Alcohol consumption inference device and alcohol consumption inference method

Abstract

An alcohol consumption inference device (100) comprises a current information acquisition unit (110) that acquires, in a time series, current information which contains biological data regarding a subject being observed, a reference output unit (130) that uses the current information accumulated in a time series manner and current information at a current time to output a reference value which indicates a feature amount for a presumably sober state during a time period determined on the basis of the current information at the current time, a current state inference unit (170) that uses the reference value for the time period determined on the basis of the current information at the current time and the current information at the current time to output a feature amount for an alcohol consumption state at the current time, and an alcohol consumption state determination unit (180) that outputs, on the basis of the feature amount for the alcohol consumption state output by the current state inference unit, the result of determination regarding the alcohol consumption state of the subject being observed.

Description

Drinking estimation device and drinking estimation method

[0001] The disclosed technology relates to a drinking estimation technology.

[0002] Conventionally, there are drinking state estimation techniques that estimate whether an observed subject, such as a vehicle driver, is drinking alcohol or not. Among the drinking state estimation techniques, there is a technique that estimates whether the observed subject is drinking alcohol or not using biological data such as the pulse rate or heart rate of the observed subject. Incidentally, Patent Document 1 describes a "drunkenness state evaluation method." The "method for assessing the state of intoxication" in Patent Document 1 uses a "Holter-type electrocardiograph" to "measure the electrocardiogram of a subject who has drunk alcohol, and perform spectral analysis of the electrocardiogram data. The changes in heart rate (HR value), high-frequency spectrum value (HF value), and LF / HF ratio (LF value: low-frequency spectrum value), which are indicators of sympathetic nerve activity, obtained from the analysis results, are compared and analyzed before and after drinking to assess the state of intoxication" (Patent Document 1's solution). By doing so, "it can be understood that the changes in heart rate, HF value, and LF / HF ratio after drinking alcohol objectively indicate the state of intoxication from various aspects, including not only the level of intoxication, but also elation, excitement, comfort (e.g., good or bad mood), and relaxation, which vary depending on the subject's constitution and the type of alcohol" (paragraph

[0032] of Patent Document 1).

[0003] Japanese Patent Application Laid-Open No. 2004-242720

[0004] However, because biological data such as pulse rate or heart rate fluctuate due to various factors, conventional techniques have a problem in that there is a high possibility of erroneously estimating whether a person is in an intoxicated state or not. The "method for assessing intoxication state" in Patent Document 1 is a method for assessing the state of intoxication in an intoxicated state, and if the heart rate fluctuates due to various factors, there is a possibility that the state of intoxication will be assessed as being in an intoxicated state, and therefore the above problem cannot be solved.

[0005] The present disclosure is intended to solve the above-mentioned problems, and aims to improve the accuracy of alcohol consumption estimation compared to the prior art.

[0006] The drinking estimation device of the present disclosure includes a current information acquisition unit that acquires current information including biometric data of the observed subject in chronological order; a reference output unit that uses the current information accumulated in chronological order and the current information at the current time to output a reference value indicating a feature amount when the observed subject would be in a non-drinking state in a time period determined based on the current information at the current time; a current state estimation unit that uses the reference value for the time period determined based on the current information at the current time and the current information at the current time to output a feature amount of the drinking state at the current time; and a drinking state determination unit that outputs a determination result of the drinking state of the observed subject based on the drinking state feature amount output by the current state estimation unit.

[0007] The present disclosure provides an advantage in that the accuracy of alcohol consumption estimation can be improved compared to conventional techniques.

[0008] FIG. 1 is a diagram illustrating a basic configuration example of a drinking alcohol estimation device 100 according to a first embodiment of the present disclosure. FIG. 2 is a diagram illustrating a configuration example of a moving-object-mounted system 1 (1A) including a drinking alcohol estimation device 100 (100A) according to a first embodiment of the present disclosure. FIG. 3 is a flowchart illustrating an example of the processing performed by the drinking alcohol estimation devices 100, 100A according to the first embodiment of the present disclosure. FIG. 4 is a flowchart illustrating an example of detailed processing of a reference output processing performed by the drinking alcohol estimation device 100A according to the first embodiment of the present disclosure. FIG. 5 is a flowchart illustrating a first example of detailed processing of a drinking alcohol determination processing performed by the drinking alcohol estimation device 100A according to the first embodiment of the present disclosure. FIG. 6 is a diagram illustrating a second configuration example of a drinking alcohol determination unit in the drinking alcohol estimation device 100A according to the first embodiment of the present disclosure. FIG. 7 is a flowchart illustrating a second example of detailed processing of a drinking alcohol determination processing performed by the drinking alcohol estimation device 100A according to the first embodiment of the present disclosure. FIG. 8 is a diagram illustrating a configuration example of a moving-object-mounted system 1 (1B) including a drinking alcohol estimation device 100 (100B) according to a second embodiment of the present disclosure. FIG. 9 is a first explanatory diagram illustrating the reference values ​​used in the drinking alcohol estimation device 100 (100B) according to the second embodiment of the present disclosure. FIG. 10A in FIG. 10 is a second explanatory diagram illustrating the reference values ​​used in the drinking alcohol estimation device 100 (100B) according to the second embodiment of the present disclosure, and FIG. 10B in FIG. 10 is a third explanatory diagram illustrating the reference values ​​used in the drinking alcohol estimation device 100 (100B) according to the second embodiment of the present disclosure. FIG. 11 is a diagram illustrating a configuration example of a mobile-mounted system 1 (1C) including a drinking alcohol estimation device 100 (100C) according to the third embodiment of the present disclosure. FIG. 12A in FIG. 12 is a first explanatory diagram illustrating the reference values ​​used in the drinking alcohol estimation device 100 (100C) according to the third embodiment of the present disclosure, and FIG. 12B in FIG. 12 is a second explanatory diagram illustrating the reference values ​​used in the drinking alcohol estimation device 100 (100C) according to the third embodiment of the present disclosure. FIG. 13 is a diagram illustrating a configuration example of a moving body mounted system 1 (1D) including a drinking estimation device 100 (100D) according to the fourth embodiment of the present disclosure.FIG. 14A in FIG. 14 is a first explanatory diagram illustrating a reference value used in the drinking estimation device 100 (100D) according to the fourth embodiment of the present disclosure, and FIG. 14B in FIG. 14 is a second explanatory diagram illustrating a reference value used in the drinking estimation device 100 (100D) according to the fourth embodiment of the present disclosure. FIG. 15 is a flowchart illustrating an example of processing by the drinking estimation device 100D according to the fourth embodiment of the present disclosure. FIG. 16 is a diagram illustrating a configuration example of a moving-body mounted system 1 (1E) including a drinking estimation device 100 (100E) according to the fifth embodiment of the present disclosure. FIG. 17 is a diagram illustrating a configuration example of a moving-body mounted system 1 (1F) including a drinking estimation device 100 (100F) according to the sixth embodiment of the present disclosure. FIG. 18 is a flowchart illustrating an example of processing by the drinking estimation device 100D according to the sixth embodiment of the present disclosure. FIG. 19 is a diagram illustrating a first example of a hardware configuration for realizing functions according to the configuration of the present disclosure. FIG. 20 is a diagram illustrating a second example of a hardware configuration for realizing functions according to the configuration of the present disclosure.

[0009] In order to explain the present disclosure in more detail, embodiments of the present disclosure will be described below with reference to the accompanying drawings.

[0010] First Embodiment In the first embodiment, a description will be given of a form in which a reference value for each time period is used to determine whether or not a person is in a drunk state.

[0011] An example configuration of a drinking alcohol estimation device according to a first embodiment of the present disclosure will be described. FIG. 1 is a diagram illustrating an example basic configuration of a drinking alcohol estimation device 100 according to a first embodiment of the present disclosure. The drinking alcohol estimation device 100 estimates the drinking state based on feature values ​​that indicate a likely non-drinking state during a time period determined using the biometric data of the observed subject. The drinking alcohol estimation device 100 shown in FIG. 1 includes a current information acquisition unit 110, a current information accumulation unit 120, a reference output unit 130, a current state estimation unit 170, and a drinking state determination unit 180.

[0012] The current information acquisition unit 110 acquires current information including biometric data of the subject in chronological order. The current information acquisition unit 110 acquires the current information from, for example, a current information output device that is an external device to the drinking estimation device 100. The current information includes the biometric data of the subject. The current information also includes time information indicating the time when the biometric data, etc., was observed. The biometric data is, for example, pulse rate or heart rate. Alternatively, the biometric data is, for example, alertness level.

[0013] The current information accumulation unit 120 accumulates current information, including biometric data of the observed subject, in chronological order. The current information accumulation unit 120 may be configured to accumulate current information acquired by the current information acquisition unit 110 in chronological order (see the solid arrow in FIG. 1 ), or to directly receive and accumulate current information from an external source (see the dashed arrow in FIG. 1 ). The current information accumulation unit 120 outputs the current information accumulated in chronological order in synchronization with the current information acquisition unit 110. Alternatively, the current information accumulation unit 120 outputs the current information accumulated in chronological order in response to a request. The current information accumulation unit 120 outputs the current information accumulated in a time period including the current time. Note that the current information accumulation unit 120 may be configured to discard past current information accumulated in chronological order after a predetermined period of time has elapsed since it was stored. Although the current information accumulation unit 120 is illustrated as an internal component of the drinking estimation device 100 in the description, it may also be configured external to the drinking estimation device 100.

[0014] The reference output unit 130 outputs a reference value that is a criterion for determining whether the observed subject is in a drinking state. The reference output unit 130 uses the current information accumulated in chronological order and the current information at the current time to output a reference value that indicates a feature amount when the observed subject is likely to be in a non-drinking state during a time period determined based on the current information at the current time. The reference output unit 130 may be configured to determine a time period and then directly calculate the reference value for that time period, or may be configured to store reference values ​​for all time periods in advance, correct them according to the determined time period, and output the corrected reference value.

[0015] The current state estimation unit 170 estimates a feature quantity indicating whether the observed subject is drinking alcohol at the current time. The current state estimation unit 170 outputs a feature quantity of the drinking state at the current time using a reference value for a time period determined based on current information at the current time and current information at the current time. The current state estimation unit 170 calculates, as the feature quantity, a value corresponding to the difference between the reference value for a time period including the current time and the pulse rate, which is biological data included in the current information at the current time, for example.

[0016] The drinking state determination unit 180 determines whether the observed subject is drinking alcohol based on the feature amounts. The drinking state determination unit 180 outputs a determination result of the drinking state of the observed subject based on the feature amounts of the drinking state output by the current state estimation unit 170. The drinking state determination unit 180, for example, analyzes changes in the feature amounts and determines whether the observed subject is drinking alcohol based on the degree of change that indicates a trend of change. Alternatively, the drinking state determination unit 180, for example, estimates a drinking score that indicates the degree of drinking from the feature amounts and determines whether the observed subject is drinking alcohol based on the drinking score.

[0017] In addition to the above components, the drinking estimation device 100 also includes a control unit (not shown), a memory unit (not shown), and a communication unit (not shown). The control unit (not shown) controls the drinking estimation device 100 as a whole and each of its components. For example, the control unit (not shown) activates the drinking estimation device 100 in response to an external command. The control unit (not shown) also controls the state of the drinking estimation device 100 (operating state = startup, shutdown, sleep, etc.). The memory unit (not shown) stores various data used by the drinking estimation device 100. For example, the memory unit (not shown) stores output (output data) from each component of the drinking estimation device 100 and outputs data requested by each component to the requesting component. The communication unit (not shown) communicates with external devices. For example, communication is performed between the drinking estimation device 100 (100A) and peripheral devices (e.g., output devices, external processing devices). For example, if the drinking alcohol estimation device 100 and the peripheral device are not connected by wire, a communication unit (not shown) has a function of communicating between the drinking alcohol estimation device 100 and the peripheral device. The communication unit (not shown) also has a function of communicating with a server device serving as an external processing device. The control unit (not shown), storage unit (not shown), and communication unit (not shown) are the same in the embodiments described below.

[0018] Next, a configuration example will be described in which the drinking alcohol estimation device according to the first embodiment of the present disclosure is applied to a mobile-mounted system. FIG. 2 is a diagram illustrating a configuration example of a mobile-mounted system 1 (1A) including a drinking alcohol estimation device 100 (100A) according to the first embodiment of the present disclosure. The mobile-mounted system 1 (1A) is a system mounted on a mobile body and can estimate whether an observed subject, such as an occupant of the mobile body, is in a drinking state. The mobile-mounted system 1 (1A) shown in FIG. 2 includes the drinking alcohol estimation device 100 (100A), a current information output device 300, and an external processing device (output device) 500. When included in the mobile-mounted system 1 (1A) shown in FIG. 2, the observed subject for drinking alcohol estimation by the drinking alcohol estimation device 100 (100A) is, for example, an occupant of the mobile body. When the mobile body is a vehicle, the observed subject is, for example, the driver of the vehicle.

[0019] The current information output device 300 outputs current information including biometric data of the observed subject in a time series. The current information output device 300 shown in FIG. 2 includes a biometric data output device 310. The biometric data output device 310 outputs the biometric data of the observed subject. The biometric data output device 310 outputs the biometric data of the observed subject at the current time in a time series over time. The biometric data output device 310 outputs biometric data such as pulse rate in a time series. For example, if the moving object is a vehicle, the biometric data output device 310 includes, for example, a driver monitoring system (DMS), a seat sensor, or a millimeter-wave radar. That is, if the moving object is a vehicle, the observed subject is a driver, and the biometric data is pulse rate, the biometric data may be the pulse rate of the observed subject measured using changes in brightness values ​​of an image of the driver captured by a DMS or the like, the pulse rate of the driver measured by a seat sensor, or the pulse rate of the driver measured by a millimeter-wave radar. When configured in this manner, the biometric data output device 310 can acquire and output biometric data of the observed subject when the observed subject is on board a moving body or driving a vehicle. The biometric data output device 310 may be configured to output biometric data observed by a portable device such as a smart watch. When configured in this manner, the biometric data output device 310 can acquire and output biometric data in everyday life other than when the observed subject is on board a moving body or driving a vehicle.

[0020] The drinking alcohol estimation device 100 (100A) estimates the drinking state based on feature values ​​that indicate a likely non-drinking state during a time period determined using the biological data of the observed subject. The drinking alcohol estimation device 100 (100A) is communicably connected to a current information output device 300 and an external processing device 500. The drinking alcohol estimation device 100 (100A) shown in FIG. 2 includes a current information acquisition unit 110A, a current information storage unit 120A, a reference output unit 130A, a current state estimation unit 170A, and a drinking state determination unit 180A.

[0021] The current information acquiring unit 110A has the same configuration as the current information acquiring unit 110 already described, and acquires current information including biological data of the observed subject in chronological order. The current information acquiring unit 110A shown in FIG. 2 acquires the current information output by the current information output device 300.

[0022] The current information storage unit 120A is configured in the same manner as the current information storage unit 120 already described, and stores current information including biological data of the subject in chronological order.

[0023] The reference output unit 130A outputs a reference value that is a criterion for determining whether the observed subject is in a drinking state. The reference output unit 130A uses current information accumulated in chronological order and current information at the current time to output a reference value indicating a feature amount when the observed subject is likely to be in a non-drinking state during a time period determined based on the current information at the current time. The reference output unit 130A may be configured to determine a time period and then directly calculate a reference value for that time period, or may be configured to store a reference value for the entire time period in advance, correct it according to the determined time period, and output the corrected reference value. The reference output unit 130A shown in FIG. 2 includes a reference value calculation unit 131A (131), a reference value storage unit 132A (132), and a reference value extraction unit 133A (133).

[0024] The reference value calculation unit 131A acquires the accumulated current information and calculates a reference value indicating a characteristic amount for a time period determined based on the accumulated current information when the person is likely to be in a non-drinking state. The reference value calculation unit 131A uses the accumulated current information to determine a time period for each piece of current information. The reference value calculation unit 131A determines a time period based on the accumulated current information and the time of the respective pieces of current information. Specifically, for example, the reference value calculation unit 131A determines a time period based on a time period such as morning or afternoon (see embodiment 2) or morning, afternoon, or night (see embodiment 2). Other time periods may also be used, for example, time periods divided into one-hour or several-hour intervals. By determining a time period in this manner, it is possible to set a reference value for each time period that takes into account fluctuations in the current information based on circadian rhythms. Alternatively, the reference value calculation unit 131A determines a time period based on the content of the accumulated current information itself. Specifically, for example, the reference value calculation unit 131A analyzes fluctuation patterns of biological data to determine a time period based on a time period such as an exercise state or a non-exercise state (see embodiment 3). By determining the time periods in this manner, it is possible to set a reference value for each time period that takes into account factors that cause fluctuations in the biometric data. For example, it is possible to set a reference value for each time period that takes into account fluctuations in the current information based on the exercise state or non-exercise state. The reference value calculation unit 131A uses the accumulated current information to analyze the pattern of the current information (e.g., pulse rate as biometric data) for each determined time period, thereby calculating a reference value as a feature amount of the current information in a state that is estimated to be a non-drinking state for each time period.

[0025] The reference value storage unit 132A stores a reference value for each time period. The reference value storage unit 132A stores and holds the reference value for each time period calculated by the reference value calculation unit 131A. The reference value storage unit 132A stores, for example, information identifying a time period and the reference value in association with each other.

[0026] The reference value extraction unit 133A outputs a reference value for a time period determined based on current information at the current time. The reference value extraction unit 133A determines a time period based on current information at the current time. The reference value extraction unit 133A extracts and outputs a reference value for the determined time period from the reference values ​​for each time period stored in the reference value storage unit 132A.

[0027] The current state estimation unit 170A estimates a feature quantity indicating whether the observed subject is drinking alcohol at the current time. The current state estimation unit 170A outputs a feature quantity of the drinking state at the current time using a reference value for a time period determined based on current information at the current time and current information at the current time. The current state estimation unit 170A calculates, for example, a value corresponding to the difference between the reference value for a time period including the current time and the pulse rate, which is biological data included in the current information at the current time, as the feature quantity.

[0028] The drinking state determination unit 180A determines whether the observed subject is drinking alcohol based on the feature quantities. The drinking state determination unit 180A outputs a determination result of the drinking state of the observed subject based on the drinking state feature quantities output by the current state estimation unit 170A. For example, the drinking state determination unit 180A analyzes changes in the feature quantities and determines whether the observed subject is drinking alcohol based on the degree of change indicating a change trend. In this case, the drinking state determination unit 180A calculates a degree of change indicating the degree of change in the drinking state feature quantities output by the current state estimation unit 170A and determines the drinking state of the observed subject using the degree of change. For example, the drinking state determination unit 180A performs threshold processing using the degree of change and a change threshold to ultimately determine whether the observed subject is drinking alcohol or not. Alternatively, the drinking state determination unit 180A estimates a drinking score indicating the degree of drinking from the feature quantities and determines whether the observed subject is drinking alcohol using the drinking score. For a configuration example in this case, see the modified example of the drinking state determination unit 180A ( FIG. 6 ) described below. The drinking level determination unit 180A performs threshold processing using the drinking score and the drinking score threshold, for example, to ultimately determine whether the observed subject is drinking alcohol or not.

[0029] The external processing device 500 is a device that acquires the drinking judgment result output by the drinking estimation device 100A and performs processing using the drinking judgment result. The external processing device 500 is an output device that outputs, for example, an image, sound, or light, and uses the image, sound, or light to warn the observed subject that they have been drinking alcohol.

[0030] An example of the processing performed by the drinking estimation device according to the first embodiment of the present disclosure will now be described. FIG. 3 is a flowchart illustrating an example of the processing performed by the drinking estimation device 100 or 100A according to the first embodiment of the present disclosure. The processing illustrated in FIG. 3 is a drinking estimation method performed by the drinking estimation device 100 or 100A. The processing performed by the drinking estimation device 100 and the drinking estimation device 100A differs from each other in that the drinking estimation device A can specify that the source of information used in the processing is the current information output device 300 and that the output destination is the external processing device 500. However, since the internal processing is similar, an example of the processing performed by the drinking estimation device 100A will be described here as a representative example.

[0031] For example, the drinking alcohol estimation device 100A shown in Fig. 2 starts the process shown in Fig. 3 when a drinking alcohol estimation start condition is satisfied, such as when the observed subject gets on a moving body or when the power source of the moving body starts to operate.

[0032] The drinking estimation device 100A then executes a current information acquisition process (step ST1100). In the current information acquisition process, the current information acquisition unit 110A of the drinking estimation device 100A acquires current information, including biometric data of the observed subject, in chronological order. The current information acquisition unit 110A acquires the current information in chronological order, for example, from a biometric data output device 310, such as a current information output device 300 external to the drinking estimation device 100. The current information acquisition unit 110A acquires the current information at each elapsed time. The drinking estimation device 100A accumulates the current information acquired in chronological order. The drinking estimation device 100A stores the current information for each time in the current information accumulation unit 120. Specifically, the current information acquisition unit 110A of the drinking estimation device 100A stores the current information for each time in the current information accumulation unit 120, thereby accumulating the current information in chronological order. Furthermore, the current information acquisition unit 110A outputs the current information at the current time to the reference output unit 130A and also to the current state estimation unit 170A. Alternatively, the current information acquisition unit 110A may output the current information at the current time to the reference output unit 130A and also to the current state estimation unit 170A via the reference output unit 130A.

[0033] The drinking estimation device 100A then executes a reference output process (step ST1200). In the reference output process, the reference output unit 130A of the drinking estimation device 100A uses the current information accumulated in chronological order and the current information at the current time to output a reference value indicating a feature amount when a person would be in a non-drinking state during a time period determined based on the current information at the current time. The reference output unit 130A acquires the current information at the current time acquired by the current information acquisition unit 110A and the current information accumulated by the current information accumulation unit 120. The reference output unit 130A uses the current information accumulated in chronological order and the current information at the current time to calculate a reference value indicating a feature amount when a person would be in a non-drinking state during a time period determined based on the current information at the current time. The reference output unit 130A outputs the calculated reference value to the current state estimation unit 170.

[0034] Here, an example of detailed processing by the reference value calculation unit 131A, the reference value storage unit 132A, and the reference value extraction unit 133A of the reference output unit 130A shown in Fig. 2 in the reference output processing by the reference output unit 130A will be described. Fig. 4 is a flowchart showing an example of detailed processing of the reference output processing in the drinking estimation device 100A according to the first embodiment of the present disclosure. For example, when the reference output unit 130A acquires current information accumulated in chronological order by the current information accumulation unit 120A, the reference output unit 130A starts the processing shown in Fig. 4 ("Start" in Fig. 4).

[0035] Next, the reference output unit 130A executes an accumulated current information acquisition process (step ST1210). In the accumulated current information acquisition process, the reference value calculation unit 131A of the reference output unit 130A acquires the current information accumulated by the current information accumulation unit 120A.

[0036] The reference output unit 130A then executes a reference value calculation process (step ST1220). In the reference value calculation process, the reference value calculation unit 131A of the reference output unit 130A calculates a reference value indicating a feature amount that would indicate a non-drinking state during a time period determined based on current information. The reference value calculation unit 131A acquires current information accumulated over the past day, for example, from the current time. The acquired current information may be current information accumulated over the past week or over the past month. The reference value calculation unit 131A determines a time period for each piece of current information using the accumulated current information. The reference value calculation unit 131A determines a time period based on the accumulated current information and the time of each piece of current information. Specifically, for example, the reference value calculation unit 131A determines a time period based on a time period such as morning or afternoon (see embodiment 2) or morning, noon, or night (see embodiment 2). Alternatively, the reference value calculation unit 131A determines a time period based on the content of the accumulated current information itself. Specifically, for example, the reference value calculation unit 131A analyzes the fluctuation pattern of the biometric data and determines the time period based on a classification such as a time period during an exercise state or a time period during a non-exercise state (see the third embodiment). The reference value calculation unit 131A uses the accumulated current information to analyze the pattern of the current information (e.g., pulse rate as biometric data) for each time period, thereby calculating, as a reference value, a feature amount of the current information in a state estimated to be a non-drinking state for each time period. The reference value calculation unit 131A outputs the calculated reference value for each time period to the reference value storage unit 132A.

[0037] Next, the reference output unit 130A executes a reference value storage process (step ST1230). In the reference value storage process, the reference value storage unit 132A of the reference output unit 130A stores and holds the reference value for each time period calculated by the reference value calculation unit 131A.

[0038] The reference output unit 130A then executes a reference value extraction process (step ST1240). In the reference value extraction process, the reference value extraction unit 133A of the reference output unit 130A outputs a reference value indicating a feature amount that would indicate a non-drinking state during a time period determined based on current information at the current time. Specifically, the reference value extraction unit 133A first acquires the current information output by the current information acquisition unit 110A. The reference value extraction unit 133A then determines a time period based on the current information at the current time. The reference value extraction unit 133A determines a time period based on a classification such as a time period such as morning or afternoon (see the second embodiment), a time period such as morning, noon, or night (see the second embodiment), or a time period such as an exercise state or a non-exercise state (see the third embodiment). The reference value extraction unit 133A then extracts a reference value corresponding to the determined time period from the reference values ​​stored in the reference value storage unit 132A. The reference value extraction unit 133A may correct a reference value that is set in advance based on the extracted reference value. The reference value output unit 130A extracts a reference value in the reference value extraction process, and outputs the extracted reference value to the current state estimation unit 170A, and then ends the process shown in Fig. 4 ("End" in Fig. 4).

[0039] The current accumulated information acquisition process (step ST1210), the reference value calculation process (step ST1220), and the reference value storage process (step ST1230) by the reference output unit 130A may be executed independently at a different timing from the reference value extraction process (step ST1240). For example, they may be executed periodically every day at a preset time, or periodically every few days, every week, or every month, or may be executed irregularly upon receiving a command. Returning to the description of FIG. 3 below.

[0040] The drinking alcohol estimation device 100A then executes a current state estimation process (step ST1300). In the current state estimation process, the current state estimation unit 170A of the drinking alcohol estimation device 100A estimates a feature quantity indicating whether the observed subject is drinking alcohol at the current time. Specifically, the current state estimation unit 170A first acquires a reference value output by the reference output unit 130A. The current state estimation unit 170A also acquires current information for the current time output by the current information acquisition unit 110A. The current state estimation unit 170A then outputs a feature quantity of the drinking state at the current time using the reference value for a time period including the current time and the current information for the current time. For example, if the reference value indicating the drinking state is the pulse rate, the current state estimation unit 170A calculates, as the feature quantity, a value corresponding to the difference between the reference value for the time period determined by the reference output unit 130A and the pulse rate, which is biometric data included in the current information at the current time. The current state estimation unit 170A outputs the feature amount of the drinking state at the current time to the drinking state determination unit 180A.

[0041] The drinking state estimation device 100A then executes a drinking state determination process (step ST1400). In the drinking state determination process, the drinking state determination unit 180A of the drinking state estimation device 100A determines whether the observed subject is drinking based on the feature amount. Specifically, the drinking state determination unit 180A acquires the drinking state feature amount at the current time output by the current state estimation unit 170A. The drinking state determination unit 180A outputs a drinking state determination result for the observed subject based on the drinking state feature amount output by the current state estimation unit 170A. For example, the drinking state determination unit 180A analyzes changes in the feature amount and determines whether the observed subject is drinking based on the degree of change indicating a change trend (see the processing example shown in FIG. 5 , described later). Alternatively, the drinking state determination unit 180A calculates a drinking score indicating the degree of drinking from the feature amount and determines whether the observed subject is drinking based on the drinking score (see the processing example shown in FIG. 7 , described later). When the drinking level determination unit 180A determines that the user is drunk in the drinking level determination process, the drinking level estimation device 100A outputs information indicating that the user is drunk to the external processing device 500.

[0042] If the drinking estimation device 100A determines in the drinking state determination process that the drinking estimation device 100A is in a non-drinking state, the device then proceeds to an end determination process (not shown). In the end determination process, a control unit (not shown) of the drinking estimation device 100A determines whether to end the processing of the drinking estimation device 100A. The control unit (not shown) determines whether to end the processing of the drinking estimation device 100A, for example, in accordance with an external end command or an execution program. If the control unit (not shown) determines not to end the processing of the drinking estimation device 100A, the device proceeds to processing in step ST1100, and repeats processing from step ST1100. If the control unit (not shown) determines to end the processing of the drinking estimation device 100A, the drinking estimation device 100A ends processing ("End" in FIG. 3).

[0043] Here, a detailed example of the drinking alcohol determination process (step ST1400) in the drinking alcohol estimation device 100A will be described. FIG. 5 is a flowchart showing a first example of the detailed process of the drinking alcohol determination process in the drinking alcohol estimation device 100A according to embodiment 1 of the present disclosure. The drinking state determination unit 180A in the drinking alcohol estimation device 100A shown in FIG. 2 starts the process shown in FIG. 5 upon acquiring the drinking state feature at the current time output by the current state estimation unit 170A ("START" in FIG. 5). The drinking state determination unit 180A then executes a feature change degree calculation process (step ST1410 "Calculate feature change degree"). In the feature change degree calculation process, the drinking state determination unit 180A calculates a change degree indicating the degree of change in the feature. The drinking state determination unit 180A then executes a change degree determination process (step ST1411 "Change degree > change degree threshold?"). If the degree of change exceeds the degree of change threshold in the change degree determination process ("YES" in step ST1411), the drinking level determination unit 180A executes a process of determining that the subject is in a drinking state (step ST1412). If the degree of change is equal to or less than the degree of change threshold in the change degree determination process ("NO" in step ST1411), the drinking level determination unit 180A executes a process of determining that the subject is in a non-drinking state (step ST1413). The drinking level determination unit 180A then executes a determination result output process (step ST1414). In the determination result output process, the drinking level determination unit 180A outputs a determination result indicating a drinking state or a non-drinking state to, for example, the external processing device 500. After outputting the determination result, the drinking level determination unit 180A ends the process shown in FIG. 5 ("END" in FIG. 5).

[0044] Here, a modified example of the drinking level determination unit 180A will be described. FIG. 6 is a diagram illustrating a second exemplary configuration of the drinking level determination unit in the drinking estimation device 100A according to the first embodiment of the present disclosure. The drinking level determination unit 180A determines the drinking level of the observed subject using a drinking score indicating the degree of drinking. The drinking level determination unit 180A illustrated in FIG. 6 includes a drinking level estimation unit 181. The drinking level estimation unit 181 estimates the drinking score indicating the degree of drinking based on the drinking level feature quantity output by the current state estimation unit 170A. That is, the drinking level determination unit 180A estimates the drinking score indicating the degree of drinking based on the drinking level feature quantity output by the current state estimation unit 170A, and determines the drinking level of the observed subject using the drinking score. The drinking level estimation unit 181 stores inference rules for estimating the drinking score. The inference rules are stored in the form of, for example, a database table, a weighted average, or a prediction model. The drinking level estimation unit 181 estimates a drinking score based on an inference rule and a feature. The inference rule and the drinking score may be generated and calculated by the following: Alternatively, the drinking level estimation unit 181 may be configured as an estimator that learns the relationship between the feature and the drinking amount using machine learning. The drinking level estimation unit 181 is constructed based on, for example, a random forest, a logistic regression, or a neural network. When configured in this way, the drinking level estimation unit 181 acquires the alertness of the observed subject, inputs the alertness and the feature of the drinking state output by the current state estimation unit 170A into a prediction model, and obtains a drinking score output by the prediction model. That is, the drinking state determination unit 180A acquires the alertness of the observed subject, inputs the alertness and the feature of the drinking state output by the current state estimation unit 170A into a prediction model, and obtains a drinking score output by the prediction model. The drinking state determination unit 180A uses the acquired drinking score to determine the drinking state of the observed subject.

[0045] Next, a processing example for the configuration shown in FIG. 6 will be described. FIG. 7 is a flowchart showing a second example of detailed processing of the drinking state determination process in the drinking state estimation device 100A according to embodiment 1 of the present disclosure. The drinking state determination unit 180A shown in FIG. 6 starts the processing shown in FIG. 7 upon acquiring the drinking state feature amount at the current time output by the current state estimation unit 170A ("START" in FIG. 7). The drinking state determination unit 180A then executes a drinking score estimation process (step ST1441). The drinking state determination unit 180A then executes a drinking score determination process (step ST1442 "Drinking score > drinking score threshold?"). If the drinking state determination unit 180A determines that the drinking score exceeds the drinking score threshold in the drinking score determination process ("YES" in step ST1442), it determines that the observed subject is drinking (step ST1443). If the drinking level determination unit 180A determines in the drinking score determination process that the drinking score is equal to or less than the drinking score threshold ("NO" in step ST1442), it determines that the observed subject is in a non-drinking state (step ST1444). The drinking level determination unit 180A then executes a determination result output process (step ST1445). In the determination result output process, the drinking level determination unit 180A outputs a determination result indicating a drinking state or a non-drinking state to, for example, the external processing device 500. After outputting the determination result, the drinking level determination unit 180A ends the process shown in FIG. 7 ("End" in FIG. 7).

[0046] The above configuration makes it possible to detect drunk driving without adding a dedicated sensor such as a breath alcohol sensor. Furthermore, by personalizing the features used for estimation, individual differences in each feature (e.g., high / low pulse rates among non-drinkers) can be absorbed, enabling accurate estimation of the condition. Furthermore, if there is a temporary increase in heart rate due to factors other than alcohol consumption (e.g., stress), the inference may be stopped to prevent false positives.

[0047] This embodiment provides an example configuration including the following: a drinking estimation device including: a current information acquisition unit (110) that acquires current information including biometric data of a subject in chronological order; a reference output unit (130) that uses the current information accumulated in chronological order and the current information at the current time to output a reference value indicating a feature amount for a case in which the subject is likely to be in a non-drinking state during a time period determined based on the current information at the current time; a current state estimation unit (170) that uses the reference value for a time period determined based on the current information at the current time and the current information at the current time to output a feature amount for the drinking state at the current time; and a drinking state determination unit (180) that outputs a determination result of the drinking state of the subject based on the drinking state feature amount output by the current state estimation unit. This provides an effect of providing a drinking estimation device that enables improved accuracy of drinking estimation compared to conventional devices.

[0048] This embodiment provides an example embodiment including the following configuration: a drinking alcohol estimation method using a drinking alcohol estimation device, the drinking alcohol estimation method comprising: a current information acquisition step in which a current information acquisition unit (110) of the drinking alcohol estimation device acquires current information including biological data of an observed subject in chronological order; a reference output step in which a reference output unit (130) of the drinking alcohol estimation device outputs, using the current information accumulated in chronological order and the current information at the current time, a reference value indicating a feature amount when the observed subject is likely to be in a non-drinking state during a time period determined based on the current information at the current time; a current state estimation step in which a current state estimation unit (170) of the drinking alcohol estimation device outputs a feature amount of the drinking state at the current time using the reference value for the time period determined based on the current information at the current time and the current information at the current time; and a drinking state determination step in which a drinking state determination unit (180) of the drinking alcohol estimation device outputs a determination result of the drinking state of the observed subject based on the drinking state feature amount output by the current state estimation unit. This provides an advantage of providing a drinking alcohol estimation method that enables improved accuracy of drinking alcohol estimation compared to conventional methods.

[0049] This embodiment further illustrates an example embodiment including the following configuration: A drinking alcohol estimation device, characterized in that the observed subject is a passenger of a moving body. This provides an advantage of providing a drinking alcohol estimation device that enables improved estimation of the drinking alcohol consumption of passengers of a moving body compared to conventional devices. Furthermore, the present disclosure provides the same advantage as the above by applying the above configuration to a moving body-mounted system or the above drinking alcohol estimation method.

[0050] This embodiment further illustrates an example embodiment including the following configuration: The observed subject is a vehicle driver, and the biometric data is the pulse rate of the observed subject measured using changes in brightness values ​​of an image of the driver, the pulse rate of the driver measured by a seat sensor, or the pulse rate of the driver measured by a millimeter-wave radar. This further provides an effect of providing a drinking alcohol estimation device that can use existing equipment for acquiring biometric data in a vehicle, thereby reducing the need to add new equipment and enabling improved estimation of the driver's drinking alcohol compared to conventional devices. Furthermore, the present disclosure provides the same effect as the above by applying the above configuration to a vehicle-mounted system or the above drinking alcohol estimation method.

[0051] This embodiment further illustrates an example embodiment including the following configuration: The drinking level determination unit (180) calculates a degree of change indicating a degree of change in the drinking state feature quantity output by the current state estimation unit, and determines the drinking state of the observed subject using the degree of change. This provides an advantage of providing a drinking level estimation device that can improve the accuracy of drinking level estimation in accordance with changes in the drinking state feature quantity. Furthermore, the present disclosure achieves the same advantage as the above by applying the above configuration to a vehicle-mounted system or the above drinking level estimation method.

[0052] This embodiment further illustrates an example embodiment including the following configuration: The drinking level determination unit (180A) estimates a drinking score indicating the degree of drinking based on the drinking state feature output by the current state estimation unit, and determines the drinking level of the observed subject using the drinking score. This provides an advantage of providing a drinking level estimation device that can improve the accuracy of drinking level estimation according to the degree of drinking. Furthermore, the present disclosure achieves the same advantage as the above by applying the above configuration to a vehicle-mounted system or the above drinking level estimation method.

[0053] This embodiment further illustrates an example embodiment including the following configuration: The drinking level determination unit (180A) acquires the alertness level of the observed subject, inputs the alertness level and the drinking state feature value output by the current state estimation unit into a prediction model to obtain the drinking score output by the prediction model, and determines the drinking state of the observed subject using the acquired drinking score. This provides an advantage of providing a drinking level estimation device that can improve the accuracy of drinking level estimation using a prediction model based on, for example, random forest, logistic regression, or neural network. Furthermore, the present disclosure achieves the same advantage as the above by applying the above configuration to a vehicle-mounted system or the above drinking level estimation method.

[0054] Embodiment 2. In the above-described embodiment 1, a form in which whether or not a person has been drinking alcohol is determined using a reference value for each time period has been described. In embodiment 2, a more detailed configuration example of the configuration according to embodiment 1 will be described. Specifically, a configuration example in which whether or not a person has been drinking alcohol is determined using a reference value for each time period based on the concept of circadian rhythm will be described. In embodiment 2, duplicated descriptions of components according to embodiment 2 that are the same as the components according to embodiment 1 already described will be omitted as appropriate.

[0055] Next, a configuration example in which the drinking alcohol estimation device according to the second embodiment of the present disclosure is applied to a mobile-mounted system will be described. FIG. 8 is a diagram illustrating a configuration example of a mobile-mounted system 1 (1B) including a drinking alcohol estimation device 100 (100B) according to the second embodiment of the present disclosure. The mobile-mounted system 1 (1B) illustrated in FIG. 8 includes the drinking alcohol estimation device 100 (100B), a current information output device 300, and an external processing device (output device) 500. The current information output device 300 and the external processing device 500 are configured similarly to the current information output device 300 and the external processing device 500 already described. Further detailed description of the current information output device 300 and the external processing device 500 will be omitted.

[0056] The drinking state estimation device 100 (100B) estimates the drinking state based on feature values ​​that indicate a non-drinking state during a time period determined using current information including the biological data of the observed subject. The drinking state estimation device 100 (100B) shown in FIG. 8 includes a current information acquisition unit 110B, a current information storage unit 120B, a reference output unit 130B, a current state estimation unit 170B, and a drinking state determination unit 180B. The current information acquisition unit 110B is configured similarly to the current information acquisition units 110 and 110A already described. The current information storage unit 120B is configured similarly to the current information storage units 120 and 120A already described. The current state estimation unit 170B is configured similarly to the current state estimation units 170 and 170A already described. The drinking state determination unit 180B is configured similarly to the drinking state determination units 180 and 180A already described.

[0057] The reference output unit 130B, in contrast to the reference output units 130 and 130A already described, outputs a reference value for each time slot in a day that includes the current time. The reference output unit 130B calculates a reference value for each time slot in a day using current information accumulated in chronological order, and outputs a reference value for the time slot that includes the current time that is determined using the current information at the current time. The reference output unit 130B shown in FIG. 8 includes a reference value calculation unit 131B, a reference value storage unit 132B, and a reference value extraction unit 133B.

[0058] The reference value calculation unit 131B acquires the accumulated current information and calculates a reference value indicating a feature amount in which a person would be in a non-drinking state during a time period determined based on the accumulated current information. The reference value calculation unit 131B uses the accumulated current information to determine a time period for each piece of current information. The reference value calculation unit 131B determines a time period based on the accumulated current information and the time of each piece of current information. Specifically, for example, the reference value calculation unit 131B determines a time period based on a time period such as morning or afternoon, or morning, noon, or night. Furthermore, the time period may be determined based on a time period other than these time periods, for example, in units of one hour or several hours. By determining a time period in this manner, a reference value for each time period can be set that takes into account fluctuations in the current information based on circadian rhythms.

[0059] The reference value storage unit 132B has a configuration similar to that of the reference value storage unit 132A already described, and stores and holds the reference values ​​for each time period calculated by the reference value calculation unit 131 B. The reference value storage unit 132B stores, for example, information identifying a time period and the reference values ​​in association with each other.

[0060] The reference value extraction unit 133B outputs a reference value for a time period determined based on current information at the current time. The reference value extraction unit 133B determines a time period based on current information at the current time. The reference value extraction unit 133B determines a time period that includes the current time. The reference value extraction unit 133B extracts and outputs a reference value for the determined time period from the reference values ​​for each time period held in the reference value storage unit 132B.

[0061] Next, a processing example of the drinking alcohol estimation device 100B according to the second embodiment of the present disclosure will be described. The processing of the drinking alcohol estimation device 100B according to the second embodiment differs from the processing of the drinking alcohol estimation device 100A according to the first embodiment mainly in the details of the reference output processing. Here, an example of detailed processing of the reference output processing will be described with reference to FIG. 4 . Additionally, the reference values ​​according to the second embodiment will be described with reference to FIGS. 9 and 10 . FIG. 9 is a first explanatory diagram for explaining the reference values ​​used in the drinking alcohol estimation device 100 (100B) according to the second embodiment of the present disclosure. FIG. 9 shows an example of a change in heart rate 2000 over a day. FIG. 10A in FIG. 10 is a second explanatory diagram for explaining the reference values ​​used in the drinking alcohol estimation device 100 (100B) according to the second embodiment of the present disclosure, and FIG. 10B in FIG. 10 is a third explanatory diagram for explaining the reference values ​​used in the drinking alcohol estimation device 100 (100B) according to the second embodiment of the present disclosure. Fig. 10A shows an example of changes in person A's heart rate 2010 in the morning and an example of changes in person A's heart rate 2020 in the afternoon. Fig. 10B shows an example of changes in person B's heart rate 2030 in the morning and an example of changes in person B's heart rate 2040 in the afternoon.

[0062] For example, when the reference output unit 130B acquires the current information accumulated in chronological order by the current information accumulation unit 120B, it starts the process shown in Fig. 4 ("Start" in Fig. 4).

[0063] Next, the reference output unit 130B executes an accumulated current information acquisition process (step ST1210). In the accumulated current information acquisition process, the reference value calculation unit 131B of the reference output unit 130B acquires the current information accumulated by the current information accumulation unit 120B.

[0064] The reference output unit 130B then executes a reference value calculation process (step ST1220). In the reference value calculation process, the reference value calculation unit 131B of the reference output unit 130B calculates a reference value indicating a feature amount that would indicate a non-drinking state during a time period determined based on current information. The reference value calculation unit 131B acquires current information accumulated over the past day from the current time, for example. The acquired current information may be current information accumulated over the past week, for example, or current information accumulated over the past month, for example. The reference value calculation unit 131B determines a time period for each piece of current information using the accumulated current information. The reference value calculation unit 131B determines a time period based on the accumulated current information and the time of each piece of current information. Specifically, for example, the reference value calculation unit 131B determines a time period based on a time period such as morning or afternoon, or a time period such as morning, noon, or night.

[0065] Here, the time period determined based on the current information will be described. As shown in FIG. 9, the heart rate 2000 fluctuates so that it is relatively low in the morning, increases toward noon, remains relatively high in the afternoon, and gradually decreases toward night. Furthermore, as shown in FIGS. 10A and 10B, in addition to fluctuations depending on the time period, the fluctuations also differ for each individual. Therefore, if a unique reference value for the feature indicated in the current information including biometric data is determined, there is a high possibility that the drinking level will be erroneously determined. Therefore, in this embodiment, a time period is determined according to each time included in the current information, and a reference value is calculated using the current information for each determined time period.

[0066] The reference value calculation unit 131B uses the accumulated current information to analyze patterns of the current information (e.g., pulse rate as biological data) for each time period, and calculates, as reference values, feature quantities of the current information in a state estimated to be a non-drinking state for each time period in a day. The reference value calculation unit 131B outputs the calculated reference values ​​for each time period in a day to the reference value storage unit 132B.

[0067] Next, the reference output unit 130B executes a reference value storage process (step ST1230). In the reference value storage process, the reference value storage unit 132B of the reference output unit 130B stores and holds the reference values ​​for each time period in a day calculated by the reference value calculation unit 131B.

[0068] The reference output unit 130B then executes a reference value extraction process (step ST1240). In the reference value extraction process, the reference value extraction unit 133B of the reference output unit 130B outputs a reference value indicating a feature amount that would indicate a non-drinking state during a time period determined based on current information at the current time. Specifically, the reference value extraction unit 133B first acquires the current information output by the current information acquisition unit 110B. The reference value extraction unit 133B then determines a time period based on the current information at the current time. The reference value extraction unit 133B determines a time period based on a time period classification, such as morning or afternoon, or morning, noon, or night. The reference value extraction unit 133B then extracts a reference value corresponding to the determined time period from the reference values ​​stored in the reference value storage unit 132B. Note that the reference value extraction unit 133B may correct a separately preset reference value based on the extracted reference value. The reference output unit 130B extracts a reference value in the reference value extraction process, outputs the extracted reference value to the current state estimation unit 170B, and then ends the process shown in FIG. 4 ("End" in FIG. 4).

[0069] The current accumulated information acquisition process (step ST1210), the reference value calculation process (step ST1220), and the reference value storage process (step ST1230) performed by the reference output unit 130B may be executed independently at a different timing from the reference value extraction process (step ST1240). For example, they may be executed periodically every day at a preset time, or periodically every few days, once a week, or once a month, or may be executed irregularly upon receiving a command.

[0070] This embodiment further illustrates an example embodiment including the following configuration: The drinking alcohol estimation device according to claim 1, wherein the reference output unit (130B) calculates a reference value for each time period of a day using current information accumulated in chronological order, and outputs a reference value for a time period including the current time using current information at the current time. This provides an advantage of providing a drinking alcohol estimation device that can perform drinking alcohol estimation using a reference value appropriate for each time period of a day, thereby improving the accuracy of drinking alcohol estimation compared to conventional devices. Furthermore, the present disclosure provides the same advantage as described above by applying the above configuration to a vehicle-mounted system or the above drinking alcohol estimation method.

[0071] Embodiment 3. In the above-described embodiment 2, as a more detailed configuration example of the configuration according to embodiment 1, a configuration example in which whether or not a person has been drinking alcohol is described using a reference value for each time period based on the concept of circadian rhythm. In embodiment 3, a configuration example in which whether or not a person has been drinking alcohol is described using a reference value for a time period in which the person is exercising will be described. In embodiment 3, duplicated descriptions of components according to embodiment 3 that are the same as the components according to embodiment 1 or embodiment 2 already described will be omitted as appropriate.

[0072] A configuration example will be described in which a drinking alcohol estimation device according to a third embodiment of the present disclosure is applied to a mobile-mounted system. FIG. 11 is a diagram illustrating a configuration example of a mobile-mounted system 1 (1C) including a drinking alcohol estimation device 100 (100C) according to the third embodiment of the present disclosure. The mobile-mounted system 1 (1C) illustrated in FIG. 11 includes the drinking alcohol estimation device 100 (100C), a current information output device 300, and an external processing device (output device) 500. The current information output device 300 and the external processing device 500 are configured similarly to the current information output device 300 and the external processing device 500 already described. Further detailed description of the current information output device 300 and the external processing device 500 will be omitted.

[0073] The drinking state estimation device 100 (100C) estimates the drinking state based on feature values ​​that indicate a non-drinking state during a time period determined using current information including the biological data of the observed subject. The drinking state estimation device 100 (100C) shown in FIG. 11 includes a current information acquisition unit 110C, a current information storage unit 120C, a reference output unit 130C, a reference value calculation unit 131C, a reference value storage unit 132C, a reference value extraction unit 133C, a current state estimation unit 170C, and a drinking state determination unit 180C. The current information acquisition unit 110C is configured similarly to the current information acquisition units 110 and 110A described above. The current information storage unit 120C is configured similarly to the current information storage units 120 and 120A described above. The current state estimation unit 170C is configured similarly to the current state estimation units 170 and 170A described above. The drinking level determination unit 180C is configured similarly to the drinking level determination units 180 and 180A already described.

[0074] The reference output unit 130C, in the reference output units 130 and 130A already described, outputs a reference value for each time period, particularly for an exercise state or a non-exercise state. The reference output unit 130C calculates a reference value for a time period when the observed subject is exercising using current information accumulated in chronological order, and outputs the reference value for a time period when the observed subject is exercising using current information at the current time when the observed subject is exercising. The reference output unit 130C shown in FIG. 11 includes a reference value calculation unit 131C, a reference value storage unit 132C, and a reference value extraction unit 133C.

[0075] The reference value calculation unit 131C acquires the accumulated current information and calculates a reference value indicating a feature amount when the person is likely to be in a non-drinking state during a time period determined based on the accumulated current information. The reference value calculation unit 131C uses the accumulated current information to determine a time period for each piece of current information. The reference value calculation unit 131C determines a time period based on the content of the accumulated current information itself. Specifically, for example, the reference value calculation unit 131C analyzes the variation pattern of biometric data and determines a time period based on a classification such as a time period for each variation factor of the biometric data. By determining a time period in this manner, it is possible to set a reference value for each time period that takes into account the variation factors of the biometric data. For example, it is possible to set a reference value for each time period that takes into account variations in the current information based on whether the person is exercising or not.

[0076] The reference value storage unit 132C has a configuration similar to that of the reference value storage unit 132A already described, and stores and holds the reference values ​​for each time period calculated by the reference value calculation unit 131C. The reference value storage unit 132C stores, for example, information identifying a time period and the reference values ​​in association with each other.

[0077] The reference value extraction unit 133C outputs a reference value for a time period determined based on current information at the current time. The reference value extraction unit 133C determines a time period based on current information at the current time. If the feature value indicated in the current information at the current time indicates an exercise state, the reference value extraction unit 133C determines a time period in which the current time is an exercise state. The reference value extraction unit 133C extracts and outputs a reference value for the determined time period from the reference values ​​for each time period stored in the reference value storage unit 132C.

[0078] A processing example of the drinking alcohol estimation device according to embodiment 3 of the present disclosure will be described. The processing of the drinking alcohol estimation device 100C according to embodiment 3 differs from the processing of the drinking alcohol estimation device 100A according to embodiment 1 mainly in the details of the reference output processing. Here, an example of detailed processing of the reference output processing will be described with reference to FIG. 4. Additionally, the reference value according to embodiment 3 will be described with reference to FIG. 12.

[0079] For example, when the reference output unit 130C acquires the current information accumulated in chronological order by the current information accumulation unit 120C, it starts the process shown in Fig. 4 ("Start" in Fig. 4).

[0080] Next, the reference output unit 130C executes an accumulated current information acquisition process (step ST1210). In the accumulated current information acquisition process, the reference value calculation unit 131C of the reference output unit 130C acquires the current information accumulated by the current information accumulation unit 120C.

[0081] The reference output unit 130C then executes a reference value calculation process (step ST1220). In the reference value calculation process, the reference value calculation unit 131C of the reference output unit 130C calculates a reference value indicating a feature amount when the person is likely to be in a non-drinking state during a time period determined based on the current information. The reference value calculation unit 131C acquires current information accumulated over the past day from the current time, for example. The acquired current information may be current information accumulated over the past week, for example, or may be current information accumulated over the past month, for example. The reference value calculation unit 131C uses the accumulated current information to determine a time period for each piece of current information.

[0082] Here, the time period determined based on current information will be described. FIG. 12A in FIG. 12 is a first explanatory diagram illustrating the reference value used in the drinking alcohol estimation device 100 (100C) according to embodiment 3 of the present disclosure. FIG. 12B in FIG. 12 is a second explanatory diagram illustrating the reference value used in the drinking alcohol estimation device 100 (100C) according to embodiment 3 of the present disclosure. FIG. 12A shows an example of a change in heart rate 3010 during post-exercise driving. FIG. 12B shows an example of a change in heart rate 3020 during normal driving. Compared to the heart rate 3020 during normal driving, the heart rate 3010 during post-exercise driving continues to be elevated during exercise for a time α minutes from the start of driving. Therefore, in this embodiment, the time period is determined based on a time period classification such as an exercise state or a non-exercise state. Note that, in the present disclosure, any time period classification that causes fluctuations in biometric data may be used, and classifications other than an exercise state or a non-exercise state may also be used.

[0083] The reference value calculation unit 131C determines a time period based on the accumulated current information and the time of each of the current information. Specifically, for example, the reference value calculation unit 131C determines a time period based on a classification such as a time period during an exercise state or a time period during a non-exercise state. The reference value calculation unit 131C uses the accumulated current information to analyze the pattern of the current information (e.g., pulse rate as biological data) for each time period, thereby calculating, as a reference value, a feature amount of the current information in a state estimated to be a non-drinking state for each time period during an exercise state or a time period during a non-exercise state. The reference value calculation unit 131C outputs the calculated reference value for each time period during an exercise state or a time period during a non-exercise state to the reference value storage unit 132C.

[0084] Next, the reference output unit 130C executes a reference value storage process (step ST1230). In the reference value storage process, the reference value storage unit 132C of the reference output unit 130C stores and holds the reference value calculated by the reference value calculation unit 131C for each time period of exercise or non-exercise.

[0085] The reference output unit 130C then executes a reference value extraction process (step ST1240). In the reference value extraction process, the reference value extraction unit 133C of the reference output unit 130C outputs a reference value indicating a feature amount that would indicate a non-drinking state during a time period determined based on current information at the current time. Specifically, the reference value extraction unit 133C first acquires the current information output by the current information acquisition unit 110C. The reference value extraction unit 133C then determines a time period based on the current information at the current time. The reference value extraction unit 133C determines a time period based on a classification, such as a time period during exercise or a time period during non-exercise. The reference value extraction unit 133C then extracts a reference value corresponding to the determined time period from the reference values ​​stored in the reference value storage unit 132C. Note that the reference value extraction unit 133C may correct a separately preset reference value based on the extracted reference value. The reference output unit 130C extracts a reference value in the reference value extraction process, outputs the extracted reference value to the current state estimation unit 170C, and then ends the process shown in FIG. 4 ("End" in FIG. 4).

[0086] The current accumulated information acquisition process (step ST1210), the reference value calculation process (step ST1220), and the reference value storage process (step ST1230) performed by the reference output unit 130C may be executed independently at a different timing from the reference value extraction process (step ST1240). For example, they may be executed periodically every day at a preset time, or periodically every few days, once a week, or once a month, or may be executed irregularly upon receiving a command.

[0087] This embodiment further illustrates an example embodiment including the following configuration: The drinking alcohol estimation device according to claim 1, wherein the reference output unit (130C) calculates a reference value for a time period when the observed subject is exercising using current information accumulated in chronological order, and outputs the reference value for a time period when the observed subject is exercising using current information at the current time when the observed subject is exercising. This provides an advantage of providing a drinking alcohol estimation device that can perform drinking alcohol estimation using a reference value for a time period when the observed subject is exercising, thereby improving the accuracy of drinking alcohol estimation compared to conventional devices. Furthermore, the present disclosure provides the same advantage as the above by applying the above configuration to a vehicle-mounted system or the above drinking alcohol estimation method.

[0088] Embodiment 4. In the above-mentioned embodiment 2, a configuration example was described in which a reference value for each time period based on the concept of circadian rhythm was used to determine whether or not a person was drinking alcohol. In the above-mentioned embodiment 3, a configuration example was described in which a reference value for a time period in which a person is exercising was used to determine whether or not a person was drinking alcohol. In embodiment 4, a configuration example in which embodiments 2 and 3 are combined will be described. In embodiment 4, duplicated descriptions of components according to embodiment 4 that are the same as components according to embodiment 1, embodiment 2, or embodiment 3 already described will be omitted as appropriate.

[0089] A configuration example when a drinking alcohol estimation device according to a fourth embodiment of the present disclosure is applied to a mobile-mounted system will be described. A configuration example of a drinking alcohol estimation device according to the fourth embodiment of the present disclosure and a mobile-mounted system including the drinking alcohol estimation device will be described. FIG. 13 is a diagram showing a configuration example of a mobile-mounted system 1 (1D) including a drinking alcohol estimation device 100 (100D) according to the fourth embodiment of the present disclosure. The mobile-mounted system 1 (1D) shown in FIG. 13 is configured to include the drinking alcohol estimation device 100 (100D), a current information output device 300, and an external processing device (output device) 500. The current information output device 300 and the external processing device 500 are configured similarly to the current information output device 300 and the external processing device 500 already described. Further detailed description of the current information output device 300 and the external processing device 500 will be omitted.

[0090] The drinking state estimation device 100 (100D) estimates the drinking state based on feature values ​​that indicate a non-drinking state during a time period determined using current information including the biological data of the observed subject. The drinking state estimation device 100 (100D) shown in FIG. 13 includes a current information acquisition unit 110D, a current information accumulation unit 120D, a reference output unit 130D, a reference value calculation unit 131D, a reference value storage unit 132D, a reference value extraction unit 133D, a reference value correction unit 134 (134D), a current state estimation unit 170D, and a drinking state determination unit 180D. The current information acquisition unit 110D is configured similarly to the current information acquisition units 110 and 110A already described. The current information accumulation unit 120D is configured similarly to the current information accumulation units 120 and 120A already described. The current state estimation unit 170D is configured similarly to the current state estimation units 170 and 170A already described. The drinking level determination unit 180D is configured similarly to the drinking level determination units 180 and 180A already described.

[0091] The reference output unit 130D, in the reference output units 130 and 130A already described, outputs a reference value for each time period of a day, specifically for a time period including the current time among the time periods in an exercise state or a time period in a non-exercise state. The reference output unit (130D) uses the current information accumulated in chronological order and the current information at the current time to output a reference value for each time period of a day, when the observed subject is exercising. That is, the reference output unit 130D can output a reference value for each time period based on a combination of time period classifications based on circadian rhythms and time period classifications based on the content of the characteristics indicated in the current information, using the time indicated in the current information. The reference output unit 130D shown in FIG. 13 includes a reference value calculation unit 131D, a reference value storage unit 132D, a reference value extraction unit 133D, and a reference value correction unit 134D.

[0092] The reference value calculation unit 131D acquires accumulated current information and calculates a reference value indicating a feature amount for a time period determined based on the accumulated current information. The reference value calculation unit 131D uses the accumulated current information to determine a time period for each piece of current information. The reference value calculation unit 131D determines a time period for each variable factor of the current information based on the accumulated current information and the time of the current information. Specifically, for example, the reference value calculation unit 131D determines a time period based on a variable factor of the current information, such as an exercise state or a non-exercise state, for each time period, such as morning or afternoon, or morning, noon, or night. By determining a time period in this manner, a reference value for each time period can be set that takes into account the variable factors of the biometric data while taking into account the variable factors of the current information based on the circadian rhythm. The reference value calculation unit 131D uses the accumulated current information to analyze the pattern of the current information (e.g., pulse rate as biometric data) for each determined time period, thereby calculating a reference value for the feature amount of the current information for a state estimated to be a non-drinking state for each time period, whether during exercise or non-exercise.

[0093] The reference value storage unit 132D has a configuration similar to that of the reference value storage unit 132A already described, and stores and holds the reference values ​​for each time period calculated by the reference value calculation unit 131D. The reference value storage unit 132D stores, for example, information identifying a time period and the reference values ​​in association with each other.

[0094] The reference value extraction unit 133D outputs a reference value for a time period determined based on current information at the current time. The reference value extraction unit 133D determines a time period based on current information at the current time. If the feature value indicated in the current information at the current time indicates an exercise state, the reference value extraction unit 133D determines a time period in which the current time is an exercise state. The reference value extraction unit 133D extracts and outputs a reference value for the determined time period from the reference values ​​for each time period stored in the reference value storage unit 132D.

[0095] The reference value correcting unit 134D corrects the reference value output by the reference value extracting unit 133D according to the driving time zone based on the feature amount during rest after long-term driving, and outputs the corrected reference value.

[0096] An example of processing by the drinking alcohol estimation device according to the fourth embodiment of the present disclosure will be described. The processing by the drinking alcohol estimation device 100D according to the fourth embodiment differs from the processing by the drinking alcohol estimation device 100A according to the first embodiment mainly in the details of the reference output processing. FIG. 14A in FIG. 14 is a first explanatory diagram for explaining the reference values ​​used in the drinking alcohol estimation device 100 (100D) according to the fourth embodiment of the present disclosure. FIG. 14B in FIG. 14 is a second explanatory diagram for explaining the reference values ​​used in the drinking alcohol estimation device 100 (100D) according to the fourth embodiment of the present disclosure. FIG. 15 is a flowchart showing an example of processing by the drinking alcohol estimation device 100D according to the fourth embodiment of the present disclosure. For example, when the reference output unit 130D acquires current information accumulated in chronological order by the current information accumulation unit 120D, the reference output unit 130D starts the processing shown in FIG. 15 ("Start" in FIG. 15).

[0097] Next, reference output unit 130D executes an accumulated current information acquisition process (step ST4210). In the accumulated current information acquisition process, reference value calculation unit 131D of reference output unit 130D acquires the current information accumulated by current information accumulation unit 120D.

[0098] The reference output unit 130D then executes a reference value calculation process (step ST4220). In the reference value calculation process, the reference value calculation unit 131D of the reference output unit 130D calculates a reference value indicating a feature amount when the person is likely to be in a non-drinking state during a time period determined based on the current information. The reference value calculation unit 131D acquires current information accumulated over the past day from the current time, for example. The acquired current information may be current information accumulated over the past week, for example, or may be current information accumulated over the past month, for example. The reference value calculation unit 131D uses the accumulated current information to determine a time period for each piece of current information.

[0099] Here, the time period determined based on the current information will be described. Fig. 14A in Fig. 14 shows an example of a change in heart rate 4010 due to exercise in a first time period. Fig. 14B in Fig. 14 shows an example of a change in heart rate 4020 due to exercise in a second time period. The heart rate 4010 due to exercise in the first time period remains high for a relatively long period of time compared to the heart rate 4020 due to exercise in the second time period. Therefore, in this embodiment, a reference value for an exercise state or a non-exercise state can be output for each time period.

[0100] The reference value calculation unit 131D determines a time period based on the accumulated current information and the time of each of the current information. Specifically, for example, the reference value calculation unit 131D determines a time period based on a classification such as an exercise state or a non-exercise state as a factor that affects the current information, for each time period, such as morning or afternoon, or morning, noon, or night. More specifically, the reference value calculation unit 131D determines a time period based on a classification such as a morning time period in which the subject is exercising, a morning time period in which the subject is not exercising, an afternoon time period in which the subject is exercising, or an afternoon time period in which the subject is not exercising. The reference value calculation unit 131D uses the accumulated current information to analyze the pattern of the current information (e.g., pulse rate as biometric data) for each of the determined time periods, thereby calculating, as a reference value, a feature amount of the current information in a state estimated to be a non-drinking state for each of the exercise and non-exercise time periods. Specifically, for example, the reference value for the morning time period when the user is exercising, the reference value for the morning time period when the user is not exercising, the reference value for the afternoon time period when the user is exercising, or the reference value for the afternoon time period when the user is not exercising is output.

[0101] Next, reference output unit 130D executes a reference value storage process (step ST4230). In the reference value storage process, reference value storage unit 132D of reference output unit 130D stores and holds the reference value for each time period calculated by reference value calculation unit 131D.

[0102] The reference output unit 130D then executes a reference value extraction process (step ST4240). In the reference value extraction process, the reference value extraction unit 133D of the reference output unit 130D outputs a reference value indicating a feature amount that would indicate a non-drinking state during a time period determined based on current information at the current time. Specifically, the reference value extraction unit 133D first acquires the current information output by the current information acquisition unit 110D. The reference value extraction unit 133D then determines a time period based on the current information at the current time. The reference value extraction unit 133D determines a time period based on, for example, a time period during which the subject is exercising in the morning, a time period during which the subject is not exercising in the morning, a time period during which the subject is exercising in the afternoon, or a time period during which the subject is not exercising in the afternoon. The reference value extraction unit 133D then extracts a reference value corresponding to the determined time period from the reference values ​​stored in the reference value storage unit 132D. The reference value extracting unit 133D may correct a reference value that is set in advance based on the extracted reference value. The reference output unit 130D extracts a reference value in the reference value extraction process and outputs the extracted reference value to the reference value correcting unit 134D.

[0103] Next, reference output unit 130D executes a reference value correction process (step ST4250). Reference value correction unit 134D of reference output unit 130D acquires the reference value output by reference value extraction unit 133D and corrects the acquired reference value according to the time period of the acquired reference value. Reference value correction unit 134D outputs the corrected reference value. After outputting the corrected reference value to current state estimation unit 170D, reference output unit 130D ends the process shown in FIG. 15 ("End" in FIG. 15).

[0104] This embodiment further illustrates an example embodiment including the following configuration: The drinking alcohol estimation device according to claim 1, wherein the reference output unit (130D) uses current information accumulated in chronological order and current information at the current time to output, when the observed subject is exercising, a reference value for each time period of a day, the reference value for when the observed subject is exercising. This provides an advantage of providing a drinking alcohol estimation device that can perform drinking alcohol estimation using a reference value appropriate for the time period of a day, the reference value for the time period when the observed subject is exercising, thereby improving the accuracy of drinking alcohol estimation compared to conventional devices. Furthermore, the present disclosure provides the same advantage as described above by applying the above configuration to a vehicle-mounted system or the drinking alcohol estimation method.

[0105] Embodiment 5. In the above-described Embodiments 1, 2, 3, and 4, configuration examples have been described in which the drinking state is estimated using biological data such as pulse rate as current information. In Embodiment 5, a configuration example will be described in which the current information includes data other than biological data such as pulse rate (heart rate). In Embodiment 5, duplicated descriptions of components according to Embodiment 5 that are the same as components according to Embodiment 1, 2, 3, or 4 already described will be omitted as appropriate.

[0106] A configuration example when a drinking alcohol estimation device according to a fifth embodiment of the present disclosure is applied to a mobile-mounted system will be described. A configuration example of a drinking alcohol estimation device according to the fifth embodiment of the present disclosure and a mobile-mounted system including the drinking alcohol estimation device will be described. FIG. 16 is a diagram showing a configuration example of a mobile-mounted system 1 (1E) including a drinking alcohol estimation device 100 (100E) according to the fifth embodiment of the present disclosure. The mobile-mounted system 1 (1E) shown in FIG. 16 is configured to include the drinking alcohol estimation device 100 (100E), a current information output device 300, and an external processing device (output device) 500. The external processing device 500 has a configuration similar to the external processing device 500 already described. Further detailed description of the external processing device 500 will be omitted.

[0107] The current information output device 300 shown in FIG. 16 includes a biometric data output device 310, a moving body information output device (vehicle information output device) 320, and a moving body surroundings information output device (vehicle surroundings information output device) 330.

[0108] The biometric data output device 310 outputs biometric data of the observed subject. The biometric data output device 310 outputs the biometric data of the observed subject at the current time in chronological order to the alcohol consumption estimation device 100E over time. The biometric data output device 310 outputs biometric data of the observed subject, such as pulse rate, alertness, facial direction, gaze, pupil size, and respiratory rate, in chronological order. For example, if the moving object is a vehicle, the biometric data output device 310 may include, for example, a driver monitoring system (DMS), a seat sensor, or a millimeter-wave radar. That is, if the moving object is a vehicle, the observed subject is a driver, and the biometric data is pulse rate, the biometric data may be the pulse rate of the observed subject measured using changes in brightness values ​​of an image of the driver captured by a DMS or the like, the pulse rate of the driver measured by a seat sensor, or the pulse rate of the driver measured by a millimeter-wave radar. Furthermore, when the moving object is a vehicle and the observed subject is a driver, the biometric data includes the observed subject's level of alertness estimated using changes in brightness values ​​in an image of the driver captured by a DMS or the like. The level of alertness can be obtained by estimating the level of alertness using the driver's eye openness, blink frequency and time, cumulative time, and yawn characteristics. The level estimation method may use an estimator that has learned the relationship between the driver's eye and mouth information and drowsiness using machine learning. When configured in this manner, the biometric data output device 310 can acquire and output biometric data of the observed subject while riding in the moving object or driving the vehicle. The biometric data output device 310 may also be configured to output biometric data observed using a portable device such as a smartwatch. When configured in this manner, the biometric data output device 310 can acquire and output biometric data from everyday life other than when riding in the moving object or driving the vehicle. Note that the biometric data is not limited to the examples provided, as long as it indicates information about the observed subject that changes when the observed subject drinks alcohol.

[0109] The mobile object information output device (vehicle information output device) 320 outputs mobile object information indicating an operation on the mobile object or the movement of the mobile object due to the operation to the drinking estimation device 100E. In other words, the current information in this embodiment further includes mobile object information, which is information indicating an operation on the mobile object by an occupant of the mobile object or the movement of the mobile object due to the operation. When the mobile object is a vehicle, the mobile object information is vehicle information. The vehicle information is based on at least one of the driver's steering operation, braking operation, or accelerator operation, or the associated vehicle movement (e.g., change in vehicle speed, change in yaw rate, change in vehicle position within the lane, etc.). In other words, when the mobile object is a vehicle, the mobile object information is based on at least one of the driver's steering operation, braking operation, or accelerator operation, or the vehicle movement associated with each operation.

[0110] The mobile object surrounding information output device (vehicle surrounding information output device) 330 acquires mobile object surrounding information indicating objects around the mobile object that should be noticed by the occupant of the mobile object, and outputs the information to the drinking estimation device 100E. That is, the current information further includes mobile object surrounding information, which is information indicating the situation around the mobile object. When the mobile object is a vehicle, the mobile object surrounding information is vehicle surrounding information. The vehicle surrounding information is vehicle surrounding information based on the types and positions of objects (e.g., pedestrians, bicycles, vehicles, traffic lights, signs, etc.) that should be noticed around the vehicle while driving, as measured by an external camera, ultrasonic sensor, radar, etc. That is, when the mobile object is a vehicle, the mobile object surrounding information is vehicle surrounding information based on the types, positions, or types and positions of objects around the vehicle that the driver should be paying attention to around the vehicle.

[0111] The drinking alcohol estimation device 100 (100E) shown in FIG. 16 includes a current information acquisition unit 110E, a current information accumulation unit 120E, a reference output unit 130E, a reference value calculation unit 131E, a reference value storage unit 132E, a reference value extraction unit 133E, a current state estimation unit 170E, and a drinking state determination unit 180E.

[0112] The current information acquisition unit 110E acquires current information including biometric data of the observed subject in chronological order. The current information acquisition unit 110E shown in FIG. 2 acquires current information output by the current information output device 300. Specifically, the current information acquisition unit 110A acquires current information output by the biometric data output device 310, the mobile object information output device (vehicle information output device) 320, and the mobile object surrounding information output device (vehicle surrounding information output device) 330. That is, the current information in this embodiment includes biometric data, mobile object information, and mobile object surrounding information.

[0113] The current information storage unit 120E is configured in the same manner as the current information storage unit 120 already described, and stores current information including biological data of the subject in chronological order.

[0114] The reference output unit 130E uses the accumulated current information to output a reference value that is a criterion for determining whether the observed subject is in a drinking state. The reference output unit 130E uses the chronologically accumulated current information and the current information at the current time to output a reference value that indicates a feature amount when the observed subject is likely to be in a non-drinking state during a time period determined based on the current information at the current time. The reference output unit 130E may be configured to determine a time period and then directly calculate a reference value for that time period, or may be configured to store a reference value for the entire time period in advance, correct it according to the determined time period, and output the corrected reference value. The reference output unit 130E shown in FIG. 16 includes a reference value calculation unit 131E (131), a reference value storage unit 132E (132), and a reference value extraction unit 133E (133).

[0115] The reference value calculation unit 131E differs from the already-described reference value calculation units 131 (131A, 131B, 131C, 131D) in that the reference value to be output takes into account or includes the alertness, mobile object information, or mobile object surrounding information. When the alertness, mobile object information, and mobile object surrounding information are taken into account in the reference value, the reference value is calculated by, for example, combining feature quantities indicated in the pulse rate, alertness, mobile object information, and mobile object surrounding information. When the alertness, mobile object information, and mobile object surrounding information are included in the reference value, multiple reference values ​​are calculated for each piece of information included in the current information based on, for example, feature quantities indicated in the pulse rate, alertness, mobile object information, and mobile object surrounding information.

[0116] The reference value storage unit 132E stores a reference value for each time period. The reference value storage unit 132E stores and holds the reference value for each time period calculated by the reference value calculation unit 131E. The reference value storage unit 132E stores, for example, information identifying a time period and the reference value in association with each other.

[0117] The reference value extraction unit 133E outputs a reference value for a time period determined based on current information at the current time. The reference value extraction unit 133E determines a time period based on current information at the current time. The reference value extraction unit 133E extracts and outputs a reference value for the determined time period from the reference values ​​for each time period stored in the reference value storage unit 132E.

[0118] The current state estimation unit 170E estimates a feature quantity indicating whether the observed subject is drinking alcohol at the current time. The current state estimation unit 170E outputs a feature quantity of the drinking state at the current time using a reference value for a time period determined based on current information at the current time and current information at the current time. The feature quantity of the drinking state may be a single feature quantity that combines each of the feature quantities indicated in the information included in the current information, or may be composed of multiple types of feature quantities including each of the feature quantities indicated in the information included in the current information. The current state estimation unit 170E calculates, as the feature quantity, a value corresponding to the difference between the reference value for the time period determined based on current information at the current time and the information included in the current information at the current time. Specifically, for example, the feature quantity is calculated as a value corresponding to the difference between the pulse rate, which is a reference value for a time period determined based on current information at the current time, and the pulse rate, which is biometric data included in the current information at the current time.

[0119] The drinking state determination unit 180E determines whether the observed subject is drinking alcohol based on the feature quantities. The drinking state determination unit 180E outputs a determination result of the drinking state of the observed subject based on the drinking state feature quantities output by the current state estimation unit 170E. For example, the drinking state determination unit 180E analyzes changes in the feature quantities and determines whether the observed subject is drinking alcohol based on the degree of change indicating a change trend. In this case, the drinking state determination unit 180E calculates a degree of change indicating the degree of change in the drinking state feature quantities output by the current state estimation unit 170E and determines the drinking state of the observed subject using the degree of change. For example, the drinking state determination unit 180E performs threshold processing using the degree of change and a change threshold to ultimately determine whether the observed subject is drinking alcohol or not. Alternatively, the drinking state determination unit 180E estimates a drinking score indicating the degree of drinking from the feature quantities and determines whether the observed subject is drinking alcohol using the drinking score. For a configuration example in this case, please refer to the modified example of the drinking state determination unit 180A already described ( FIG. 6 ). The drinking level determination unit 180A performs threshold processing using the drinking score and the drinking score threshold, for example, to ultimately determine whether the observed subject is drinking alcohol or not.

[0120] The processing of the drinking alcohol estimation device according to the fifth embodiment of the present disclosure differs in the number of types of information treated as current information, but is similar to the processing example already described, and therefore a detailed description of the processing example will be omitted here.

[0121] In this embodiment, the current information for estimating the drinking status of the observed subject includes observed subject information (mobile body occupant information, driver information), mobile body information (vehicle information), and mobile body surroundings information (vehicle surroundings information). The observed subject information includes alertness, pulse rate, facial direction, gaze, pupil size, respiratory rate, and reaction time to the surroundings. The vehicle information includes vehicle speed, steering angle, accelerator force, and braking. For example, the observed pulse, alertness, reaction time, and vehicle information tend to differ depending on the drinking status as follows: Pulse rate: low → high Alertness: high → low Reaction time to the surroundings: short → long Vehicle information (roughness of steering operation): low → high Vehicle information (variability in speed change): small → large Based on this concept, the reference value or feature value at the current time can be calculated by quantifying the changes from a non-drinking state (reference value) for the above multiple pieces of information included in the current information. The reference value or the feature value of the current time can be calculated for each of the multiple pieces of information included in the current information, or can be calculated as a single value by combining the multiple pieces of information included in the current information.

[0122] This embodiment further illustrates an example embodiment including the following configuration: The current information further includes mobile body information indicating an operation performed by an occupant of the mobile body on the mobile body or the movement of the mobile body due to the operation. This provides an advantage of providing a mobile body drinking estimation device that can comprehensively estimate drinking levels by taking into account the operation performed by the occupant of the mobile body on the mobile body or the state of the movement of the mobile body due to the operation, thereby enabling improved estimation of drinking levels of a mobile body occupant compared to conventional devices. Furthermore, the present disclosure achieves the same advantage as the above by applying the above configuration to a mobile body-mounted system or the above drinking estimation method.

[0123] This embodiment further illustrates an example embodiment including the following configuration: The moving body is a vehicle, and the moving body information is vehicle information based on at least one of a steering operation, a braking operation, or an accelerator operation by the vehicle driver, or vehicle motion associated with each operation. This provides an advantage of providing a drinking alcohol estimation device that can comprehensively estimate drinking alcohol consumption by taking into account the vehicle driver's operation on the vehicle or the state of vehicle motion due to the operation, thereby enabling improved estimation of the driver's drinking alcohol consumption compared to conventional devices. Furthermore, the present disclosure achieves the same advantage as the above by applying the above configuration to a moving body-mounted system or the above drinking alcohol estimation method.

[0124] This embodiment further illustrates an example embodiment including the following configuration: a drinking alcohol estimation device, wherein the current information further includes mobile-object surrounding information, which is information indicating the situation around the mobile object. This provides an advantage of providing a drinking alcohol estimation device that can comprehensively estimate drinking alcohol consumption by further taking into account the situation around the mobile object, thereby enabling improved estimation of drinking alcohol consumption by a mobile object occupant compared to conventional methods. Furthermore, the present disclosure provides the same advantage as the above by applying the above configuration to a mobile-object mounted system or the above drinking alcohol estimation method.

[0125] This embodiment further illustrates an example embodiment including the following configuration: The drinking alcohol estimation device according to claim 9, wherein the moving body is a vehicle, and the moving body surroundings information is vehicle surroundings information based on the type, location, or type and location of objects around the vehicle that the driver is likely to pay attention to. This provides an advantage of providing a drinking alcohol estimation device that can comprehensively estimate drinking alcohol consumption by further taking into account the situation around the vehicle, thereby enabling improved estimation of the driver's drinking alcohol consumption compared to conventional methods. Furthermore, the present disclosure provides the same advantage as the above by applying the above configuration to a moving body-mounted system or the above drinking alcohol estimation method.

[0126] Sixth Embodiment. In the above-described embodiments, the case where multiple subjects are observed has not been mentioned. However, as shown in FIGS. 10A and 10B in FIG. 10, the biometric data and characteristics of fluctuations may differ depending on the subject. In the sixth embodiment, a configuration example in which reference values ​​for each time period can be managed for each subject is described. In this embodiment, for each feature used in the determination, the characteristics of a driver in a sober state are recorded in advance, and changes in the characteristics due to drinking are captured, thereby absorbing individual differences. However, because there are individual differences in the expression of each feature, the absolute values ​​of the feature cannot be used directly in the calculation. Therefore, for each feature used in the determination, the characteristics of a driver in a sober state are recorded in a sober state feature storage unit in advance, and the feature detection unit captures changes in the features from a sober state, thereby achieving drinking detection that absorbs individual differences.

[0127] In embodiment 6, among the components of embodiment 6, duplicate explanations will be omitted as appropriate for components that are similar to the components of embodiment 1, embodiment 2, embodiment 3, embodiment 4, or embodiment 5 that have already been described.

[0128] A configuration example in which a drinking alcohol estimation device according to a sixth embodiment of the present disclosure is applied to a mobile-mounted system will be described. FIG. 17 is a diagram illustrating a configuration example of a mobile-mounted system 1 (1F) including a drinking alcohol estimation device 100 (100F) according to the sixth embodiment of the present disclosure. The mobile-mounted system 1 (1F) is configured by adding a personal authentication function and a configuration that enables the use of a reference value for each individual using the results of the personal authentication to the already-described mobile-mounted system 1 (). The mobile-mounted system 1 (1F) shown in FIG. 17 is configured to include the drinking alcohol estimation device 100 (100F), a current information output device 300, and an external processing device (output device) 500. The external processing device 500 has a configuration similar to the already-described external processing device 500. Further detailed description of the external processing device 500 will be omitted.

[0129] The current information output device 300 outputs personal identification information for identifying each observed subject in addition to the functions of the current information output device 300 already described. The current information output device 300 shown in Fig. 17 includes a biological data output device 310, a mobile object information output device (vehicle information output device) 320, and a mobile object surrounding information output device (vehicle surrounding information output device) 330.

[0130] The biometric data output device 310 is configured by, for example, a DMS, and has a function of performing personal authentication for the occupant of the vehicle in addition to the functions of the biometric data output device 310 already described. The biometric data output device 310 outputs personal identification information for identifying the subject of observation as a result of the personal authentication.

[0131] The moving body information output device (vehicle information output device) 320 has the same configuration as the moving body information output device (vehicle information output device) 320 already described.

[0132] The moving object surrounding information output device (vehicle surrounding information output device) 330 has the same configuration as the moving object surrounding information output device (vehicle surrounding information output device) 330 already described.

[0133] The drinking alcohol estimation device 100 (100F) has the function of managing individual standards for each observed subject in addition to the functions of the drinking alcohol estimation device 100 in the embodiment already described. The drinking alcohol estimation device 100 (100F) shown in FIG. 17 includes a current information acquisition unit 110F, a current information accumulation unit 120F, a standard output unit 130F, a standard value calculation unit 131F, a standard value storage unit 132F, a standard value extraction unit 133F, a current state estimation unit 170F, a drinking state determination unit 180F, and an individual standard management unit 190 (190F). The following will mainly describe new features of this embodiment, and will omit descriptions of features already described.

[0134] The individual reference management unit 190F manages individual references for each observed subject. The individual reference management unit 190F acquires personal identification information that identifies the observed subject, and if a reference value corresponding to the personal identification information is stored, commands the reference output unit 130F to output the stored reference value. The individual reference management unit 190F acquires personal identification information that identifies the observed subject, and if it is necessary to register a reference value corresponding to the personal identification information, commands the reference output unit 130F to associate the personal identification information with a reference value for each time period and store them.

[0135] In addition to the functions of the reference value storage unit 132 already described, the reference value storage unit 132F in the reference output unit 130F stores personal identification information and a reference value in association with each other in response to a command from the individual reference management unit 190F. Furthermore, the reference value storage unit 132F outputs the reference value associated with the personal identification information in response to a request from the reference value extraction unit 133F.

[0136] In addition to the functions of the reference value extraction unit 133 already described, the reference value extraction unit 133F in the reference output unit 130F requests the reference value associated with personal identification information from the reference value storage unit 132F in response to an instruction from the individual reference management unit 190F, and acquires the reference value output by the reference value storage unit 132F.

[0137] Next, a processing example of the drinking alcohol estimation device according to the sixth embodiment of the present disclosure will be described. FIG. 18 is a flowchart showing an example of processing by the drinking alcohol estimation device 100F according to the sixth embodiment of the present disclosure. The processing shown in FIG. 18 is part of the drinking alcohol estimation method performed by the drinking alcohol estimation device 100F. For example, the drinking alcohol estimation device 100F shown in FIG. 17 starts the processing shown in FIG. 18 when a drinking alcohol estimation start condition is satisfied, such as when the observed subject boards a moving body and personal authentication is performed, or when the drinking alcohol estimation process by the drinking alcohol estimation device 100F is started. (Start)

[0138] The drinking alcohol estimation device 100F then executes a personal identification information acquisition process (step ST6610). In the personal identification information acquisition process, the individual reference management unit 190F of the drinking alcohol estimation device 100F acquires personal identification information that identifies the observation subject.

[0139] The drinking estimation device 100F then executes an individual criterion presence / absence determination process (step ST6620, "Is an individual criterion present?"). In the individual criterion presence / absence determination process, the individual criterion management unit 190F of the drinking estimation device 100F uses the personal identification information to refer to the reference value storage unit 132F and determines whether a reference value corresponding to the personal identification information needs to be stored or updated. For example, the individual criterion management unit 190F determines whether a reference value corresponding to the personal identification information needs to be stored by determining whether the reference value corresponding to the personal identification information is stored in the reference value storage unit 132F. Alternatively, for example, the individual criterion management unit 190F determines whether a reference value needs to be updated by referring to the date and time when the reference value corresponding to the personal identification information was stored.

[0140] When the drinking estimation device 100F determines that there is no individual criterion in the individual criterion presence / absence determination process (step ST6620, "NO"), it then executes an individual criterion storage command process (step ST6630). In the individual criterion storage command process, the individual criterion management unit 190F of the drinking estimation device 100F commands the reference value storage unit 132F to store the reference value calculated by the reference value calculation unit 131F in association with the personal identification information.

[0141] When the drinking estimation device 100F determines that an individual criterion is present in the individual criterion presence / absence determination process ("YES" in step ST6620), the drinking estimation device 100F then executes an individual criterion use command process (step ST6640). In the individual criterion use command process, the individual criterion management unit 190F of the drinking estimation device 100F commands the reference value extraction unit 133F to acquire the reference value associated with the personal identification information and stored in the reference value storage unit 132F.

[0142] After issuing a command to the reference value storage unit 132F or the reference value extraction unit 133F in the individual reference value storage command process, the drinking alcohol estimation device 100F ends the process shown in FIG. 18 and waits (END).

[0143] This embodiment further illustrates an example embodiment including the following configuration: The drinking alcohol estimation device according to any one of claims 1 to 4 further includes an individual reference management unit (190F) that acquires personal identification information identifying the observed subject and, if a reference value corresponding to the personal identification information is stored, commands the reference output unit to output the stored reference value. This provides an advantage of providing a drinking alcohol estimation device that can perform drinking alcohol estimation using the reference value stored for each individual, thereby enabling improved drinking alcohol estimation for each observed subject compared to conventional methods. Furthermore, the present disclosure provides the same advantage as the above by applying the above configuration to a mobile-mounted system or the drinking alcohol estimation method.

[0144] This embodiment further illustrates an example embodiment including the following configuration: The drinking alcohol estimation device according to claim 11, wherein the individual reference management unit (190F) acquires personal identification information identifying the observed subject, and, if registration of a reference value corresponding to the personal identification information is necessary, commands the reference output unit to associate the personal identification information with a reference value for each time period and store the associated information. This provides an advantage of providing a drinking alcohol estimation device that can register or update a reference value for each observed subject as needed, thereby enabling improved drinking alcohol estimation for each observed subject compared to conventional devices. Furthermore, the present disclosure provides the same advantage as described above by applying the above configuration to a mobile-mounted system or the drinking alcohol estimation method.

[0145] Here, a hardware configuration for realizing the functions of the present disclosure will be described. Fig. 19 is a diagram showing a first example of a hardware configuration for realizing the functions of the configuration of the present disclosure. Fig. 20 is a diagram showing a second example of a hardware configuration for realizing the functions of the configuration of the present disclosure. The drinking estimation devices 100, 100A, 100B, 100C, 100D, 100E, and 100F of the present disclosure are each realized by the hardware shown in Fig. 19 or Fig. 20.

[0146] 19 , each of the drinking alcohol estimation devices 100, 100A, 100B, 100C, 100D, 100E, and 100F includes, for example, a processor 10001, a memory 10002, an input / output interface 10003, and a communication circuit 10004. The processor 10001 and the memory 10002 are installed in, for example, a computer. The memory 10002 stores the computer, current information acquisition units 110, 110A, 110B, 110C, 110D, 110E, and 110F, current information accumulation units 120, 120A, 120B, 120C, 120D, 120E, and 120F, reference output units 130, 130A, 130B, 130C, 130D, 130E, and 130F, reference value calculation units 131 (131A, 131B, 131C, 131D, 131E, and 131F), reference value storage units 132 (132A, 132B, 132C, 132D, 132E, 132F), a reference value extraction unit 133 (133A, 133B, 133C, 133D, 133E, 133F), a reference value correction unit 134 (134D), a current state estimation unit 170, 170A, 170B, 170C, 170D, 170E, 170F, a drinking state determination unit 180, 180A, 180B, 180C, 180D, 180E, 180F, a drinking degree estimation unit 181, an individual reference management unit 190 (190F), and a program for functioning as a control unit (not shown) are stored.The processor 10001 reads out and executes the program stored in the memory 10002, whereby the current information acquisition units 110, 110A, 110B, 110C, 110D, 110E, 110F, the current information accumulation units 120, 120A, 120B, 120C, 120D, 120E, 120F, the reference output units 130, 130A, 130B, 130C, 130D, 130E, 130F, the reference value calculation units 131 (131A, 131B, 131C, 131D, 131E, 131F), the reference value storage unit 132 ( 132A, 132B, 132C, 132D, 132E, 132F), reference value extraction unit 133 (133A, 133B, 133C, 133D, 133E, 133F), reference value correction unit 134 (134D), current state estimation unit 170, 170A, 170B, 170C, 170D, 170E, 170F, drinking state determination unit 180, 180A, 180B, 180C, 180D, 180E, 180F, drinking level estimation unit 181, individual reference management unit 190 (190F), and a control unit (not shown) are realized. Furthermore, a storage unit (not shown) is realized by memory 10002 or another memory (not shown). Furthermore, a communication unit (not shown) is realized by communication circuit 10004.

[0147] The processor 10001 is, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a microprocessor, a microcontroller, or a DSP (Digital Signal Processor). The memory 10002 may be a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable Read Only Memory), or flash memory, or a magnetic disk such as a hard disk or flexible disk, or an optical disk such as a CD (Compact Disc) or DVD (Digital Versatile Disc), or a magneto-optical disk. The processor 10001 and the memory 10002 or the communication circuit 10004 are connected in a state capable of transmitting data to each other. The processor 10001, memory 10002, and communication circuit 10004 are connected via an input / output interface 10003 so as to be capable of transmitting data to and from other hardware.

[0148] Alternatively, in the drinking alcohol estimation devices 100, 100A, 100B, 100C, 100D, 100E, 100F, the current information acquisition units 110, 110A, 110B, 110C, 110D, 110E, 110F, the current information accumulation units 120, 120A, 120B, 120C, 120D, 120E, 120F, the reference output units 130, 130A, 130B, 130C, 130D, 130E, 130F, the reference value calculation units 131 (131A, 131B, 131C, 131D, 131E, 131F), the reference value storage units 132 (132A, 132B, 132C, 132D, 132E, 132F), 32D, 132E, 132F), reference value extraction unit 133 (133A, 133B, 133C, 133D, 133E, 133F), reference value correction unit 134 (134D), current state estimation units 170, 170A, 170B, 170C, 170D, 170E, 170F, drinking state determination units 180, 180A, 180B, 180C, 180D, 180E, 180F, drinking level estimation unit 181, individual reference management unit 190 (190F), and the functions of a control unit (not shown) may be realized by a dedicated processing circuit 20001, as shown in FIG. 20 .

[0149] The processing circuit 20001 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), an FPGA (Field-Programmable Gate Array), a SoC (System-on-a-Chip), or a system LSI (Large-Scale Integration), etc. In addition, a storage unit (not shown) is realized by the memory 20002 or another memory (not shown). The memory 20002 may be a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable Read Only Memory), or flash memory, or a magnetic disk such as a hard disk or flexible disk, or an optical disk such as a CD (Compact Disc) or DVD (Digital Versatile Disc), or a magneto-optical disk. The communication circuit 20004 implements a communication unit (not shown). The processing circuit 20001 and the memory 20002 or the communication circuit 20004 are connected in a state where they can transmit data to each other. In addition, the processing circuit 20001, the memory 20002, and the communication circuit 20004 are connected in a state where they can transmit data to other hardware via the input / output interface 20003.In addition, in the drinking alcohol estimation devices 100, 100A, 100B, 100C, 100D, 100E, 100F, the current information acquisition units 110, 110A, 110B, 110C, 110D, 110E, 110F, the current information accumulation units 120, 120A, 120B, 120C, 120D, 120E, 120F, the reference output units 130, 130A, 130B, 130C, 130D, 130E, 130F, the reference value calculation units 131 (131A, 131B, 131C, 131D, 131E, 131F), the reference value storage units 132 (132A, 132B, 132C, The functions of the reference value extraction unit 133 (133A, 133B, 133C, 133D, 133E, 133F), reference value correction unit 134 (134D), current state estimation unit 170, 170A, 170B, 170C, 170D, 170E, 170F, drinking state determination unit 180, 180A, 180B, 180C, 180D, 180E, 180F, drinking level estimation unit 181, individual reference management unit 190 (190F), and a control unit (not shown) may be realized by separate processing circuits, or may be realized together in a processing circuit. Similarly, the functions of the video acquisition unit 301A, operation determination unit 302A, and a control unit (not shown) in the occupant monitoring device 300A may be realized by separate processing circuits, or may be realized together in a processing circuit. Similarly, the functions of the operation information collecting unit 601E and the control unit (not shown) in the server device 600E may be realized by separate processing circuits, or may be realized together by a processing circuit.

[0150] Or, in the drinking estimation devices 100, 100A, 100B, 100C, 100D, 100E, 100F, the current information acquisition units 110, 110A, 110B, 110C, 110D, 110E, 110F, the current information accumulation units 120, 120A, 120B, 120C, 120D, 120E, 120F, the reference output units 130, 130A, 130B, 130C, 130D, 130E, 130F, the reference value calculation units 131 (131A, 131B, 131C, 131D, 131E, 131F), the reference value storage units 132 (132A, 132B, 132C, 132D, 132E, 132F), Some of the functions of the reference value extraction unit 133 (133A, 133B, 133C, 133D, 133E, 133F), reference value correction unit 134 (134D), current state estimation unit 170, 170A, 170B, 170C, 170D, 170E, 170F, drinking state determination unit 180, 180A, 180B, 180C, 180D, 180E, 180F, drinking level estimation unit 181, individual reference management unit 190 (190F), and a control unit (not shown) may be realized by the processor 10001 and memory 10002, and the remaining functions may be realized by the processing circuit 20001.

[0151] It should be noted that, within the scope of this disclosure, the embodiments may be freely combined, any component of each embodiment may be modified, or any component of each embodiment may be omitted.

[0152] The present disclosure can improve the accuracy of alcohol consumption estimation compared to conventional methods, and is therefore suitable for use in, for example, an alcohol consumption estimation device that estimates the alcohol consumption of occupants in a moving body, or an alcohol consumption estimation device that estimates the alcohol consumption of a driver in a vehicle.

[0153] 1 (1A, 1B, 1C, 1D, 1E, 1F) Mobile mounted system, 100, 100A, 100B, 100C, 100D, 100E, 100F Drinking estimation device, 110, 110A, 110B, 110C, 110D, 110E, 110F Current information acquisition unit, 120, 120A, 120B, 120C, 120D, 120E, 120F Current information storage unit, 130, 130A, 130B, 130C, 130D, 130E, 130F Reference output unit, 131 (131A, 131B, 131C, 131D, 131E, 131F) Reference value calculation unit, 132 (132A, 132B, 132C, 132D, 132E, 132F) Reference value storage unit, 133 (133A, 133B, 133C, 133D, 133E, 133F) Reference value extraction unit, 134 (134D) Reference value correction unit, 170, 170A, 170B, 170C, 170D, 170E, 170F Current state estimation unit, 180, 180A, 180B, 180C, 180D, 180E, 180F Drinking state determination unit, 181 Drinking level estimation unit, 190 (190F) Individual reference management unit, 300 Current information output device, 310 Biometric data output device, 320 Mobile information output device (vehicle information output device), 330 Mobile object surrounding information output device (vehicle surrounding information output device), 500 external processing device (output device), 2000 heart rate (example of change in heart rate throughout the day), 2010 heart rate (example of change in heart rate of person A in the morning), 2020 heart rate (example of change in heart rate of person A in the afternoon), 2030 heart rate (example of change in heart rate of person B in the morning), 2040 heart rate (example of change in heart rate of person B in the afternoon), 3010 heart rate (example of change in heart rate when driving after exercise), 3020 heart rate (example of change in heart rate when driving normally), 4010 heart rate (example of change in heart rate due to exercise in a first time period), 4020 heart rate (example of change in heart rate due to exercise in a second time period), 10001 processor, 10002 memory, 10003 input / output interface, 10004 communication circuit, 20001 Processing circuit, 20002 memory, 20003 input / output interface, 20004 communication circuit.

Claims

1. A drinking estimation device comprising: a current information acquisition unit that acquires current information including biometric data of the observed subject in chronological order; a reference output unit that uses the current information accumulated in chronological order and the current information at the current time to output a reference value indicating a feature amount for when the observed subject would be in a non-drinking state during a time period determined based on the current information at the current time; a current state estimation unit that uses the reference value for a time period determined based on the current information at the current time and the current information at the current time to output a feature amount for the drinking state at the current time; and a drinking state determination unit that outputs a determination result for the drinking state of the observed subject based on the feature amount for the drinking state output by the current state estimation unit.

2. The drinking estimation device according to claim 1, characterized in that the reference output unit calculates a reference value for each time period in a day using current information accumulated in chronological order, and outputs a reference value for a time period including the current time using current information at the current time.

3. The drinking estimation device according to claim 1, characterized in that the reference output unit calculates a reference value for a time period when the observed subject is exercising using current information accumulated in chronological order, and outputs the reference value for a time period when the observed subject is exercising using current information at the current time when the observed subject is exercising.

4. The drinking estimation device according to claim 1, characterized in that the reference output unit uses current information accumulated in chronological order and current information at the current time to output reference values ​​for each time period in a day when the observed subject is exercising.

5. The drinking alcohol estimation device according to any one of claims 1 to 4, wherein the observed person is a passenger of a moving body.

6. The drinking estimation device according to any one of claims 1 to 4, characterized in that the observed person is a driver of a vehicle, and the biometric data is the pulse rate of the observed person measured using changes in brightness values ​​of an image of the driver, the pulse rate of the driver measured by a seat sensor, or the pulse rate of the driver measured by a millimeter-wave radar.

7. The drinking level estimation device according to claim 5, characterized in that the current information further includes mobile body information that indicates operations performed on the mobile body by an occupant of the mobile body or the movement of the mobile body due to such operations.

8. The drinking estimation device according to claim 7, characterized in that the moving body is a vehicle, and the moving body information is vehicle information based on at least one of the steering operation, braking operation, or accelerator operation by the vehicle driver, or vehicle movement associated with each operation.

9. The drinking level estimation device according to claim 7, wherein the current information further includes mobile object surrounding information, which is information indicating the situation around the mobile object.

10. The drinking estimation device of claim 9, wherein the moving body is a vehicle, and the moving body surroundings information is vehicle surroundings information based on the type, location, or type and location of objects around the vehicle that the driver may be likely to pay attention to.

11. The drinking estimation device according to any one of claims 1 to 4, further comprising an individual standard management unit that acquires personal identification information that identifies the observed subject, and, if a standard value corresponding to the personal identification information is stored, commands the standard output unit to output the stored standard value.

12. The drinking estimation device described in claim 11, characterized in that the individual standard management unit acquires personal identification information that identifies the observed subject, and if it is necessary to register a standard value corresponding to the personal identification information, instructs the standard output unit to associate the personal identification information with a standard value for each time period and store it.

13. The drinking state estimation device according to any one of claims 1 to 4, characterized in that the drinking state determination unit calculates a degree of change indicating the degree of change in the drinking state feature output by the current state estimation unit, and determines the drinking state of the observed subject using the degree of change.

14. The drinking state estimation device described in any one of claims 1 to 4, characterized in that the drinking state determination unit estimates a drinking score indicating the degree of drinking based on the drinking state feature output by the current state estimation unit, and determines the drinking state of the observed subject using the drinking score.

15. The drinking state determination unit of the drinking estimation device described in claim 14, characterized in that it acquires the level of alertness of the observed subject, inputs the level of alertness and the feature of the drinking state output by the current state estimation unit into a prediction model to acquire the drinking score output by the prediction model, and determines the drinking state of the observed subject using the acquired drinking score.

16. A drinking alcohol estimation method using an alcohol consumption estimation device, comprising: a current information acquisition step in which a current information acquisition unit of the drinking alcohol estimation device acquires current information including biological data of the observed subject in chronological order; a reference output step in which a reference output unit of the drinking alcohol estimation device uses the current information accumulated in chronological order and the current information at the current time to output a reference value indicating a feature amount when the observed subject would be in a non-drinking state during a time period determined based on the current information at the current time; a current state estimation step in which a current state estimation unit of the drinking alcohol estimation device outputs a feature amount of the drinking state at the current time using the reference value for the time period determined based on the current information at the current time and the current information at the current time; and a drinking state determination step in which a drinking state determination unit of the drinking alcohol estimation device outputs a determination result of the drinking state of the observed subject based on the drinking state feature amount output by the current state estimation unit.

Images