Exhalation measurement device and authentication method

Abstract

This exhalation measurement device comprises: an exhalation volume measurement unit (48) that measures the amount of exhalation blown by a subject; a pattern identification unit (102) that identifies an exhalation volume pattern indicating the measured amount of exhaled breath over time; and an authentication unit (104) that authenticates the subject on the basis of a comparison between the exhalation volume pattern identified on the basis of the exhalation blown upon authentication of the subject and the exhalation volume pattern registered in advance.

Description

Breath measurement device and authentication method

[0001] The present disclosure relates to a breath measurement device that measures the breath of a subject and an authentication method.

[0002] Conventionally, in order to prevent drunk driving by a driver (subject), a technology is known that uses an alcohol detector to measure the breath alcohol concentration of the subject to confirm whether or not the subject has consumed alcohol (see, for example, Patent Document 1). In such alcohol tests, it is anticipated that fraudulent measurements may be taken by a substitute, such as by someone other than the subject who has not drunk alcohol exhaling into the alcohol detector. To prevent such fraud, for example, the technology described in Patent Document 1 authenticates the driver of a vehicle based on an image of the driver's face.

[0003] JP 2010-000941 A

[0004] The technology described in Patent Document 1 requires expensive devices such as cameras, and there is a problem in that it is costly to incorporate anti-tampering functions into equipment that is expected to be in the low-priced range, such as an alcohol concentration measurement device, which is a type of breath measurement device.

[0005] In view of the above, an object of the present disclosure is to provide a technology that effectively prevents a subject from impersonating another person during breath measurement while suppressing an increase in the cost of a breath measurement device.

[0006] In order to solve the above problem, a certain aspect of the breath measurement device of this embodiment comprises an exhalation volume measurement unit that measures the volume of exhaled breath blown in by the subject, a pattern identification unit that identifies an exhalation volume pattern that shows the measured volume of exhaled breath over time, and an authentication unit that authenticates the subject based on a comparison of the identified exhalation volume pattern based on the exhaled breath blown in during authentication of the subject with a registered exhalation volume pattern that has been registered in advance.

[0007] Another aspect of the method of this embodiment comprises measuring the amount of exhaled air blown in by a subject whose exhaled air is to be measured, identifying an exhaled air volume pattern that indicates the measured amount of exhaled air over time, and authenticating the subject based on a comparison of the identified exhaled air volume pattern based on the exhaled air blown in during authentication of the subject with a registered exhaled air volume pattern that has been registered in advance.

[0008] Any combination of the above components, and conversion of the expression of this embodiment into a method, device, system, recording medium, computer program, etc. are also valid aspects of this embodiment.

[0009] According to this embodiment, it is possible to provide a technology that effectively prevents impersonation of a subject in alcohol concentration measurement while suppressing an increase in the cost of an alcohol concentration measurement device, which is an example of a breath measurement device.

[0010] 7 is a diagram illustrating an example of the configuration of an alcohol concentration measurement system according to an embodiment. FIG. 8 is a diagram illustrating an example of the hardware configuration of an alcohol concentration measurement device according to an embodiment. FIG. 9 is a functional block diagram of a processing unit according to an embodiment. FIG. 10 is a flowchart of processing for registering exhalation volume patterns and voice patterns in a concentration measurement device. FIG. 11 is a diagram illustrating examples of exhalation volume patterns and voice patterns. FIG. 12 is a flowchart of processing for authenticating a subject and measuring an alcohol concentration in an alcohol concentration measurement device according to an embodiment. FIG. 13 is a diagram illustrating exhalation volume patterns and voice patterns registered in advance and identified exhalation volume patterns and voice patterns. FIG. 14 is a diagram illustrating the results of AD conversion of the exhalation volume patterns and voice patterns in FIG. 7. FIG. 15 is a diagram illustrating each calculation result in a calculation unit.

[0011] In the following description, an alcohol concentration measurement device will be used as an example of a breath measurement device. Fig. 1 is a diagram showing an example of the configuration of an alcohol concentration measurement system 1 according to an embodiment. The alcohol concentration measurement system 1 of Fig. 1 includes an administrator terminal 10, a management server 20, a subject terminal 30, and an alcohol concentration measurement device 40. The administrator terminal 10, the management server 20, and the subject terminal 30 can communicate with each other via a wireless network NW such as the Internet.

[0012] The manager terminal 10 is a terminal device used by an alcohol concentration measurement manager P1. The manager terminal 10 may be, for example, a personal computer, a PDA (Personal Digital Assistant), a mobile phone, a smartphone, or a tablet terminal.

[0013] The management server 20 receives the measurement results of the alcohol concentration measurement device 40 from the subject terminal 30 and stores them in a storage device (not shown) of the management server 20. When the management server 20 receives an access request from the administrator terminal 10, it provides the measurement results to the administrator terminal 10. The measurement results here include, for example, at least one of the measured alcohol concentration and the pass / fail result of the alcohol test when the alcohol concentration is compared with the test reference value. The measurement results may further include information on the date and time of the measurement and information that can identify the subject.

[0014] The subject terminal 30 is a terminal device used by the subject P2, whose breath alcohol concentration is to be measured. The subject terminal 30 may be, for example, a personal computer, PDA, mobile phone, smartphone, tablet terminal, or in-vehicle device. The subject terminal 30 communicates data with the administrator terminal 10 and the management server 20 via a wireless network NW. The subject terminal 30 communicates with the alcohol concentration measurement device 40 via wired or wireless communication (e.g., infrared communication, Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.). In the following description, the subject terminal 30 is assumed to be a smartphone.

[0015] The alcohol concentration measurement device 40 measures the alcohol concentration in the breath of the subject P2. The alcohol concentration measurement device 40 includes a mouthpiece 41 that is replaceably attached to the alcohol concentration measurement device 40 so that the subject P2 can blow his or her breath into the alcohol concentration measurement device 40. The subject can blow his or her breath into the alcohol concentration measurement device 40 while holding the mouthpiece 41 in his or her mouth.

[0016] 2 is a diagram illustrating an example of the hardware configuration of the alcohol concentration measurement device 40. The alcohol concentration measurement device 40 includes a concentration measurement unit 42, an output unit 43, an input unit 44, a communication unit 45, a power switch 46, a processing unit 47, a breath volume measurement unit 48, and a voice detection unit 49.

[0017] The concentration measurement unit 42 measures the alcohol concentration in the breath of the person being measured. When the person being measured holds the mouthpiece 41 between their mouths and blows in, the alcohol concentration in the breath is measured by the concentration measurement unit 42. The breath is then discharged from an exhaust port (not shown). The concentration measurement unit 42 can apply known principles, such as a gas sensor, for measuring the alcohol concentration in breath, so detailed explanations will be omitted.

[0018] The output unit 43 outputs various types of information. The output unit 43 includes, for example, a display device configured by a liquid crystal display, a light bulb, an LED (light emitting diode), an EL element (electroluminescence element), or a combination thereof, and an audio output device such as a speaker.

[0019] Various information based on operations by the subject is input to the input unit 44. The input unit 44 may input various information based on the touch position on a touch panel, or may input various information based on the operation of an operation button. The input unit 44 of this embodiment includes a registration start button 44A and a measurement start button 44B.

[0020] The communication unit 45 communicates with the subject terminal 30 via wired or wireless communication (for example, infrared communication, Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.) The communication unit 45 may also communicate with the administrator terminal 10, the management server 20, and other devices not shown.

[0021] The power switch 46 is configured to be able to switch the power of the alcohol concentration measurement device 40 on and off.

[0022] The processing unit 47 executes various processes in the alcohol concentration measurement device 40 .

[0023] The exhaled air volume measuring unit 48 measures the volume of exhaled air (hereinafter referred to as exhaled air volume) of the subject. The exhaled air volume measuring unit 48 is configured, for example, by a known exhaled air volume sensor or pressure sensor.

[0024] The voice detection unit 49 detects the voice emitted when the subject exhales. The voice detection unit 49 is composed of, for example, a microphone. Because voice is easily affected by ambient noise, it is desirable to use a directional microphone or adjust its installation position. For example, the microphone serving as the voice detection unit 49 is placed near the concentration measurement unit 42 and the exhaled breath volume measurement unit 48.

[0025] 3 is a functional block diagram of the processing unit 47 according to the embodiment. The processing unit 47 includes a determination unit 101, a pattern identification unit 102, a calculation unit 103, an authentication unit 104, an output control unit 105, a transmission unit 106, and a storage unit 107. These components can be realized by a combination of hardware and software resources, or by hardware resources alone. Examples of hardware resources that can be used include a CPU, ROM, RAM, GPU (Graphics Processing Unit), DSP (Digital Signal Processor), ISP (Image Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), and other LSIs. Examples of software resources that can be used include programs such as firmware.

[0026] The determining unit 101 executes various determination processes in the processing unit 47. Details of the determination processes will be described later.

[0027] The pattern identification unit 102 identifies an exhalation volume pattern that indicates the amount of exhaled air blown in by the subject over time, based on the amount of exhaled air measured by the exhalation volume measurement unit 48. The pattern identification unit 102 also identifies a voice pattern that indicates the voice uttered by the subject over time, based on the voice detected by the voice detection unit 49. In the following description, the exhalation volume pattern identified by the pattern identification unit 102 may be referred to as a specific exhalation volume pattern, and the voice pattern identified by the pattern identification unit 102 may be referred to as a specific voice pattern. In this embodiment, the exhalation volume pattern and the voice pattern need to be registered in advance in the memory unit 107 before the subject's breath is subjected to alcohol detection. In the following description, the exhalation volume pattern registered in advance in the memory unit 107 may be referred to as a registered exhalation volume pattern, and the voice pattern registered in advance in the memory unit 107 may be referred to as a registered voice pattern. The exhalation volume pattern and the voice pattern will be described later.

[0028] The calculation unit 103 calculates a certainty factor regarding the subject based on the identified breath volume pattern and voice pattern. The certainty factor in this embodiment will be described later.

[0029] The authentication unit 104 authenticates the subject based on a comparison of an exhalation volume pattern determined based on the exhaled air blown into the subject during authentication with a pre-registered exhalation volume pattern. The authentication unit 104 of this embodiment authenticates the subject based on a comparison of a voice pattern determined based on the subject's speech during authentication with a pre-registered voice pattern. The authentication unit 104 of this embodiment authenticates the subject based on the degree of certainty.

[0030] The output control unit 105 controls the output of the output unit 43. For example, the output control unit 105 controls the display of the display device in the output unit 43 and the audio output of the audio output device.

[0031] The transmitting unit 106 transmits various types of information to the subject terminal 30 via the communication unit 45. For example, the transmitting unit 106 can transmit to the subject terminal 30 the authentication result of the subject according to the authentication method of this embodiment, the alcohol concentration measurement result, and the like.

[0032] The storage unit 107 stores programs and various data for executing various processes of the processing unit 47. The storage unit 107 of this embodiment stores data on exhaled breath volume and voice. The storage unit 107 of this embodiment registers exhaled breath volume patterns and voice patterns for authentication in advance.

[0033] The process of registering breath volume patterns and voice patterns in the alcohol concentration measurement device 40 according to the embodiment will be described with reference to Fig. 4. Fig. 4 is a flowchart of the process S100 for registering breath volume patterns and voice patterns in the alcohol concentration measurement device 40 according to the embodiment.

[0034] In step S101, the judgment unit 101 judges whether or not the registration start button 44A in the input unit 44 has been operated while the power of the alcohol concentration measurement device 40 is on. If the registration start button 44A has not been operated (No in step S101), the process S100 returns to step S101. If the registration start button 44A has been operated (Yes in step S101), the process S100 proceeds to step S102.

[0035] In the flowchart of process S100, it is described that one test subject is registered for one alcohol concentration measurement device 40, but this is not limited to this. Two or more test subjects may be registered for one alcohol concentration measurement device 40. In this case, a step of selecting a test subject (not shown) may be carried out before the process of step S101.

[0036] In step S102, the output control unit 105 reads predetermined registration guidance information from the storage unit 107 and causes the output unit 43 to output the registration guidance information. By outputting the registration guidance information, a message such as "Please speak while blowing into the mouth to register a pattern" is displayed or output as sound by the output unit 43.

[0037] In step S103, the determination unit 101 determines whether or not the subject's blowing of breath is detected, for example, based on the measurement results of the breath volume measurement unit 48. If no blowing of breath is detected (No in step S103), process S100 returns to step S102. If blowing of breath is detected (Yes in step S103), process S100 proceeds to step S104, where the subject begins blowing breath and speaking. After starting blowing breath, the subject can speak at any timing and end blowing and speaking at any timing. The subject may start speaking simultaneously with the start of blowing breath, or may end speaking simultaneously with the end of blowing breath. Because there is no freedom of speech during blowing, it is inevitable that the subject will produce growling sounds such as "uh" or "hmm." Hereinafter, the act of speaking while blowing breath for pattern registration may be referred to as a pattern registration operation.

[0038] In step S104 , the storage unit 107 starts recording data on the exhaled air volume and voice based on the measurement result of the exhaled air volume measurement unit 48 and the detection result of the voice detection unit 49 .

[0039] In step S105, the determination unit 101 determines whether or not the subject has finished blowing their breath into the exhaled air, based on the measurement result of the exhaled air volume measurement unit 48. If the subject has not finished blowing their breath into the exhaled air (No in step S105), the process S100 returns to step S105. If the subject has finished blowing their breath into the exhaled air (Yes in step S105), the process S100 proceeds to step S106.

[0040] In step S105, the determination unit 101 may further determine whether the subject has finished blowing in their breath based on the detection results of the voice detection unit 49. Here, if the authentication unit 104 authenticates the subject based on a comparison of voice patterns, the subject needs to speak while blowing in their breath for pattern registration. Therefore, in step S105, if the determination unit 101 determines that the subject has not spoken based on the detection results of the voice detection unit 49, it may determine that the subject has not finished blowing in their breath (No in step S105). In other words, the determination unit 101 determines whether the subject is speaking based on the voice detected by the voice detection unit 49. Here, the determination unit 101 may use known voice analysis technology to determine whether the subject is speaking.

[0041] If the determination unit 101 determines that the subject has finished blowing in the exhaled air based on the measurement result of the exhaled air volume measurement unit 48, but determines that the subject has not spoken based on the detection result of the voice detection unit 49, the process may return to step S102, and the output control unit 105 may output registration guidance information. At this time, for example, the output unit 43 may display or output a message such as "For pattern registration, be sure to speak while blowing in the exhaled air."

[0042] In step S106, the storage unit 107 ends the recording of the breath volume and voice data.

[0043] In step S107, the pattern identification unit 102 identifies an exhalation volume pattern and a voice pattern from the exhalation volume and voice data recorded for the pattern registration operation. The exhalation volume pattern here refers to, for example, time-series data on the exhalation volume obtained from the start to end of the exhalation blowing. The voice pattern here refers to, for example, time-series data on the volume and frequency of the subject's vocalization obtained from the start to end of the exhalation blowing. The voice pattern may also include sounds other than the subject's vocalization, such as sounds generated by the subject's exhalation being input to the voice detection unit 49. However, in any case, the voice pattern should be information that can detect the timing at which the subject's vocalization changes.

[0044] FIG. 5 shows an example of an exhalation volume pattern 51 and a voice pattern 61. In the example of the exhalation volume pattern 51 and the voice pattern 61 shown in FIG. 5, the start time of the exhalation is t0, the start time of speaking is t1, the time immediately after the start of speaking is t2, the time when the speaking ends is t3, and the time when the exhalation ends is tx. For example, at the start time t1 of speaking, it is assumed that the exhalation volume increases simultaneously with the speaking. Therefore, characteristics are likely to appear in the exhalation volume pattern 51 and the voice pattern 61 from the start time t1 of speaking to the end time t3 of speaking, particularly in the period from the start time t1 of speaking to the time t2 immediately after the start of speaking. In other words, individual characteristics are likely to appear particularly in the period from the start time t1 of speaking to the time t2 immediately after the start of speaking. Therefore, by using the exhalation volume pattern 51 and the voice pattern 61 when such blowing and speaking are performed simultaneously for authentication, authentication accuracy can be improved.

[0045] In addition, if the pattern registration operation is performed multiple times as described below in step S108, the identified exhalation volume pattern and voice pattern are stored in memory unit 107 in association with an identifier indicating which pattern registration operation has been performed.

[0046] In step S108, the determination unit 101 determines whether the pattern registration operation has been performed a specified number of times (e.g., five times). For example, the determination unit 101 stores the number of times in the storage unit 107 as the first pattern registration operation performed after the pattern registration button is operated in step S101, and increments the number of pattern registration operations by one each time the pattern registration operation is repeated in steps S102 to S108, updating the number in the storage unit 107. The determination unit 101 determines that the pattern registration operation has been performed a specified number of times when the counted number of pattern registration operations reaches the specified number. Note that if a breathing volume pattern and a voice pattern that significantly deviate from the breathing volume pattern and voice pattern obtained in the first pattern registration operation are obtained in the second or subsequent pattern registration operations, the breathing volume pattern and voice pattern may be deleted, the pattern registration operation may be invalidated, and the number of times may not be counted. If the pattern registration operation has not been performed the specified number of times (No in step S108), the process S100 returns to step S102, and steps S102 to S108 are repeated until the specified number of times is reached. If the pattern registration operation has been performed the specified number of times (Yes in step S108), the process S100 proceeds to step S109.

[0047] In step S109, the pattern identification unit 102 determines the exhalation volume pattern and voice pattern for registration by integrating the exhalation volume patterns and voice patterns for the specified number of times identified for each pattern registration operation into a single exhalation volume pattern and voice pattern, respectively. For example, the pattern identification unit 102 reads the exhalation volume pattern and voice pattern for each pattern registration operation from the storage unit 107 using an identifier associated with the exhalation volume pattern and voice pattern for each pattern registration operation. The exhalation volume pattern and voice pattern may be integrated using a known integration method for time-series data.

[0048] For example, the pattern identification unit 102 registers the respiration volume pattern and voice pattern obtained in the first pattern registration operation as a base pattern. Then, the pattern identification unit 102 calculates the differences between the respiration volume patterns and voice patterns from the second time to a specified number of times and the base pattern. More specifically, the pattern identification unit 102 divides the respiration volume patterns and voice patterns to be compared into fixed time units, and calculates the differences between the base pattern and other patterns for each time unit. Then, the pattern identification unit 102 calculates the average or standard of the differences between the base pattern and other patterns for each time unit, and registers them as calibration values. The above is an example of the integration process of the respiration volume pattern and voice pattern by the pattern identification unit 102.

[0049] When registering breath volume patterns and voice patterns, degradation of authentication accuracy due to noise can be suppressed by integrating multiple breath volume patterns and voice patterns obtained by performing the pattern registration operation multiple times. When the pattern registration operation is performed multiple times, the pattern obtained in each pattern registration operation is used for accuracy adjustment for calibration purposes such as a margin when performing approximate calculations, only if it is not significantly different from the pattern obtained in the first pattern registration operation.

[0050] In step S110, the storage unit 107 registers the determined exhalation volume pattern and voice pattern. In conjunction with registering the exhalation volume pattern and voice pattern, the storage unit 107 may delete the exhalation volume and voice data, as well as the exhalation volume pattern and voice pattern, stored for each pattern registration operation in steps S104 to S106. When two or more subjects are registered for one alcohol concentration measurement device 40, the storage unit 107 registers the determined exhalation volume pattern and voice pattern in association with the selected subject.

[0051] After step S110, the process S100 in the alcohol concentration measurement device 40 ends.

[0052] The above description has been given assuming that only one subject uses the alcohol concentration measurement device 40, but this is not necessarily the case. In cases where multiple subjects use the alcohol concentration measurement device 40, the output control unit 105 may request input of information identifying the subject in step S101. The information identifying the subject is, for example, a unique subject ID. The subject ID is assumed to have been registered in advance by, for example, the administrator terminal 10. More specifically, when the registration start button 44A is operated, the output control unit 105 causes the output unit 43 to output a message for inputting information identifying the subject. If it is determined that the subject has input the subject ID using, for example, the input unit 44, processing S100 proceeds to step S102.

[0053] Furthermore, if there are multiple subjects using the alcohol concentration measurement device 40, when the registration start button 44A is operated (Yes in step S101), the judgment unit 101 may assign the subject ID instead of having the subject enter the subject ID.

[0054] If there are multiple subjects using the alcohol concentration measurement device 40, in step S110 the storage unit 107 links the determined breath volume pattern and voice pattern to the subject ID input in step S101 and registers them. Here, the transmission unit 106 may transmit information linking the subject ID with the determined breath volume pattern and voice pattern to the administrator terminal 10 and management server 20 via the subject terminal 30.

[0055] The process for authenticating a subject and measuring an alcohol concentration in the alcohol concentration measurement device 40 according to the embodiment will be described with reference to Fig. 6. Fig. 6 is a flowchart of the process S200 for authenticating a subject and measuring an alcohol concentration in the alcohol concentration measurement device 40 according to the embodiment.

[0056] In step S201, the judgment unit 101 judges whether or not the measurement start button 44B in the input unit 44 has been operated while the power of the alcohol concentration measurement device 40 is on. If the measurement start button 44B has not been operated (No in step S201), the process S200 returns to step S201. If the measurement start button 44B has been operated (Yes in step S201), the process S200 proceeds to step S202.

[0057] In the flowchart of processing S200, if there are two or more subjects to be registered for one alcohol concentration measurement device 40, a step of selecting subjects (not shown) may be carried out before processing step S201.

[0058] In step S202, the output control unit 105 reads predetermined authentication guidance information from the storage unit 107 and causes the output unit 43 to output the authentication guidance information. By outputting the authentication guidance information, for example, a message such as "Please speak while blowing in the pattern registered for authentication" is displayed or output as voice by the output unit 43.

[0059] Thereafter, steps S203 to S207 are executed. Here, the subject to be authenticated speaks while blowing in a pattern similar to the registered exhalation volume pattern and voice pattern. Hereinafter, this action of speaking while blowing in for authentication purposes may be referred to as the authentication action. Steps S203 to S207 are the same as steps S103 to S107 described above, and therefore their description will be omitted.

[0060] In step S208, the calculation unit 103 calculates a certainty factor for the subject based on the breath volume pattern and voice pattern identified in step S207. The certainty factor here refers to the probability that the subject who performed the authentication operation is the subject who performed the pattern registration operation (and is not an impersonator). For example, the calculation unit 103 compares the breath volume pattern and voice pattern identified in step S207 with previously registered breath volume patterns and voice patterns, and calculates the certainty factor based on the comparison results. When two or more subjects are registered for one alcohol concentration measurement device 40, the calculation unit 103 compares the breath volume pattern and voice pattern identified in step S207 with the breath volume pattern and voice pattern associated with the selected subject and registered, and calculates the certainty factor based on the comparison results. A specific method for calculating the certainty factor will be described below with reference to FIGS. 7 to 9.

[0061] 7 illustrates examples of breath volume patterns and voice patterns registered in advance and breath volume patterns and voice patterns identified in step S207. Fig. 7(a) shows breath volume pattern 52 registered in process S100 and breath volume pattern 53 identified in step S207. Fig. 7(b) shows voice pattern 62 registered in process S100 and voice pattern 63 identified in step S207. The calculation unit 103 performs analog-to-digital (AD) conversion on the breath volume patterns 52 and 53 and the voice patterns 62 and 63.

[0062] Figure 8 shows the results of AD conversion of the breath volume pattern and voice pattern of Figure 7. Figure 8(a) shows an AD conversion result 54 of the breath volume pattern 52 registered in process S100 and an AD conversion result 55 of the breath volume pattern 53 identified in step S207. Figure 8(b) shows an AD conversion result 64 of the voice pattern 62 registered in process S100 and an AD conversion result 65 of the voice pattern 63 identified in step S207.

[0063] For example, the calculation unit 103 calculates the degree of deviation between the AD conversion results 54, 55 of the breath volume patterns for a plurality of different breath volume-related parameters. The breath volume-related parameters include, for example, at least one of the total breath volume from the start of breath blowing to the end of breath blowing (hereinafter referred to as the total breath volume), the average breath volume per predetermined time interval (hereinafter referred to as the average breath volume), the amount of change in breath volume per unit time (hereinafter referred to as the breath volume change rate), and the duration of breath blowing from the start of breath blowing to the end of breath blowing (hereinafter referred to as the breath blowing time). For example, the calculation unit 103 calculates the difference (hereinafter referred to as the parameter difference) between each breath volume-related parameter of the AD conversion results 54, 55 of the breath volume patterns, calculates the difference between this difference and a threshold (hereinafter referred to as the threshold difference), and weights the threshold difference by the weight assigned to each breath volume-related parameter, thereby calculating the degree of deviation for each breath volume-related parameter. 8A, the calculation unit 103 of this embodiment calculates a deviation degree 71 of the rate of change in the exhaled air volume and a deviation degree 72 of the total exhaled air volume. For example, the deviation degree is calculated using the following formula (1).

[0064] [Degree of deviation] = [Threshold difference] × [Weight] / [Threshold] Equation (1) Furthermore, for example, the calculation unit 103 calculates the degree of deviation between the AD conversion results 64, 65 of the voice patterns for a plurality of different voice-related parameters. The voice-related parameters include, for example, at least one of an average volume per predetermined time interval (hereinafter referred to as average volume), a change in volume per unit time (hereinafter referred to as volume change rate), an average frequency per predetermined time interval (hereinafter referred to as average frequency), and a speech start timing and a speech end timing (hereinafter referred to as voice timing). The calculation unit 103 calculates a parameter difference between each voice-related parameter of the AD conversion results 54, 55 of the voice patterns, calculates a threshold difference between this difference and a threshold, and weights the threshold difference by the weight assigned to each voice-related parameter, thereby calculating the degree of deviation for the voice-related parameter. 8(b), the calculation unit 103 of this embodiment uses the above formula (1) to calculate the deviation degree of audio timing 73, the deviation degree of average frequency 74, and the deviation degree of average volume 75. The calculation unit 103 of this embodiment calculates the deviation degree of audio timing using the sum of the deviation of the vocalization start timing and the deviation of the vocalization end timing as a parameter difference.

[0065] Furthermore, the calculation unit 103 may calculate the degree of deviation between the AD conversion results 64, 65 of the voice patterns by comparing the timings at which the presence or absence of vocalization detected by the determination unit 101 changes between the AD conversion results 64, 65 of the voice patterns. For example, in the case of the voice pattern 61 shown in Fig. 5 , the timings at which the presence or absence of vocalization detection changes are the vocalization start time t1 and the vocalization end time t3.

[0066] The calculation unit 103 calculates a confidence level based on the degree of deviation calculated for each of the multiple breath volume-related parameters and the voice-related parameters. For example, the calculation unit 103 calculates the average value of the calculated degrees of deviation as the confidence level. FIG. 9 illustrates calculation results by the calculation unit 103. As shown in FIG. 9 , the calculation unit 103 calculates the degrees of deviation for the breath volume change rate, total breath volume, voice timing, average frequency, and average volume to be 4%, 6%, 16%, 4%, and 36%, respectively. The calculation unit 103 calculates the average of these degrees of deviation to obtain a confidence level of 13.2%.

[0067] In step S109, if the pattern identification unit 102 has registered base patterns and calibration values ​​for the respiratory volume pattern and voice pattern, the calculation unit 103 may calculate the confidence level further based on the registered base patterns and calibration values ​​for the respiratory volume pattern and voice pattern.

[0068] In step S209, the authentication unit 104 determines whether the calculated certainty is equal to or greater than a threshold value. If the certainty is equal to or greater than the threshold value (Yes in step S209), the process proceeds to step S210.

[0069] In step S210, the authentication unit 104 completes the authentication of the subject.

[0070] In step S211, the concentration measurement unit 42 measures the alcohol concentration in the breath blown in in steps S203 to S205.

[0071] In step S212, the output control unit 105 outputs the measurement results of the alcohol concentration measurement. For example, the output control unit 105 displays the measured alcohol concentration and the pass / fail result of the alcohol test when the measured alcohol concentration is compared with the test reference value on the display device of the output unit 43. Also, in step S212, the transmission unit 106 may transmit the measurement results to the administrator terminal 10 and the management server 20 via the subject terminal 30. If two or more subjects are registered for one alcohol concentration measurement device 40, the output control unit 105 may further include information that can identify the selected subject in the measurement results and transmit the results to the administrator terminal 10 and the management server 20.

[0072] Returning to step S209, if the confidence level is not equal to or greater than the threshold value (No in step S209), the process S200 proceeds to step S213. In step S213, the output control unit 105 reads predetermined re-authentication guidance information from the storage unit 107 and causes the output unit 43 to output the re-authentication guidance information. By outputting the re-authentication guidance information, for example, a message such as "Authentication failed. Please operate the measurement start button again and speak while blowing in," is displayed or audibly output by the output unit 43. Also, in step S213, the transmission unit 106 may transmit to the administrator terminal 10 and the management server 20 via the subject terminal 30 that the identity authentication of the subject has failed.

[0073] After steps S212 and S213, the process S200 ends.

[0074] If the subject ID has been acquired in step S101, such as when multiple subjects use the alcohol concentration measurement device 40, the output control unit 105 causes the output unit 43 to output a guide for inputting the subject ID when the measurement start button 44B is operated (Yes in step S201). If it is determined that the subject has input the subject ID using the input unit 44 or the like, and the exhalation volume pattern and voice pattern associated with the input subject ID have been registered in the storage unit 107, processing S200 proceeds to step S202.

[0075] If there are multiple subjects using the alcohol concentration measurement device 40, in step S212 the transmitting unit 106 may transmit information linking the subject ID and the measurement results to the administrator terminal 10 and the management server 20 via the subject terminal 30.

[0076] Modifications will be described below.

[0077] Although the subject is authenticated using both the breath volume pattern and the voice pattern, the present invention is not limited to this, and the subject may also be authenticated based on the breath volume pattern alone.

[0078] In the embodiment, an example in which the subject speaks during the period from the start to the end of the breath blowing has been described, but the present invention is not limited to this. A voice pattern may be identified by a voice uttered at a timing other than the period from the start to the end of the breath blowing.

[0079] In the embodiment, the alcohol concentration measurement device 40 communicates with the administrator terminal 10 and the management server 20 via the subject terminal 30, but this is not limited to this and the device may also communicate directly with the administrator terminal 10 and the management server 20.

[0080] In the embodiment, the breath alcohol concentration of the subject is measured using the breath blown in during the authentication operation, but this is not limited to this. After the authentication of the subject is completed through the authentication operation, the breath alcohol concentration of the subject may be measured by blowing breath again.

[0081] In the embodiment, the exhalation volume pattern and the voice pattern are identified by integrating a plurality of exhalation volume patterns and a plurality of voice patterns obtained by a plurality of pattern registration operations, but the present invention is not limited to this. The exhalation volume pattern and the voice pattern may be identified by one exhalation volume pattern and one voice pattern obtained by a single pattern registration operation.

[0082] The method of authenticating a subject using a breath volume pattern and a voice pattern is not limited to the example using the confidence factor described above. For example, a subject may be authenticated using a breath volume pattern and a voice pattern by a general pattern matching method. Furthermore, a subject may be authenticated using a breath volume pattern and a voice pattern by an inference model obtained by general machine learning.

[0083] In the embodiment, the confidence level is expressed by a numerical value expressed as a percentage such as 0 to 100%, but is not limited to this and may be expressed, for example, as a binary value.

[0084] In the embodiment, the exhalation volume pattern and the voice pattern are registered based on the breath actually blown in by the subject, but this is not limited to this, and the exhalation volume pattern and the voice pattern registered in the storage unit 107 may be preset. For example, the alcohol concentration measurement system 1 may have a function of transferring the exhalation volume pattern and the voice pattern already registered in the alcohol concentration measurement device 40 before replacement to the new alcohol concentration measurement device 40 after replacement when the number of years of use or the number of times of use exceeds an upper limit and replacement is necessary.

[0085] In the embodiment, an example is shown in which the subject holds the mouthpiece 41 in his / her mouth and blows in his / her breath, but this is not limiting. For example, the alcohol concentration measurement device 40 does not need to have the mouthpiece 41, and the subject may bring his / her mouth close to the alcohol concentration measurement device 40, blow in his / her breath, and speak.

[0086] The effects of the embodiment will be described below.

[0087] The alcohol concentration measurement device 40 of this embodiment includes an exhaled breath volume measurement unit 48 that measures the volume of breath blown in by the subject, a pattern identification unit 102 that identifies an exhaled breath volume pattern that indicates the measured volume of breath over time, and an authentication unit 104 that authenticates the subject based on a comparison of the identified exhaled breath volume pattern based on the breath blown in during authentication of the subject with a pre-registered exhaled breath volume pattern. This configuration enables authentication of the subject without the need for expensive equipment such as a camera. Therefore, it is possible to effectively prevent subject impersonation during alcohol concentration measurement while suppressing an increase in the cost of the alcohol concentration measurement device 40.

[0088] The alcohol concentration measurement device 40 of this embodiment further includes a voice detection unit 49 that detects voices uttered by the subject, a pattern identification unit 102 that identifies a voice pattern representing the detected voice, and an authentication unit 104 that authenticates the subject based on a comparison of the identified voice pattern with pre-registered voice patterns. This configuration allows the breath volume pattern and voice pattern to be identified through a series of operations. Therefore, authentication accuracy can be effectively improved while reducing the operational burden of authenticating the subject.

[0089] In the alcohol concentration measurement device 40 of this embodiment, the pattern identification unit 102 identifies a voice pattern based on the voice detected from the start of the breath-in process to the end of the breath-in process. Here, characteristics tend to appear in the breath volume pattern and the voice pattern during the period from the start of the speech to the end of the speech, particularly from the start of the speech to the time immediately after the speech starts. Therefore, with this configuration, the breath volume pattern and the voice pattern 61 obtained when the breath-in process and the speech are being performed simultaneously can be used for authentication, thereby improving authentication accuracy.

[0090] In the alcohol concentration measurement device 40 of the embodiment, the concentration measurement unit 42 measures the alcohol concentration contained in the breath blown in during authentication. With this configuration, authentication of the subject and alcohol detection can be performed simultaneously, thereby improving convenience.

[0091] The alcohol concentration measurement method of the embodiment includes measuring the amount of breath exhaled by a subject, whose breath alcohol concentration is to be measured, identifying a breath volume pattern that indicates the measured breath volume over time, and authenticating the subject based on a comparison of the identified breath volume pattern based on the breath exhaled during authentication with a pre-registered breath volume pattern. This configuration effectively prevents subjects from impersonating others during alcohol concentration measurement while suppressing an increase in the cost of the alcohol concentration measurement device 40.

[0092] Up to this point, the alcohol concentration measurement device 40 and the alcohol concentration measurement system 1 have been used as examples for the explanation, but this is not a limitation. For example, the breath measurement device may be a device for analyzing components in breath and utilizing them for diagnosis, a heated tobacco device, a device for measuring lung capacity, etc. Even with such a device, it is possible to effectively prevent subjects from impersonating others while suppressing increases in costs.

[0093] The above-described embodiments are merely examples, and it will be understood by those skilled in the art that various modifications are possible in the combination of each component and each treatment process, and that such modifications are also within the scope of the present disclosure.

[0094] The present disclosure relates to a breath measurement device that measures the breath of a subject and an authentication method.

[0095] 1 Alcohol concentration measurement system, 10 Administrator terminal, 20 Management server, 30 Subject terminal, 40 Alcohol concentration measurement device, 41 Mouthpiece, 42 Concentration measurement unit, 43 Output unit, 44 Input unit, 45 Communication unit, 46 Power switch, 47 Processing unit, 48 Breath volume measurement unit, 49 Voice detection unit, 101 Judgment unit, 102 Pattern identification unit, 103 Calculation unit, 104 Authentication unit, 105 Output control unit 106 Transmission unit, 107 Memory unit.

Claims

1. A breath measurement device comprising: an exhaled breath volume measurement unit that measures the volume of exhaled breath blown in by the subject; a pattern identification unit that identifies an exhaled breath volume pattern that indicates the measured volume of exhaled breath over time; and an authentication unit that authenticates the subject based on a comparison between the identified exhaled breath volume pattern based on the exhaled breath blown in during authentication of the subject and a registered exhaled breath volume pattern that has been registered in advance.

2. The breath measurement device of claim 1, further comprising a voice detection unit that detects voice uttered by the subject, wherein the pattern identification unit identifies a voice pattern that indicates the detected voice over time, and wherein the authentication unit authenticates the subject based on a comparison of the specific voice pattern identified based on the subject's utterances with a pre-registered voice pattern during authentication.

3. The breath measuring device according to claim 2, wherein the pattern identifying unit identifies the voice pattern based on the voice detected from the time when the blowing of the breath into the breath begins to the time when the blowing of the breath into the breath ends.

4. The breath measuring device of claim 2, wherein the judgment unit judges whether or not the subject's speech has been detected based on the sound detected by the sound detection unit, and the authentication unit compares the timing at which the presence or absence of the speech detection changes between the specific sound pattern and the registered sound pattern, thereby authenticating the subject further based on the degree of deviation between the specific sound pattern and the registered sound pattern.

5. A breath measurement device according to any one of claims 1 to 4, further comprising a concentration measurement unit that measures the concentration of alcohol contained in the breath blown in during the authentication.

6. An authentication method comprising: measuring the amount of exhaled air blown in by a subject whose exhaled air is to be measured; identifying an exhaled air volume pattern that indicates the measured amount of exhaled air over time; and authenticating the subject based on a comparison of the identified exhaled air volume pattern based on the exhaled air blown in during authentication of the subject with a registered exhaled air volume pattern that has been registered in advance.

Images