decreasing according to the physical condition. A high authentication rate can be maintained without depending on physical condition fluctuations by extracting time fluctuation patterns and adding new feature amounts of the patterns as templates.

At that time, in order to prevent an increase in the number of templates to be registered, a stranger is mistakenly accepted as the person himself/herself, or an increase in the false acceptance rate, the template registration determination unit (107) additionally registers a template. I try to limit the number.

This embodiment shows an embodiment different from the first embodiment. In the first embodiment, the number of templates that can be added is limited by the template registration determination unit (107) based on the evaluation result of the risk evaluation management table, but in the present embodiment, another method is provided.

That is, an attribute information measuring unit is newly provided in the biometric authentication device (101) described in FIG. A correlation with the feature amount of the variation pattern extracted by the extraction unit (106) is obtained, and when it is determined that a correlation is recognized between them, the feature of the biological variation pattern extracted by the template registration determination unit (107) Quantities (mean or median) can be registered as templates. Other functions are the same as those of the first embodiment.

FIG. 29 is a functional block diagram of the biometric authentication device (2901) of this embodiment.

The biometric authentication device (2901) of this embodiment includes a biometric information measurement unit (2902), a biometric information storage unit (2903), a feature amount extraction unit (2904), a template management unit (2906), a biometric information variation pattern extraction unit ( 2907), template registration determination unit (2908), attribute information measurement unit (2909) for acquiring attribute information at the time of biometric information measurement, personal authentication unit (2913), operation unit (2910), communication unit (2911), display unit (2912). Since the functions of the above units are basically the same as those of the first embodiment, explanations of common parts are omitted, and different parts are mainly explained below.

FIG. 30 shows the hardware configuration of the biometric authentication device (2901) described in the functional blocks of FIG.

The difference from the hardware configuration of the first embodiment (FIG. 2) is that a body temperature/acceleration sensor (3001) and GPS (3002) are newly added to realize an attribute information measurement unit (2909). Note that the speaker (3003) is not essential.

A processing program and various tables (FIGS. 31, 32, and 33) described in this embodiment are stored in a storage device. Processing programs and tables are temporarily loaded into RAM (205) and executed. Data and tables (FIG. 34) generated during processing are stored in the RAM (205). Also, the added template and template management table are stored in the storage device (204). The biometric information authentication device (2901) shown in FIG. 29 can also be implemented in the forms shown in FIGS. 22 to 28, as in the case of the first embodiment.

The parts different from the first embodiment will be described below. FIG. 31 shows the types of attribute data (3101) measured by the attribute information measuring unit (2909) newly added in this embodiment.

The attribute data is data other than the electrocardiogram measured during the electrocardiogram measurement, and relates to the user to be measured. For example, the user attribute (3102) includes body temperature (3103), body water ratio (3104), and body fat percentage (3105). It also includes wake-up time (3107), meal time (3108), bedtime (3109), and exercise time (3110) related to user's habit attribute (3106).

Next, the structure of the biological information management table (3200) managed by the biological information information storage unit (2903) will be described with reference to FIG.

In the biometric information management table (3200), in addition to the biometric information measurement times (start time 3201 and end time 3202) and measurement data (3203) similar to those in Example 1, attribute data (3204) at the time of biometric information measurement is recorded. and memorize it. In the example of FIG. 32, the user's biometric information (body water ratio 3205, body fat percentage 3206) and activity event (3207) are recorded during biometric information measurement. Activity events include, for example, various events that occur when a user leads a daily life, such as getting up and eating.

Next, using FIG. 33, an example of the variation pattern extraction table (3300) is shown. The fluctuation pattern extraction table (3300) describes the type of fluctuation pattern (3301) and the characteristics of the waveform of the fluctuation pattern (3302).

For example, pattern A is characterized by an RR interval (interval between R waves) of 0.6 seconds or less (3303). Pattern B has an RR interval of 1.0 seconds or longer (3304), and pattern C has no discrimination of P waves (3305). In this way, the fluctuation pattern extraction table (3300) defines extraction conditions for fluctuation patterns. Although these three types of patterns are described in FIG. 33, the present invention is not limited to these.

The template registration determination unit (2908) refers to the waveform characteristics (3302) of the variation pattern extraction table (3300) of FIG. Determine if it corresponds to the pattern. FIG. 34 shows a table for registering results of pattern determination (hereinafter referred to as pattern determination result table). This table includes feature quantity (3401), measurement time (3407), attribute information (3402), variation pattern (3403), analysis status (3404, either unanalyzed or analyzed), correlation analysis result (3406, have an item for recording correlated attribute data types). This table is stored and held in the storage device (204) by the template registration determination unit (2908).

When the template registration determination unit (2908) performs pattern determination of the feature amount input from the biometric information variation pattern extraction unit (2907), the determination result and the measurement input to the biometric information variation pattern extraction unit (2907) The time and attribute information are recorded in the table of FIG. The analysis status (3404) value is unanalyzed as an initial value.

As described in the first embodiment, the feature amount is the average value of the interval, amplitude, and angle features in the electrocardiogram waveform. All of the 20 feature amounts (501 to 512, 806, 520 to 523, 531 to 533) described in Example 1 are recorded (feature variables in [ ] of 3401). The item of the feature amount in the table of FIG. 34 contains a numerical value for each feature amount, but since it is difficult to understand in the explanation, it is shown using a character string for the sake of convenience.

The template registration determination unit (2908) analyzes the correlation between the feature amount and the attribute data (in FIG. 34, body temperature, body water ratio, and body fat percentage) of the attribute information (3402).

Next, using FIG. 36, a configuration example of the template management table (3600) stored in the storage device (204) and managed by the template management unit (2906) is shown.

The template management table (3600) includes items of template number (3601), measurement time (measurement start time 3602, measurement end time 3603), waveform characteristics (3604), attribute information (3605), and frequency of use (3606). . Items other than waveform characteristics (3604) and attribute information (3605) are the same as in the first embodiment. As the attribute information data (attribute data) to be registered, an average value for each pattern is registered. In the item of attribute information (3605), refer to the correlation result (3406) in FIG. 34 and record the type of attribute data that has been found to be correlated with the feature amount (3607 indicates that there was a correlation with body temperature). example shown).

Next, with reference to FIG. 37, the details of the process of analyzing the correlation between the attribute data and the feature amount in the template registration determination unit (2908), which characterizes this embodiment, will be described.

The template registration determination unit (2908) refers to the variation pattern extraction table (3300), specifies the type of variation pattern of the feature amount extracted by the biometric information variation pattern extraction unit (2907), and outputs the result to the pattern determination result table ( 34) (step 3701).

Next, the template registration determination unit (2908) obtains the total number of variation pattern types from the pattern determination result table (3400) in FIG. 34, and substitutes that value for T. The template registration determination section (2908) also sets a pattern counter value N and sets the value to 0 (step 3702).

Next, the template registration determination unit (2908) selects a pattern of unanalyzed data from the pattern determination result table (3400) in FIG. 34 and adds 1 to the pattern counter N (step 3703). Then, one piece of attribute information data for which correlation analysis has not been performed is selected, and the correlation with the feature amount of the pattern is analyzed (step 3704).

For example, the template registration determination unit (2908) uses the correlation formula shown in FIG. Determine the presence or absence of correlation.

The presence/absence of correlation is determined for each component of the feature quantity. The correlation formula in FIG. 35 is a formula for seeing the correlation between two variables X (for example, feature quantity RP) and Y (for example, body temperature T of attribute data). Using the value of r (3501), which is the result of this value, it is confirmed whether or not there is a correlation between the attribute data and the component of the feature amount (step 3705). For example, when the absolute value of r is greater than 0.4 (r>0.4 is the range of values that are generally judged to be "correlated" or higher), it can be judged that there is correlation. The value of r for judging the presence or absence of correlation is set to r>0.4, but is not limited to this. The biometric information authentication device may be configured to be arbitrarily set.

In the example of FIG. 34, the template registration determination unit (2908) calculates the value of r (3501) by correlating the RP variable, which is one of the feature values, with the temperature T variable of the attribute data. For example, let the value of variable RP in FIG. 34 be variable X (X0=RP1, X1=RPN) in FIG. 35, and let the value of variable T in FIG. 34 be variable Y (Y0=T1, Y1=TN) in FIG. In this case, N=2 in FIG. 35), the correlation coefficient r is calculated, and whether or not there is a correlation is analyzed.

In step 3705, when the template registration determination unit (2908) determines that there is no correlation between the selected component of the feature amount and the attribute data (step 3705; No), another feature amount (component) and the attribute data (step 3706).

As a result of the analysis in step 3706, the template registration determination section (2908) determines whether or not there is a correlation between another feature quantity (component) and the attribute data (step 3707).

In step 3707, when the template registration determination unit (2908) determines that there is no correlation (step 3707; No), it determines whether or not there is an unanalyzed component (step 3708).

In step 3708, when the template registration determination unit (2908) determines that there is a component that has not been analyzed (step 3708; Yes), the process returns to step 3706. If there is no unanalyzed configuration (step 3708; No), it is further determined whether there is an unanalyzed attribute (step 3709). If the template registration determination unit (2908) determines that there is an attribute that has not been analyzed (step S3709; Yes), it returns to step 3704. If it determines that there is no attribute that has not been analyzed (step S3709; No), , to step 3713 .

On the other hand, if it is determined in step 3707 that there is a correlation between another feature amount (component) and the attribute data (step 3707; Yes), the template registration determination unit (2908) performs pattern determination in FIG. The type of attribute data for which correlation is recognized is stored in the correlation result (3406) of the result table (3400) (step 3710).

The template registration determination unit (2908) inputs the average value of the feature amount of the pattern to be registered as a template to the template management unit (2906), and the template management unit (2906) registers it as a template (step 3711). The template registration determination unit (2908) inputs the measurement time, attribute information, waveform characteristics (3302), and correlation result (3406) of the relevant pattern to the template management unit (2906). ), the template management section (2906) registers it in the template management table (3600) of FIG. 36 (step 3712). The template registration determination unit (2908) determines whether or not correlation analysis has been performed for all patterns (T<=N) (step 3713), and if it is determined that correlation analysis has been performed for all patterns (step 3713; Yes). , terminate the process. When the template registration determination unit (2908) determines that correlation analysis has not been performed for all patterns (step 3713; No), the process returns to step 3703. In the flow of FIG. 37, the correlation analysis was performed on all the constituent elements of the feature amount and the attribute data. good.

In this embodiment, the configuration and functions of a biometric authentication system having the biometric authentication devices described in the first and second embodiments as constituent elements will be described.

The difference between the first embodiment and the second embodiment is that the biometric information variation pattern extraction unit is implemented in a smart phone with abundant resources or a server on the cloud, and the biometric authentication device cooperates with the smart phone or server to detect the biometric information variation. This is where the pattern can be extracted. The above smartphones and servers are configured from general computers equipped with storage devices such as CPUs and HDDs as hardware.

Another difference from the first and second embodiments is that the biometric authentication device is newly provided with an attribute information acquisition unit. In the second embodiment, the attribute information measurement unit (2909) acquires the attribute data. Acquire peripheral data at the time of biometric information measurement and attribute data indicating information about the user. Other functions are the same as those of the first and second embodiments.
The differences from the first and second embodiments will be described below.
<Configuration of biometric authentication system (Figs. 38 and 40)>

FIG. 38 shows the biometric authentication system of this embodiment. The biometric authentication system of this embodiment includes a biometric authentication device (3801), a smartphone (3830) paired with the biometric information device, a server (3840) in which the biometric authentication device is registered, an external sensor (3850) that can be connected to a network, ) and biological information measurement equipment (3860). Here, pairing means pairing of devices in communication and pairing on applications (the same applies hereinafter).

The biometric authentication device (3801) includes a biometric information measurement unit (3802), a biometric information storage unit (3803), a feature amount extraction unit (3804), a template management unit (3805), and a sensor ( 3833), an attribute information acquisition unit (3807) that acquires attribute information from external sensors (3850) and biometric information measuring devices (3860) connected to the network, and a measurement environment recorded as attribute information obtained during biometric information measurement. A template registration determination unit (3808) that examines the correlation between the attribute information such as the above and the feature amount of the extracted variation pattern and registers the feature amount of the correlated variation pattern as a template (Fig. 8), the personal authentication unit (3815 ), an operation unit (3809), a communication unit (3810), and a display unit (3811). Since the functions of the respective parts are basically the same as those of the first and second embodiments, descriptions of common parts are omitted, and different parts are mainly explained below.

The smartphone (3830) extracts a biometric information variation pattern from the biometric information stored in the biometric information storage unit (3831), which stores biometric information measured by the biometric authentication device (3801), and the biometric information storage unit (3831). A biological information variation pattern extraction unit (3832) for obtaining the feature amount and a sensor (3833) are provided.

Similarly, the server (3840) also stores the biometric information stored in the biometric information storage unit (3841), which stores the biometric information measured by the biometric authentication device (3801). and a biometric information variation pattern extraction unit (3842) for extracting the variation pattern and obtaining its feature amount, and a sensor (3843).

The difference from the first and second embodiments is that the biometric information variation pattern extraction unit (3832, 3842) is implemented in a resource-rich smart phone (3830) or a cloud server (3840), and a biometric authentication device (3801) , a smart phone (3830) and a server (3840) are being operated in cooperation with each other. The biometric information variation pattern is analyzed by uploading the biometric information measured in the biometric authentication device (3801) from the biometric authentication device (3801) to the smart phone (3830) or server (3840).

In Example 1 and Example 2, the template (FIG. 8) in the biometrics authentication device (101, 2901) was utilized when extracting the variation pattern. In the third embodiment, similarly, the template (FIG. 8) may be copied to the smart phone (3830) or server (3840) and utilized to extract the variation pattern. However, if the template (FIG. 8) cannot be taken out of the biometric authentication device (3801), first obtain the amplitude spectrum of the standard electrocardiogram waveform, and extract patterns having amplitude spectra different from the amplitude spectrum as fluctuation patterns. You may

Another difference from the first and second embodiments is that an attribute information acquisition unit (3807) is provided on the biometric device (3801) side. In the second embodiment, the attribute information measurement unit (2909) acquires attribute data, but the biometric information authentication device (3801) uses an external sensor (3850) and biometric information measurement connected through the communication unit (3810). Peripheral data at the time of biometric information measurement of the biometric authentication device (3801) and attribute data indicating information on the user of the biometric authentication device (3801) are acquired from the device (3860), the smartphone (3830), and the server (3840). If the external sensor (3850) does not exist, similar information may be obtained from the sensor (3833) or sensor (3843).

The attribute data may be acquired from the attribute information measurement unit (2909) provided inside the biometric authentication device (3801) as in the second embodiment.

The biometric authentication device (3801) in FIG. 38 is paired with the smart phone (3830) on the application. In addition, the biometric authentication device (3801) is registered in advance in the server (3840) on the linked cloud through the web screen displayed on the display unit of the smart phone (3830). This registration can be performed, for example, by utilizing the Fast IDentity Online (FIDO) authentication platform.

FIG. 40 shows a communication method between the biometric authentication device (3801) and an external device such as a sensor (3850) in the biometric authentication system of this embodiment. An example of using wireless communication Bluetooth/WiFi (both registered trademarks, the same shall apply hereinafter) as communication means is shown. The biometric authentication device (3801) in the biometric authentication system of this embodiment can communicate with a smart phone (3830) using Bluetooth/WiFi. Also, the biometric authentication device (3801) can communicate with a communicable biometric information measuring device (3860) or an external sensor (3850) using Bluetooth. Also, it is possible to connect to the IoT gateway (4001) or WiFi access point (4002) to connect to the Internet (4003) and cooperate with the cloud server (3840) on the Internet (4003). In FIG. 40, an example in which Bluetooth/WiFi is used as a means of communication has been described. (3840) is not limited to these communication means.
<Attribute Data Type Table Managed by Attribute Information Acquisition Unit (Fig. 39)>

FIG. 39 shows an example of data types (3901) measured by the attribute information acquisition unit (3807). Since data can be acquired from external sensors (3850), biological information measuring devices (3860), smartphones (3830), and servers (3840), more than the attribute information measurement unit (2909) of Example 2 measures types of data can be obtained. As for the data of the user attributes (3902), data measured simultaneously by other biological information measuring devices (3860) connected to the network can be acquired. In addition, the data of the surrounding information (3903), for example, are obtained by wireless communication via an external sensor (3850), such as weather (3904), air pressure (3905), humidity (3906), illuminance (3907), noise level (3908). ) can be obtained.
<Processing Flow of Attribute Information Acquisition Unit (Fig. 43)>

Processing of the attribute information acquisition unit (3807) will be described with reference to FIG. When the biometric authentication device (3801) starts measuring biometric information, it determines whether or not there is a connectable sensor or device through the communication unit (3810) (step 4301). If the biometric authentication device (3801) determines that there is no available device (step 4301; No), it waits for a certain period of time (step 4302), determines whether there is a sensor or device that can be connected again (step 4301), If it is determined that there is a connectable device (step 4301; Yes), the connectable device is connected using Bluetooth or WiFi (step 4303). The biometric authentication device (3801) stores attribute data indicating information about the user of the biometric information measuring device (3860) in the template registration determining section (3808) (step 4304). In the first embodiment, attribute data is input by the user, but some data (for example, body weight) can be acquired from an external device via wireless communication. Next, the biometric device (3801) acquires the data detected by the sensors (3850, 3843, 3833) and stores them in the biometric information storage unit (3803) together with the biometric information measurement data (step 4305).

Returning to FIG. 43, the biometric authentication device (3801) determines whether or not to continue biometric information measurement (step 4306), and if it determines not to continue (step 4306), the process ends. On the other hand, if the biometric authentication device (3801) determines to continue (step 4306; Yes), it determines whether there is a connectable sensor or device (step 4307), and if it determines that there is no connectable sensor or device (Step 4307), the process ends. On the other hand, when the biometric authentication device (3801) determines that there is a connectable sensor or device (step 4307; Yes), the process returns to step 4305.

In addition, the types of sensors and devices that can be connected to the IoT gateway (4001) and the types of data that can be obtained from them are recorded, and if such information can be obtained from the connected biometric authentication device (3801), they , the user or the biometrics authentication device (3801) may actively go to obtain the data.
<Processing Flow of Attribute Information Acquisition Unit (Selecting Biometric Information Measuring Device Sensor) (FIG. 41)>

Processing when there are many biometric information measuring devices and sensors that can be connected to the biometric authentication apparatus of the present embodiment and information from all biometric information measuring devices and sensors cannot be acquired will be described.

FIG. 41 illustrates an example of selecting attribute data to be stored according to the user's activity response. The reason why attention is focused on the activity content is that the factors affecting the living body change greatly depending on the activity content. For example, factors affecting the living body differ during eating, commuting, and sleeping. During meals, it is highly possible that the type and amount of food ingested have a great effect on the living body. In addition, if you are commuting, there is a high possibility that the degree of traffic congestion, airborne components in the vehicle, etc. will have an effect. Also, during sleep, there is a high possibility that the temperature, noise, illuminance, etc. of the room will have an effect. Therefore, if the number of pieces of attribute data that can be acquired/stored at one time is limited, the attribute data to be acquired should be changed according to the activity of the user.

That is, the biometrics authentication system of this embodiment detects the user's activity by wirelessly communicating with the activity meter (step 4101), determines whether the activity is waking up (step 4102), If it is determined that it is waking up (step 4102; Yes), for example, biological information such as body temperature, pulse, respiration, and blood pressure, and environmental attribute data at the time of waking up, such as illuminance and outside temperature, are preferentially read from the activity meter, It is recorded in the storage device (204) in the biometric device (step 4103).

If the biometric authentication device determines that the activity is not getting up (step 4102; No), it further determines whether the activity is eating (step 4104), and if it determines that the activity is eating (step 4104). ; Yes), for example, the attribute data relating to the ingested food is preferentially read from the activity meter and recorded in the storage device (204) in the biometric authentication device (step 4105).

Further, if the biometric authentication device determines that it is not eating (step 4104; No), it determines whether or not the person is commuting/moving (step 4106). ; Yes), for example, environment-related attribute data such as the degree of congestion, outside temperature, chemical components in the air, etc. are preferentially read from the activity meter, and recorded in the storage device (204) in the biometric authentication device (step 4107).

Also, if the biometric authentication device determines that it is not commuting/moving (step 4106; No), it determines whether it is sleeping (step 4108), and if it determines that it is sleeping (step 4108; Yes), for example, biometric information such as body temperature, pulse, respiration, and blood pressure, and attribute data related to the environment such as illuminance, noise, and outside temperature are preferentially read from the activity meter, and recorded in the storage device (204) in the biometric authentication device. (step 4109). As described above, the biometric authentication device of this embodiment selects and records the type of data to be acquired from among the connectable biometric information measuring devices and sensors according to the activity of the user. FIG. 41 shows a method of selecting attribute data for getting up, eating, commuting/moving, and sleeping, but activities handled by the biometric authentication device are not limited to these. When there are many types of attribute data in the selected biometric information or environmental information, attribute data in which a large change is recognized (the amount of change determined to be a large change is defined in advance) is acquired.

Activity may be detected using an activity meter as described above, or may be detected by referring to a user's schedule recorded in a smartphone or the like. Activities with little movement, such as eating, can be detected by other biometric information (heartbeat fluctuations, mouth movements, etc.).

In FIG. 41, the attribute data is selected according to the activity of the user, but as another method, the type of attribute data to be acquired may be changed according to the time period of the day. It is generally known that living organisms have a circadian rhythm. Living organisms change in approximately 24-hour cycles, and there is a constant rhythm of change. It is known that the time of onset of myocardial infarction and arrhythmia also has a circadian rhythm. For this reason, during times when diseases and arrhythmias are likely to develop due to circadian rhythms (mainly in the morning and evening), biological information and attribute data that affect circadian rhythms (e.g., light intensity when waking up, external data, etc.) temperature, etc.) may be acquired.

Although the method of selecting the attribute data is determined based on the general tendency, the attribute data to be acquired may be selected according to the physical condition of each user or the state of affliction of the disease. For example, in the case of a user suffering from angina pectoris, since it is known that exercise causes a change in cardiac activity, attribute data relating to the amount of cardiac activity is acquired. A user with a respiratory disease acquires mainly attribute data related to atmospheric components and humidity. A person with high blood pressure mainly obtains attribute data such as the salt concentration and fat content of the ingested food. Also, if the genetic constitution is known in advance by genetic analysis or the like, the attribute data to be acquired may be selected using the result.

If the types of attribute data that have a large impact on individual users are known from the analysis results of big data on users, the results may be used for selection. Also, a user who is susceptible to seasonal environmental factors such as hay fever may change the type of attribute data to be acquired for each season. In the case of the elderly, it is important to manage fluids in order to maintain their physical condition, and since dehydration has a greater impact on the heart, priority is given to attribute data related to fluid intake. good too. Also, when the time difference or the environment changes due to travel or the like, the change may be detected by GPS or the like, and attribute data (such as illuminance) that affects jet lag may be selected.

The method for selecting different types of attribute data has been described above, but biometric information measuring devices that can acquire the same type of attribute data, and biometric information measuring devices and sensors that acquire data when there are many sensors are selected. I will explain how.

Attribute data related to a living body is acquired from a biological information measuring device or sensor capable of acquiring the most recent data. Attribute data on ingested food is acquired from biological information measuring devices and sensors that can acquire the most recent data in terms of time. What is taken into the body through the respiratory system is obtained from the biological information measuring device or sensor that is spatially closest to the body. Attribute data related to the surrounding environment is acquired from the biological information measuring device or sensor that is closest in spatial distance.

As the user moves through the space, sensors that can acquire surrounding environmental data change significantly. In this case, in particular, data from sensors that are spatially close to the user and have detected a large change are preferentially acquired. In particular, changes in the outside temperature are known to place a heavy burden on the heart, so temperature changes should be accurately acquired.

Although the method of selecting attribute data to be acquired has been described above, focusing on the content that has a physical impact, it is also necessary to consider psychological factors. For this reason, for example, the type of attribute data to be selected may be changed in response to a life event such as advancement to higher education, employment, marriage, or the like that greatly changes life. Life events are recorded by the user. Prioritize acquisition of attribute data relating to lifestyle habits, especially before and after an event that significantly changes one's lifestyle. In the case of events such as moving, where the surrounding environment (temperature and rainfall) changes significantly, attribute data related to the surrounding environment is preferentially acquired.

In addition, if you are living alone due to the death of a family member, the amount of communication you have with others is important for maintaining your mental health. Prioritize acquisition of attribute data related to Regarding psychological factors, if the types of attribute data that have a large influence on individual users are known from the analysis results of big data on users, the results may be used to select attribute data.

<Biometric information variation pattern extraction processing flow between smartphone/server biometric authentication device (Fig. 42)>

Finally, with reference to FIG. 42, the biological information variation pattern extraction processing flow between the smart phone (3830)/server (3840) and the biometric authentication device (3801) will be described.

FIG. 42 shows the processing flow of the biometric authentication device (3801), smart phone (3830), and cloud server (3840) in the biometric authentication system of this embodiment. The process starts at an arbitrary timing after personal authentication is completed in the biometric authentication device (3801). The biometric authentication device (3801) in the biometric authentication system of this embodiment starts communication by WiFi (step 4201). Next, the biometric device (3801) connects to the smart phone (3830) or WiFi access point (step 4202). When the connection is completed, the biometric authentication device (3801) transmits the biometric information in the biometric information storage unit (3803) to the smart phone (3830) or the cloud server (3840) (step 4203). The data structure in the biometric information storage section (3803) is as shown in FIG.

The server (3840) on the smart phone (3830) cloud analyzes the variation pattern (step 4204). The biometric authentication device (3801) receives the variation pattern analysis result from the smart phone (3830) or the cloud server (3840) (step 4205). The biometric authentication device (3801) determines whether or not the reception has been completed (step 4206), and when it is determined that the reception has been completed (step 4206; Yes), ends the process. On the other hand, if the biometric authentication device (3801) determines that it has not been completed (step 4206; No), it returns to step 4205 and continues to obtain the variation pattern analysis result from the smartphone (3830) or the cloud server (3840). receive.

In the above description, WiFi is used for communication between the biometric authentication device and the smartphone, but Bluetooth may also be used.

FIG. 38 shows an example in which the template registration determination unit (3808) is provided in the biometric authentication device (3801). You may make it possible. Biometric information on the smart phone (3830) may be transmitted to the server (3840) after analysis. After transmission to the server (3840), the area vacated may be overwritten with new biometric information. If the biometric information is not stored after analysis, the biometric information after analysis on the smart phone (3830) or server (3840) may be overwritten with new biometric information.

In this embodiment, the configuration and functions of a medical service to which the biometric authentication device and system of the first, second, and third embodiments are applied will be described.

The difference from Embodiments 1 to 3 is that the biometric authentication device (4401) is equipped with a predicted waveform pattern management section (4411) and a disease diagnosis section (4409), enabling detection of heart disease and arrhythmia. . In addition, the feature quantity extraction unit (4404) is newly provided with a function of extracting the amplitudes of the Q, ST, and T waves on a baseline basis, and these feature quantities are included in templates and predicted waveform patterns. Also, the predicted waveform pattern is generated from the initial registration template and the characteristics of each variation pattern.

The difference from the third embodiment is that the cooperating smart phone (4430) and the server (4440) on the cloud are further provided with disease countermeasure examination units (4433, 4443) so that countermeasures can be examined when a disease develops. In addition, the smart phone (4430) and server (4440) that cooperate with the biometric authentication device (4401) can cooperate with the server (4450) of another cooperation organization to support the user.
<Configuration of biometric authentication system (Fig. 44)>

FIG. 44 shows another embodiment of the biometric authentication system of this example. The biometric authentication system of this embodiment includes a biometric authentication device (4401), a smartphone (4430) paired with the biometric information device (4401), a server (4440) in which the biometric authentication device (4401) is registered, and a biometric authentication device. It comprises a server (4440) in which (4401) is registered, a cooperation server (4450) that cooperates with a smartphone (4430), a sensor group (4470), and a biological information measuring device group (4480).

The biometric authentication device (4401) includes a biometric information measurement unit (4402), a biometric information storage unit (4403), a feature amount extraction unit (4404), a template management unit (4406), an attribute information acquisition unit (4407), and a template registration unit. A determination unit (4408), a person authentication unit (4421), a disease diagnosis unit (4409) that diagnoses a disease by analyzing the measurement result of biological information, and a feature waveform at the time of disease onset from the user's initial registration template (Fig. 45 ) and manage them, it comprises a predicted waveform pattern management unit (4411), an operation unit (4412), a communication unit (4413), and a display unit (4414). Since the functions of the above sections are basically the same as those of Embodiments 1 to 3, explanations of common parts are omitted, and different parts are mainly explained below.

The smart phone extracts a biometric information variation pattern from the biometric information stored in the biometric information storage unit (4431) and the biometric information stored in the biometric information storage unit (4431), which stores biometric information measured by the biometric authentication device, and calculates the feature amount. It is equipped with a biological information variation pattern extraction unit (4432) to be obtained, a disease countermeasure examination unit (4433) for examining countermeasures when a disease develops, and a sensor (4434).

Similarly, the server (4440) also stores the biometric information measured by the biometric authentication device (4401), the biometric information stored in the biometric information storage unit (4441), and the biometric information from the biometric information recorded in the biometric information storage unit (4441). It includes a biometric information variation pattern extraction unit (4442) that extracts the variation pattern of and obtains its feature quantity, a disease countermeasure examination unit (4443), and a sensor (4444).

The data structure (FIG. 32) of the biometric information storage units (4403, 4431, 4441) and the structure of the template management table in the biometric information authentication device (4401) are the same as in the second embodiment (FIG. 36). The biometric authentication device (4401) is registered in advance with the server (4440) to be linked. In addition, the smart phone (4430) paired with the biometric authentication device (4401) is also registered in the cooperating server (4440). In addition, the paired smartphone (4430) is also registered in the cooperating institution server (4450).

If serious illness develops, the help of someone close to you may be needed until ambulance help arrives. For this reason, rescuer information (for example, user information such as name, address, telephone number, etc.) for identifying not only the user but also the person (4460) who can request the user's rescue cooperation is sent to the server (4440). be registered in advance.
<Rescue helper registration screen (Fig. 49)>

FIG. 49 is an example of a screen for registering rescue cooperators (4460). The screen is, for example, input from an input device such as a keyboard of the server (4440) and displayed on a display device such as a display of the server (4440). In this example, the case where the rescue collaborator (4460) is registered from the server (4440) is described. It may be registered with the server (4440) via the network. As shown in FIG. 49, the screen is provided with a registration column for rescue cooperators, and user information related to rescue cooperators such as name, age, and contact information is displayed. In this way, mainly contact information is registered from the screen so that the user can be requested to help in the event of an emergency.
<Table for Managing Fluctuation Pattern Features and Angina Pectoris Diagnosis Index (Figs. 48 and 47)>

The biometric authentication device (4401) stores a table (FIG. 48) for managing the characteristics of variation patterns in the storage device (204) in the predicted waveform pattern management unit (4411). The characteristics of this variation pattern are mainly the characteristics of disease and patterns such as arrhythmia requiring attention.

There is an index that serves as a guideline for diagnosing the type of heart disease or arrhythmia based on the characteristics of changes in the waveform of an electrocardiogram. For example, in angina pectoris, a change occurs between the S wave and the T wave (hereinafter referred to as ST). FIGS. 47A and 47B show changes in electrocardiogram waveforms at the onset of angina pectoris. The ST is normally located at the base line (4703), but when angina pectoris develops, this portion descends and becomes negative (ST horizontal 4701, ST descend 4702). The above features are described in a table (FIG. 48) stored by the predicted waveform pattern management unit (4411).

FIG. 48 shows the configuration of a table stored in the storage device (204) by the predicted waveform pattern management unit (4411). This table includes types of fluctuation patterns (4801), fluctuation pattern features and index feature amounts (4802), disease names (4805) that can be diagnosed by the fluctuation pattern feature amounts, and urgency of the disease (4803). and countermeasures (4804) are recorded. The countermeasure (4804) described here differs depending on the service content of the service organization (4450) that cooperates with the biometric device (4401).
<Description of Feature Amounts for Disease Diagnosis and Feature Amounts Used in Authentication, Configuration of Templates and Pattern-Specific Waveform Feature Amounts (FIG. 45)>

Note that the feature amount used for the index of disease and the feature amount used for authentication do not necessarily match. For example, four feature values of RP_A (520), RQ_A (521), RS_A (522), and RT_A (523) have been described as amplitude feature values used for authentication in FIG. 5B. , these indexes do not contain the feature quantity indicating the amplitude from the baseline (548) for the Q wave (542), S wave (544), and T wave (545). For this reason, in the biometrics authentication device (4401) shown in FIG. In addition to the feature quantity to be analyzed, it is added to a part of the feature quantity (4802) serving as an index in the table (FIG. 48) of the predicted waveform pattern management unit (4411). Since there is no peak for ST_A (525), the average amplitude of ST was recorded. In addition, since the amplitude is observed based on the base line, the fluctuation above the base line (548) is positive and the fluctuation below the base line (548) is negative. Since the baseline-based waveform feature amount analysis function is not included in the biometric authentication devices of Embodiments 1 to 3, the feature amount extraction unit (4404) performs baseline-based analysis of Q, ST, and T waves. It is necessary to newly provide a function to extract the amplitude of

FIG. 45 shows the configuration of the template and pattern waveform feature amount. In the biometrics authentication device (4401), in addition to the features (4500) for authentication in the first to third embodiments, an analysis index (4501) for disease/arrhythmia is newly added. Three baseline-based Q, S, and T wave amplitudes Q_A (4502), ST_A (4503), and T_A (4504) are added. However, the disease/arrhythmia analysis index (4501) is not limited to these three.

The predicted waveform pattern management unit (4411) refers to the table (Fig. 48) and uses the initial registration template (registered in step 1005, identified as ID = 000 in Fig. 45) to determine waveform fluctuations due to disease, arrhythmia, etc. Then, the feature amount of the waveform for each pattern (FIG. 45) is generated by predicting the waveform feature amount when the This processing is performed immediately after the initial registration processing shown in FIG. 10 is completed. However, when a new feature pattern is recognized, the server (4440) adds a new feature to the variation pattern management table (FIGS. 48 and 50) of the smartphone (4430), server (4440), and biometric authentication device (4401). patterns can be registered. Furthermore, in response to this, the biometric authentication device (4401) may newly generate and register a predicted waveform pattern corresponding to that pattern.

FIG. 45 is also an example of the feature amount of pattern A (4801). Since pattern A (4801) is characterized by ST_A being 0.1 or less, the ST_A value of the waveform feature amount of pattern A is -0.1 or less (4505). Other feature amounts are the same as those of the initial registration template (registered in step 1005 and identified as ID=000 in FIG. 45). Although the pattern A has the same other feature amounts, the feature amounts affected by the pattern differ, and not all patterns are limited to a specific feature amount like the pattern A.
<Table for Registering Disease Diagnosis Processing and Diagnosis Results (Fig. 46)>

The biometric authentication device (4401) measures an electrocardiogram with the biometric information measurement unit (4402), and stores the result together with the measurement time and attribute information in the biometric information storage unit (4403). Immediately after, the feature amount extraction unit (4404) reads the electrocardiogram data (3203), the measurement times (3201, 3202), and the attribute information (3204) stored in the information storage unit (4403), and after calculating the feature amount , the result, measurement time, and attribute information are input to the disease diagnosis unit (4409).

Next, the disease diagnosis unit (4409) compares the input feature amount with the pattern-specific waveform feature amount (described in the structure of FIG. 45) of the predicted waveform pattern management unit (4411). The disease diagnosis unit (4409) judges whether or not the part of the feature quantity that serves as an index meets the set conditions, and the other feature quantity falls within the set threshold value in the same way as at the time of authentication. Determine whether or not it fits. The disease diagnosis unit (4409) determines that a pattern has been detected if it matches the range of the set conditions and thresholds, and that a disease characterized by that pattern has been detected. Then, the results are recorded in a table (FIG. 46) consisting of registration numbers, disease names, and occurrence times.

FIG. 46 shows the configuration of the table. In the registration number (4601), a unique number is assigned to the result of pattern detection, and the number is registered. A disease name characterized by the detected pattern is registered in the disease name (4602) (the relationship between the pattern and the disease is determined with reference to the table shown in FIG. 48). The time of occurrence (4603, 4604) describes the time of measurement (measurement start time and measurement end time) input to the disease diagnosis unit (4409) together with the feature quantity.
<Example of processing after disease detection and information sent by server of service provider (Fig. 52)>

Next, processing after a disease is detected by the disease diagnosis unit (4409) will be described. When the biometric authentication device (4401) detects the onset of a disease, it transmits information on the diagnosis result, information on the user's situation, and information on the surrounding situation to the pre-registered server (4440).

FIG. 52 shows an example of information sent to the server. (A) of FIG. 52 is an example of information about the diagnosis result. It relates to the detected disease name (5201), variation pattern (5202), urgency level (5203) of the disease, and occurrence time (5204). The detected disease name (5201) and urgency (5203) are acquired by the disease diagnosis unit (4409) by referring to the table (FIG. 48) in the predicted waveform management unit. Also, the time of occurrence (5204) is acquired from within the disease diagnosis unit (4409).

FIG. 52B is an example of information about user status. Consciousness level (5205), age/gender (5206), vital information (blood pressure 5207, pulse rate 5208, respiratory rate 5209, body temperature 5216), situation at the time of onset (5210 during exercise, etc.), presence or absence of registration of rescue collaborators (5211) , GPS information (5215).

The level of consciousness (5205) may be determined by the content of the diagnosed disease, or may be determined by attribute data acquired by the attribute information acquisition unit (4407) in the biometric authentication device (4401). For the age and sex (5206), those registered by the user (stored in the template registration determination unit 4408) at the start of using the device are used. Vital information (blood pressure 5207, pulse rate 5208, respiratory rate 5209, body temperature 5216), condition at onset (5210), and GPS information (5215) are obtained from attribute information input to the disease diagnosis unit (4409). Information about the presence or absence of rescue cooperator registration (5211) uses the information first registered by the user (the contents of FIG. 49 are stored in the biometric information storage unit 4403).

(C) of FIG. 52 is an example of peripheral information at the time of disease onset. This also uses the attribute information input into the disease diagnosis section (4409).
<Functions of Disease Response Investigation Department and Internally Managed Tables (Fig. 50)>

Next, the functions of the disease treatment examination units (4433, 4443) provided on the smartphone (4430) or server (4440) will be described. The disease treatment investigation unit (4433, 4443) receives information on diagnosis results ((A) in FIG. 52), information on user status ((B) in FIG. 52), and peripheral information transmitted from the biometric authentication device (4401). Based on information such as (52 (C)), countermeasures are considered. Specific processing will be described later.

FIG. 50 shows the configuration of a table stored and managed in the storage device (204) by the disease response examination units (4433, 4443) on the smart phone (4430) and server (4440). This table stores a table similar to that of the biometric authentication device (4401). This makes it possible to perform a follow-up test on the sent diagnosis result (however, for the follow-up test, the biometric information needs to be sent separately from the biometric authentication device 4401 in order to share the biometric information).

In the biometric authentication device (4401) as well, countermeasures (4804) are determined in the table (FIG. 48) in the predicted waveform pattern management unit (4411), but there are countermeasures that can be dealt with by the biometric authentication device (4401) alone. It has been described. On the other hand, in the tables of the disease response study departments (4433, 4443) of the smart phone (4430) and the server (4440), countermeasures are set in consideration of the cooperating organizations (4450) and the relief cooperators (4460). In the disease response examination departments (4433, 4443) of the smartphone (4430) and the server (4440), information on the user situation at the time of onset (contents of (B) in FIG. 52) and the situation of the partner institution (4450) (degree of congestion, etc. ) to determine the final countermeasure.
<Processing Sequence When Detecting Onset of Myocardial Infarction (Fig. 51)>

FIG. 51 shows the processing sequence upon detecting the onset of myocardial infarction in the biometric authentication system of this embodiment. The components are a biometric authentication device (4401), a smart phone (4430), a server (4440, service provider), a cooperating agency server (4450, cooperating service agency), and a rescue cooperator (4460). A rescue cooperator (4460) has a mobile terminal such as a smart phone that can communicate with the server (4440).

A biometric authentication device (4401) detects the onset of myocardial infarction (5101). Upon receiving the result, the alarm (speaker, 3003) in the device is turned on (5102).

Next, the information on the diagnosis result ((A) in FIG. 52), the information on the user situation ((B) in FIG. 52), and the peripheral information at the onset of the disease ((C) in FIG. 52) are sent to the smartphone (4430) and the server. (4440, service provider) (5103, 5104). The processing in the biometric authentication device (4401) up to this point does not require any operation by the user, and is performed automatically.

The smart phone (4430) transmits the information received from the biometrics authentication device (4401) as it is to the server (4440) of the registration destination (5105). The reason for transmitting information from both the biometric authentication device (4401) and the smart phone (4430) to the server (4440) is to reduce the risk of disconnection of the communication line.

The server (4440, service provider) that has received the information and the disease countermeasure examination units (4433, 4443) of the smartphone (4430) receive information on the diagnosis result ((A) in FIG. 52) and information on the user situation ((A) in FIG. 52). (B) of (B)), receives peripheral information ((C) of FIG. 52) at the time of disease onset, outputs study information obtained by examining countermeasures, and requests the cooperation institution server (4450, cooperation service institution) to activate an ambulance. (5106, 5107).

For example, the disease countermeasure examination unit (4433, 4443) reads the disease name (for example, angina pectoris) by referring to the information on the diagnosis result, and furthermore, reads the disease name (for example, angina pectoris) as a medical treatment item handled by a pre-registered medical institution. It is determined whether the disease name is included, and if it is determined that the disease name is included, the contact information (e.g., email address) of the medical institution is notified of a request for an ambulance, Information on the diagnosis result, information on the user's situation, and peripheral information at the onset of the disease are attached to the notification. In the case of a disease with the highest degree of urgency, such as myocardial infarction, an ambulance call may be requested from both the smart phone (4430) and the server (4440) of the service provider. In this way, each of the disease countermeasure examination units sends a request to the cooperation server communicating with the external server based on the detection result, which is the diagnosis result of the user's heart disease diagnosed by the disease diagnosis unit of the biometric authentication device, to help the user. Notify the request for cooperation.

In addition, the disease countermeasure examination unit (4433, 4443) transmits a help request notification to the pre-registered user's mobile terminal (for example, e-mail address) of the rescue cooperator (5108). At the same time, the disease countermeasure examination department (4433, 4443) notifies Automated External Defibrillator (AED) information to the mobile terminal of the rescue cooperator (5109). For example, the disease countermeasure examination unit (4433, 4443) transmits information indicating the installation location of the nearest AED to the portable terminal. The location of the nearest AED may be determined by comparing, for example, the GPS information included in the information on the user situation with a map of pre-registered AED locations.

Upon receiving the notification, the portable terminal of the rescue cooperator (4460) notifies (reports) the status of response to the request to the server (4440, service provider) (5110). The response status is input from the input unit (for example, touch panel) of the mobile terminal by the relief cooperator. The response status includes, for example, the date and time when the relief was provided, what kind of relief was provided, and other information indicating the details of the actual relief provided by the relief cooperator. In the case of serious cases such as myocardial infarction, instructions are not given directly to the user because it is difficult for the person himself/herself to deal with the situation.
<Example of information sent to the user by the server of the service provider (Fig. 53)>

The disease countermeasure examination unit (4443) of the server (4440, service provider) transmits the diagnosis result including medical institution information, cause analysis result, and countermeasure information as a result of the examination to the user's smart phone (4430). FIG. 53 shows an example of medical institution information, cause analysis results, and countermeasure information sent from the server (4440, service provider) to the user.

The server (4430, service provider) notifies the caution level of the detected disease or arrhythmia (5301), and provides information (5302) on medical institutions capable of diagnosing and treating the detected disease or arrhythmia. This information is sent from the user, GPS information (5215) in the user situation ((B) of FIG. 52), etc. is received. All you have to do is provide the information of the medical institution. The cause analysis (5303) displays the results of analyzing the causes of the detected disease and arrhythmia. This analysis result is obtained by transmitting biometric information measured from the biometric authentication device (4401) to the smartphone (4430) and server (4440) and sharing the data with the server (4450) that cooperates, as in Example 3. It may be analyzed.

In the cause analysis result (5303) in FIG. 53, fatigue and stress are also analyzed as causes. The heart beats under the influence of both sympathetic and parasympathetic nerves. Therefore, the balance between the sympathetic nerve and the parasympathetic nerve can be analyzed by analyzing the measurement data of the electrocardiogram, and the degree of stress and fatigue can be analyzed according to the state of the balance.

A common method is as follows. Analysis can be performed by obtaining the distribution of the intervals between the R peaks of the electrocardiogram (hereinafter referred to as RRI, 4506) and obtaining the power spectrum density for each frequency component of the distribution graph. The integrated value of the density of the power spectrum of the low-frequency component 0.05-0.15Hz reflects the activity of both the sympathetic and parasympathetic nerves. On the other hand, the integrated value of the density of the power spectrum of 0.15-0.45Hz is an index of the function of the parasympathetic nervous system. The RRI coefficient of variation (RRI standard deviation divided by RRI average) is an index of autonomic nerve activity. By analyzing the indices as described above, the degree of stress and fatigue can be analyzed by analyzing the activity of the autonomic nerves and the activity of the sympathetic/parasympathetic nerves.

The above analysis can be carried out by utilizing the cooperating institution (4450) and the analysis result can be utilized, or by providing an analysis function inside the biometric authentication device (4401). Moreover, it is also possible to install and implement the function on the smart phone (4430) and server (4440) which cooperate. Returning to FIG. 53, as for the content of the countermeasure, the result of the cause analysis (5303) is received and the countermeasure (advice) for resolving it is described (5304). These countermeasures may be registered in advance in association with the above analysis results.

Example 5 shows an example of an application utilizing the biometric authentication device and system of Examples 1-4.
<Example of application to medical service and insurance service (Fig. 54)>

FIG. 54 shows an example of an application implemented by a user of the biometric authentication device of this embodiment who subscribes to both medical service and insurance service.

The user (5401) registers the biometric authentication device of this embodiment and the terminal (e.g., the email address of the mobile terminal such as a smart phone) that receives notifications from the medical service in the server of the medical service to which he subscribes. Next, the user (5401) uses the registered biometric authentication device to transmit biometric information measured after identity authentication to the server of the medical service facility (5402). The server of the medical service facility (5402) refers to the biometric information received from the registered biometric authentication device and analyzes signs of disease. When a symptom of a disease is detected, the result is notified to the above contacts, and the information of the medical institution is provided to encourage the patient to visit the hospital.

The user (5401) operates the pre-registered terminal or the like to execute payment processing for paying the service usage fee for this service. As for the settlement method, an application is installed in advance, and various conventionally known settlement methods can be used. The user (5401) also subscribes to an insurance service institution (5403) affiliated with the medical service institution (5402), and operates the terminal to execute settlement processing for paying insurance premiums. The server of the insurance service institution (5403) receives, with the permission of the user (5401), information indicating the health management status of the insurance subscriber from the server of the medical service institution (5402), confirm. The server of the insurance service agency (5403), upon confirming the health care status of the insured person, executes a process of cashing back part of the above paid insurance premiums, and transmits the result to the above terminal. The reason for executing such processing is that there is an advantage that the risk of contracting a disease is reduced for the insured person who uses the services of the affiliated medical institution for health management.
<Example of application to pension payment service (Fig. 55)>

FIG. 55 shows an example of an application when the biometrics authentication system of this embodiment is applied to pension payment. The Japan Pension Service (5502) supplies the biometric authentication device of this embodiment to the pensioner (5501) (at the start of pension receipt).

A pensioner (5501) measures an electrocardiogram using the biometric authentication device of this embodiment and registers a template. The biometric authentication device in which the template is registered is registered in advance on the server of the Japan Pension Service (5502) (it can also be registered using a paired smartphone).

The pensioner (5501) uses the biometric authentication device of the present embodiment to perform personal authentication in the month in which the pension is received. result can be sent).

The server of the Japan Pension Service (5502) judges that the payment of the pension can be permitted when the pensioner (5501) is authenticated, and sends the pension to the pre-registered pensioner (5501) and the financial institution (5503). Execute the transfer process. It should be noted that the above authentication can also serve as confirmation of existence.

The pensioner (5501) authenticates himself/herself at the financial institution (5503) that registers the pension transfer. For example, if cooperative registration with the biometric authentication device of the present embodiment is carried out in advance in the management server of an ATM of a bank, the biometric authentication device of the present embodiment that has been authenticated (and authentication OK continues) The pension can be received from the pension transfer account of the pension recipient (5501) simply by holding the card over the biometric authentication terminal of the ATM.

The biometrics authentication system of this embodiment can confirm the existence of the user as well as perform personal authentication. Therefore, it is possible to prevent fraudulent receipt of pensions after the death of the pensioner.
<Other application examples of this embodiment>

Other applications of the biometric authentication device of this embodiment include payment, entry/exit management, access control to information equipment and data, car door and engine keys, authentication and health management of drivers of public institutions, and solitary people. monitoring services, etc.

As described above, according to the biometric authentication device, the biometric authentication system, and the mobile terminal shown in each of the above embodiments, it is possible to automatically create/register a template according to physical condition. In that case, it can be implemented without adding new parts to the apparatus. In addition, it is possible to improve the personal authentication rate without depending on the physical condition, and reduce the risk of increasing the false acceptance rate due to registering an unnecessary template.

Incidentally, the processing performed by each device such as the biometric authentication device, smartphone, and server shown in each of the above embodiments is actually performed by an arithmetic device such as the CPU (203) and a memory such as the storage device (204). The functions of the above units are realized by reading out the programs stored in the , loading them into a main storage device (not shown) and executing them. The program is recorded in a computer-readable recording medium such as a CD-ROM (Compact Disc Read Only Memory) or a DVD (Digital Versatile Disc) as a file in an installable format or an executable format and provided. It may be configured to be provided or distributed by storing it on another computer connected to a network such as the Internet and downloading it via the network.

101 biometric authentication device 102 biometric information measurement unit 103 biometric information storage unit 104 feature amount extraction unit 105 template management unit 106 biometric information variation pattern extraction unit 107 template registration determination unit 115 personal authentication unit

Claims (5)

an electrocardiogram information measuring unit for measuring electrocardiogram information;
a template management unit that registers and manages feature information extracted from the electrocardiogram information;
an authentication unit that performs user authentication processing by comparing feature information managed by the template management unit with feature information extracted from newly measured electrocardiogram information;
Attribute for obtaining a predetermined condition when electrocardiographic information is measured, which is a predetermined time including morning and evening, or a condition related to any item of temperature, noise, or illuminance around the user. an information acquisition unit;
a biometric information variation pattern extraction unit that extracts feature information outside the user's standard range;
a template registration determination unit that determines whether the feature information extracted by the biometric information variation pattern extraction unit can be used in user authentication,
prompting the user to measure an electrocardiogram for a short period of time if the feature information is not registered in the template management unit;
registering first feature information based on the electrocardiogram information obtained by the measurement in the template management unit, performing user authentication processing using the first feature information registered in the template management unit;
Furthermore, after the first feature information is registered in the template management unit , the biometric information variation pattern extraction unit uses second feature information different from the first feature information, and Acquiring the second feature information extracted from the second electrocardiogram information obtained by electrocardiogram measurement,
when the template registration determination unit determines that the second characteristic information can be used in user authentication, the template management unit registers the second characteristic information as an additional template;
A biometrics authentication system, wherein user authentication processing is performed using the first and second characteristic information registered in the template management unit. The biometric authentication device according to claim 1,
The biometric authentication device, wherein the authentication unit holds the authentication result of the user until the biometric authentication device is removed from the user. The biometric authentication device according to claim 2,
Equipped with a display for viewing,
displaying the authentication result held by the authentication unit on the display as an authentication state;
A biometric authentication device characterized by: The biometric authentication device according to claim 1,
Equipped with a display for viewing,
displaying the registration status of the feature information registered in the template management unit;
A biometric authentication device characterized by: The biometric authentication device according to any one of claims 1 to 4,
The biometric authentication device is a wearable information terminal worn on the arm,
The electrocardiogram information measuring unit
Provided on the back surface of the wearable information terminal and a surface different from the back surface,
A biometric authentication device characterized by:

Images