Data generation device, data generation method, and program
The data generation device addresses noise in measurement data by using multiple sensors and accuracy calculation to generate precise determination data, enabling accurate object state assessment despite noise.
Patent Information
- Application Number
- JP2021059698
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-03-31
- Publication Date
- 2024-12-09
- Estimated Expiration
- 2041-03-31
AI Technical Summary
Measurement data often includes noise due to external factors such as vibration and external light, making it difficult to accurately determine the state of an object, and existing technologies do not provide effective methods to handle noise in measurement data.
A data generation device that includes a first sensor for acquiring measurement data, a second sensor for measuring noise factors, an accuracy calculation unit to determine measurement accuracy, and a data output unit to provide accuracy information, allowing for the generation of determination data that accurately reflects the state of the object despite noise.
The device can generate data that accurately determines the state of an object even when measurement data contains noise, ensuring precise analysis by incorporating noise factor data into the determination process.
Smart Images

Figure 0007596020000001 
Figure 0007596020000002 
Figure 0007596020000003
Abstract
Description
[Technical field]
[0001] The present disclosure relates to a data generation device, a data generation method, and a program. [Background technology]
[0002] In recent years, devices have been known that measure an object in a non-contact manner to obtain measurement data of the object. For example, Patent Document 1 discloses a device that obtains measurement data by measuring the brain waves of a subject in a non-contact manner by irradiating the subject with infrared rays and photographing the subject irradiated with the infrared rays with a camera. Such measurement data is used, for example, to determine the state of the object. [Prior art documents] [Patent documents]
[0003] [Patent Document 1] International Publication No. 2020 / 203914 Summary of the Invention [Problem to be solved by the invention]
[0004] However, measurement data may contain noise due to external factors such as vibration of the device and external light. If noise is included in the measurement data, it may be difficult to accurately determine the state of the target object. Therefore, technology to remove noise from the measurement data has been studied, but removal may be difficult depending on the noise included.
[0005] Therefore, it is sometimes desirable to accurately determine the state of an object even if the measurement data contains noise. However, Patent Document 1 does not disclose how to determine the state of an object when the measurement data contains noise.
[0006] Therefore, the present disclosure provides a data generating device, a data generating method, and a program capable of generating data that allows accurate determination of the state of an object even if the measurement data contains noise. [Means for solving the problem]
[0007] A data generating device according to one embodiment of the present disclosure is a data generating device that generates judgment data for judging a state of an object, and includes a first acquisition unit that acquires measurement data of the object from a first sensor, a second acquisition unit that acquires first factor data from a second sensor different from the first sensor, in which a first noise factor that may become noise in the measurement data is measured, an accuracy calculation unit that calculates the measurement accuracy of the measurement data based on the first factor data, and a data output unit that outputs the judgment data in which the measurement data is associated with accuracy information indicating the measurement accuracy.
[0008] A data generation method according to one aspect of the present disclosure is a data generation method for generating judgment data for judging a state of an object, which method comprises the steps of: obtaining measurement data of the object from a first sensor; obtaining first factor data measuring a first noise factor that may become noise in the measurement data from a second sensor different from the first sensor; calculating a measurement accuracy of the measurement data based on the first factor data; and outputting the judgment data in which the measurement data is associated with accuracy information indicating the measurement accuracy.
[0009] A program according to one aspect of the present disclosure is a program for causing a computer to execute the above-described data generation method. Effect of the Invention
[0010] According to one aspect of the present disclosure, it is possible to realize a data generating device or the like that can generate data that allows accurate determination of the state of an object even if measurement data contains noise. [Brief description of the drawings]
[0011] [Figure 1] FIG. 1 is a diagram showing a configuration of a data generation system according to the first embodiment. [Diagram 2] FIG. 2 is a block diagram illustrating a functional configuration of the information processing device according to the first embodiment. [Diagram 3] FIG. 3 is a flowchart showing the operation of the data generating system according to the first embodiment. [Figure 4] FIG. 4 is a diagram showing determination data and noise levels according to the first embodiment. [Diagram 5] FIG. 5 is a diagram showing a first conversion table between external factor data and noise levels according to the first embodiment. [Figure 6] FIG. 6 is a flowchart showing the operation of correcting the delay due to time correlation shown in FIG. [Figure 7] FIG. 7 is a diagram showing the relationship between ΔT and the correlation coefficient according to the first embodiment. [Figure 8] FIG. 8 is a diagram showing a second conversion table between noise level and accuracy according to the first embodiment. [Figure 9] FIG. 9 is a diagram showing data with accuracy information according to the first embodiment. [Figure 10] FIG. 10 is a flowchart showing the operation of the determination process shown in FIG. [Figure 11] FIG. 11 is a diagram showing data for the first determination unit according to the first embodiment. [Figure 12] FIG. 12 is a diagram showing data for the second determination unit according to the first embodiment. [Figure 13] FIG. 13 is a block diagram illustrating a functional configuration of the information processing device according to the second embodiment. As shown in FIG. [Figure 14] FIG. 14 is a flowchart showing the operation of the data generating system according to the second embodiment. [Figure 15] FIG. 15 is a diagram showing a third conversion table between the first external factor data and the noise level according to the second embodiment. [Figure 16]FIG. 16 is a diagram showing a fourth conversion table between the second external factor data and the noise level according to the second embodiment. [Figure 17] FIG. 17 is a flowchart showing the operation of correcting the delay due to time correlation shown in FIG. [Figure 18] FIG. 18 is a diagram showing a fifth conversion table between noise level and accuracy according to the second embodiment. [Figure 19] FIG. 19 is a diagram illustrating a configuration of a data generation system according to the third embodiment. [Figure 20] FIG. 20 is a block diagram showing functional configurations of an information processing device and a server device according to the third embodiment. [Figure 21] FIG. 21 is a flowchart showing the operation of the data generating system according to the third embodiment. [Figure 22] FIG. 22 is a flowchart showing the operation of the personal authentication shown in FIG. [Figure 23] FIG. 23 is a flowchart showing the operation of the server device according to the third embodiment. [Figure 24] FIG. 24 is a flowchart showing the operation of calculating the noise level of the external factor shown in FIG. [Diagram 25] FIG. 25 is a flowchart showing the operation of calculating the accuracy shown in FIG. [Figure 26] FIG. 26 is a flowchart showing the operation of the determination process shown in FIG. [Figure 27] FIG. 27 is a block diagram showing functional configurations of an information processing device and a server device according to the fourth embodiment. As shown in FIG. [Figure 28] FIG. 28 is a flowchart showing the operation of the information processing device according to the fourth embodiment. [Figure 29] FIG. 29 is a flowchart showing the operation of the server device according to the fourth embodiment. [Diagram 30] FIG. 30 is a diagram illustrating a configuration of a data generation system according to the fifth embodiment. [Diagram 31]FIG. 31 is a flowchart showing the operation of the data generation system according to the fifth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0012] A data generating device according to one embodiment of the present disclosure is a data generating device that generates judgment data for judging a state of an object, and includes a first acquisition unit that acquires measurement data of the object from a first sensor, a second acquisition unit that acquires first factor data from a second sensor different from the first sensor, in which a first noise factor that may become noise in the measurement data is measured, an accuracy calculation unit that calculates the measurement accuracy of the measurement data based on the first factor data, and a data output unit that outputs the judgment data in which the measurement data is associated with accuracy information indicating the measurement accuracy.
[0013] As a result, the measurement accuracy based on the first factor data in which a first noise factor that may become noise in the measurement data is measured is added to the judgment data. For example, the measurement data is used to judge the object according to the measurement accuracy, so that the judgment can be performed accurately. Therefore, the data generating device can generate data that can accurately judge the state of the object even if the measurement data contains noise.
[0014] Also, for example, the measurement data and the first factor data may be time-series data measured over a first period, and the device may further include a delay correction unit that corrects a time delay between the measurement data and the first factor data, and the data output unit may output the judgment data in which the measurement data and the accuracy information have been associated with each other and the time delay has been corrected.
[0015] This allows accurate temporal correspondence between the measurement data and the first factor data. For example, even if noise changes over time, accurate temporal correspondence between the measurement data and the measurement accuracy can be achieved, so that even if the measurement data contains noise, data that can accurately determine the state of the object can be generated. Therefore, the data generating device can generate determination data that can more accurately determine the state of the object even if the measurement data contains noise.
[0016] In addition, for example, the measurement device may further include a third acquisition unit that acquires second factor data, in which a second noise factor that may become noise in the measurement data, from a third sensor different from the first sensor and the second sensor, and the calculation unit may further calculate the measurement accuracy based on the second factor data.
[0017] This allows the measurement accuracy to be calculated based on two types of external factors, so that the measurement accuracy can be calculated more accurately. For example, when there are multiple noise factors that can cause noise in the measurement data, it is possible to prevent the accuracy of the calculation of the measurement accuracy from decreasing.
[0018] Also, for example, the second factor data may be time-series data measured over a second period at least a portion of which overlaps with the first period, and the delay correction unit may further correct a time delay between the measurement data and the second factor data.
[0019] This allows accurate temporal correspondence between the measurement data and the second factor data. For example, even if noise changes over time, accurate temporal correspondence between the measurement data and the measurement accuracy can be achieved, so that even if the measurement data contains noise, data that can accurately determine the state of the object can be generated. Therefore, the data generating device can generate determination data that can more accurately determine the state of the object even if the measurement data contains noise.
[0020] Also, for example, the device may further include a noise level calculation unit that calculates the noise level from the first factor data acquired by the second acquisition unit based on a first table in which the first factor data and the noise level that the first factor data imparts to the measurement data are associated, and the accuracy calculation unit may calculate the measurement accuracy of the measurement data acquired by the first acquisition unit from the noise level calculated by the noise level calculation unit using a second table in which the noise level and the measurement accuracy are associated.
[0021] This makes it possible to easily calculate the measurement accuracy by conversion using the first table and the second table.
[0022] Also, for example, the information processing device may further include a first update unit that updates at least one of the first table and the second table.
[0023] As a result, at least one of the first table and the second table is updated, so that the measurement accuracy can be calculated more accurately from the noise.
[0024] Also, for example, the device may further include a determination unit that determines the state of the object based on the determination data output from the data output unit.
[0025] This allows the data generating device to consistently perform processing up to the state of the target object.
[0026] Also, for example, the object may be a person, and the device may further include an authentication result acquisition unit that acquires an authentication result of the person.
[0027] As a result, the determination data is generated by processing according to the authentication result, so that it is possible to generate determination data that can more accurately determine the state of the object.
[0028] Also, for example, the authentication device may further include a second update unit configured to update at least one of a determination method of the determination unit and a threshold value used for the determination, the method being determined based on the authentication result.
[0029] As a result, at least one of the determination method and the threshold value used in the determination by the determination unit is updated according to the authentication result, so that a determination appropriate to the person can be made.
[0030] Also, for example, the judgment unit may have a first judgment unit that judges a first state of the object and a second judgment unit that judges a second state different from the first state, and the data output unit may output the same judgment data to the first judgment unit and the second judgment unit.
[0031] This makes it possible to determine multiple states of an object from a single determination data.
[0032] Also, for example, the first judgment unit may extract a first portion from the measurement data that satisfies the first accuracy based on a first accuracy corresponding to the judgment of the first state, and judge the first state based on the extracted first portion, and the second judgment unit may extract a second portion from the measurement data that satisfies the second accuracy based on a second accuracy corresponding to the judgment of the second state, and judge the second state based on the extracted second portion.
[0033] This makes it possible to accurately determine whether the first state or the second state is present.
[0034] Also, for example, the second state may be a state in which the danger to the object is less than that of the first state, and the device may further include a first notification unit that makes an emergency call when the first determination unit determines that the first state is abnormal, and a second notification unit that outputs a sound when the second determination unit determines that the second state is abnormal.
[0035] This allows the notification mode to be changed depending on the determination result, making it possible to perform notification according to the determination result.
[0036] Also, for example, the data generating device may be mounted in a vehicle, and the target object may be a driver of the vehicle.
[0037] This makes it possible to generate judgment data that can accurately judge the state of the object even in a vehicle with a lot of noise that reduces the accuracy of the measurement data.
[0038] Also, for example, the data generation device may be mounted on a vehicle, the object may be a driver of the vehicle, the first judgment unit may determine whether or not the driver has difficulty driving the vehicle, and the second judgment unit may determine whether or not the driver is dozing off or looking away.
[0039] This makes it possible to accurately determine whether the driver is having difficulty driving the vehicle and whether the driver is dozing off or looking away.
[0040] Also, for example, the data generation device may be mounted on a vehicle, the target object may be a driver of the vehicle, the first judgment unit may judge the driver's drowsiness, and the second judgment unit may judge the driver's emotions or whether the driver is ill.
[0041] This allows an accurate determination of the driver's drowsiness and the driver's emotions or whether the driver is ill.
[0042] Also, for example, the first sensor may be a millimeter wave sensor or an infrared camera, and the second sensor may be an acceleration sensor that measures the acceleration of the vehicle.
[0043] This makes it possible to use acceleration, which is noise that can be an external factor for a millimeter wave sensor or infrared camera, to generate judgment data that can accurately determine the state of the target object even if the measurement data contains noise.
[0044] Also, for example, the data generation device may be mounted on a vehicle, the first sensor may be an infrared camera, the second sensor may be an acceleration sensor that measures the acceleration of the vehicle, and the third sensor may be a luminance sensor that measures the luminance around the vehicle.
[0045] This makes it possible to generate data that can accurately determine the state of an object, even if the measurement data contains noise, by using acceleration and brightness, which can be external factors of the infrared camera.
[0046] Also, for example, information identifying the first noise factor may be acquired from the first sensor.
[0047] This allows the data generating device to have a simpler configuration since it does not need to store information that identifies the first noise factor.
[0048] A data generation method according to an embodiment of the present disclosure is a data generation method for generating judgment data for judging a state of an object, which includes acquiring measurement data of the object from a first sensor, acquiring first factor data of a first noise factor that may become noise in the measurement data from a second sensor different from the first sensor, calculating a measurement accuracy of the measurement data based on the first factor data, and outputting the judgment data in which the measurement data is associated with accuracy information indicating the measurement accuracy. A program according to an embodiment of the present disclosure is a program for causing a computer to execute the above data generation method.
[0049] This provides the same effects as the data generating device described above.
[0050] These general or specific aspects may be realized by a system, a method, an integrated circuit, a computer program, or a non-transitory recording medium such as a computer-readable CD-ROM, or may be realized by any combination of the system, the method, the integrated circuit, the computer program, or the recording medium. The program may be stored in the recording medium in advance, or may be supplied to the recording medium via a wide area communication network including the Internet.
[0051] Hereinafter, the embodiment will be specifically described with reference to the drawings.
[0052] The embodiments described below are all comprehensive or specific examples. The numerical values, shapes, components, the arrangement and connection of the components, steps, and the order of steps shown in the following embodiments are merely examples and are not intended to limit the present disclosure. For example, the numerical values are not expressions that express only the strict meaning, but expressions that include a substantially equivalent range, for example, a difference of about a few percent. Furthermore, among the components in the following embodiments, components that are not described in the independent claims are described as optional components.
[0053] In addition, each figure is a schematic diagram and is not necessarily illustrated precisely. Therefore, for example, the scales in each figure do not necessarily match. In addition, in each figure, substantially the same configurations are given the same reference numerals, and duplicated explanations are omitted or simplified.
[0054] Furthermore, in this specification, terms indicating relationships between elements, such as "identical," terms indicating the shape of an element, such as "rectangle," numerical values, and numerical ranges are not expressions that only express a strict meaning, but are expressions that also include a substantially equivalent range, for example, a difference of about a few percent (e.g., about 5%).
[0055] (Embodiment 1) Hereinafter, the data generating system according to the present embodiment will be described with reference to FIGS.
[0056] [1-1. Data generation system configuration] First, the configuration of a data generating system according to the present embodiment will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the functional configuration of a data generating system 1 according to the present embodiment.
[0057] In the following, an example will be described in which the object measurement sensor 100 is mounted on a vehicle, and the object to be measured by the object measurement sensor 100 is the driver of the vehicle. In the vehicle environment, there are many factors that cause noise to impair the accuracy of data acquired by the object measurement sensor 100, and even if a high-precision sensor is installed, it is difficult to acquire high-precision data. In particular, when the frequency band of the target data to be acquired includes the frequency band of noise, it is difficult to separate the noise from the data. In addition, the noise generated while the vehicle is running is generally larger than the noise generated while the vehicle is stopped. In other words, in a vehicle, the target data to be acquired may include noise that changes over time. With such target data, it is difficult to accurately determine the state of the object. Therefore, the data generation system 1 according to one embodiment of the present disclosure generates judgment data that can accurately determine the state of the object even from target data that includes noise that changes over time. Hereinafter, the judgment data is also referred to as data with accuracy information.
[0058] Specifically, the data generation system 1 is an information processing system that generates data with accuracy information that associates determination data based on object data obtained by measuring an object using the object measurement sensor 100 with the measurement accuracy of the determination data based on external factor data obtained by measuring the external factors of the object measurement sensor 100 using the external factor measurement sensor 200. The object data is an example of measurement data obtained by measuring an object using the object measurement sensor 100. Hereinafter, the measurement accuracy of the determination data, i.e., the measurement accuracy of the object data, will also be referred to simply as accuracy.
[0059] It should be noted that the object measuring sensor 100 is not limited to being mounted on a vehicle, and the object is not limited to being a driver.
[0060] As shown in FIG. 1, the data generating system 1 includes a sensor 100 for measuring an object, a sensor 200 for measuring an external factor, an information processing device 300, and a notification device 400.
[0061] The object measurement sensor 100 is mounted on a vehicle and measures the driver (the driver's state), which is an example of an object. The object measurement sensor 100 may be, for example, a sensor that is mounted on the vehicle in advance (for example, mounted when the vehicle is manufactured). The object measurement sensor 100 measures, for example, the driver's biological information. The object measurement sensor 100 measures the driver in a non-contact manner, but is not limited to this, and may measure the driver in a contact manner. The object measurement sensor 100 is an example of a first sensor, and is realized by, for example, a millimeter wave sensor, a camera, etc. The camera may be, for example, an infrared camera.
[0062] Furthermore, the object measurement sensor 100 may store information for identifying noise factors that may be included in the object data obtained by measuring the driver. The object measurement sensor 100 may output the information to the information processing device 300 when communication with the information processing device 300 is established. The information for identifying noise factors is information for identifying external factors that may become noise during measurement by the object measurement sensor 100, such as vehicle vibrations and external light. Note that the information for identifying noise factors may vary depending on the type of the object measurement sensor 100, the object on which the object measurement sensor 100 is mounted, and the like.
[0063] The external factor measuring sensor 200 measures an external factor that may become noise in the target data of the object measuring sensor 100. It can also be said that the external factor measuring sensor 200 measures an external factor that reduces the measurement accuracy when the object measuring sensor 100 measures an object. The external factor measuring sensor 200 may be, for example, a sensor that is pre-installed in the vehicle, or a sensor that is retrofitted to the vehicle. The external factor measuring sensor 200 measures the external factor in a non-contact manner, but is not limited thereto, and may measure the external factor by contacting an object that may become the external factor. The external factor is an example of a first noise factor.
[0064] The external factor measuring sensor 200 is an example of a second sensor, and is a sensor different from the object measurement sensor 100. The external factor measuring sensor 200 is realized by an acceleration sensor that measures the acceleration of the vehicle, a luminance sensor that measures the luminance around the vehicle, etc. The luminance sensor may measure, for example, the luminance around the object measurement sensor 100 (for example, the luminance of external light incident on the object measurement sensor 100). Note that external light is light other than the light emitted by the object measurement sensor 100.
[0065] For example, the external factor measurement sensor 200 does not measure an object. For example, the external factor data does not include data obtained by measuring an object. The external factor data measured by the external factor measurement sensor 200 is an example of first factor data.
[0066] The information processing device 300 generates judgment data for judging the state of an object. The information processing device 300 acquires object data from the object measurement sensor 100, acquires external factor data from the external factor measurement sensor 200, and generates judgment data based on the object data and the external factor data. For example, the information processing device 300 generates accuracy information for the object data obtained by measuring an object, taking into account the influence of noise due to external factors in the object data, and generates judgment data by performing a process of attaching the accuracy information to the object data. The information processing device 300 is mounted on a vehicle, for example, but may be located remotely from the vehicle.
[0067] The information processing device 300 may acquire information for identifying a noise factor from the target measurement sensor 100, or may store the information. The information processing device 300 is communicably connected to a plurality of sensors mounted on the vehicle, and may determine a sensor identified based on information for identifying a noise factor from the plurality of sensors as the external factor measurement sensor 200.
[0068] The notification device 400 notifies the user of the information processing result of the information processing device 300. The notification device 400 may output (present) the information processing result to the user by an image, a sound, or the like. The notification device 400 includes, for example, a display device, a sound output device, and the like. The user may be the driver, or may be a monitor who monitors (for example, remotely monitors) the driver.
[0069] Next, a detailed configuration of the information processing device 300 will be described with reference to Fig. 2. Fig. 2 is a block diagram showing the functional configuration of the information processing device 300 according to this embodiment.
[0070] As shown in FIG. 2, the information processing device 300 includes a precision-added information generating unit 310, a determining unit 370, and an output unit 380.
[0071] The precision-assigned information generating unit 310 generates judgment data based on the target data and the external factor data. The precision-assigned information generating unit 310 includes a first acquiring unit 320, an analyzing unit 330, a second acquiring unit 340, a precision information generating unit 350, and a precision information integrating unit 360. The precision-assigned information generating unit 310 is an example of a data generating device.
[0072] The first acquisition unit 320 acquires object data obtained by measuring an object from the object measurement sensor 100. The first acquisition unit 320 is configured to include, for example, a communication interface for communicating with the object measurement sensor 100. The object data is time-series data obtained by measuring an object over a first period of time, and includes time information indicating the time when the object was measured. The time information includes, for example, a timestamp. The timestamp is a record of the time when certain data was acquired.
[0073] The analysis unit 330 performs a process of converting the target data obtained by measuring the target object into a format used by the determination unit 370 for determination. The analysis unit 330 converts the target data into a feature quantity indicating the feature of the target object contained in the target data. For example, when the object measurement sensor 100 is a millimeter wave sensor or the like, the analysis unit 330 converts the target data measured by the millimeter wave sensor or the like into a heart rate (heart rate waveform), which is an example of a feature quantity. The feature quantity may be any of the driver's head position, posture, line of sight, blink interval, number of yawns, trunk behavior, steering time, and the like.
[0074] The analysis unit 330 may also have a filter (not shown) that removes a predetermined component from the converted feature amount. The filter is realized by, for example, a band-pass filter.
[0075] The second acquisition unit 340 acquires external factor data, in which external factors that may become noise in the target data are measured, from an external factor measurement sensor 200 different from the target measurement sensor 100. The second acquisition unit 340 is configured to include, for example, a communication interface for communicating with the external factor measurement sensor 200. Note that the external factor data is time-series data obtained by measuring the external factors over a first period of time, and includes time information indicating the time at which the external factors were measured.
[0076] The accuracy information generating unit 350 calculates the measurement accuracy of the target data acquired by the first acquiring unit 320, based on the external factor data acquired by the second acquiring unit 340. The accuracy information generating unit 350 has a noise information calculating unit 351, a time delay correcting unit 352, and an accuracy information calculating unit 353. The accuracy information generating unit 350 is an example of an accuracy calculating unit.
[0077] The noise information calculation unit 351 calculates the noise level included in the target data based on the external factor data. The noise information calculation unit 351 calculates the noise level from the external factor data acquired by the second acquisition unit 340 based on, for example, a table in which the external factor data and the noise level that the external factor data gives to the target data are associated (for example, see the first conversion table in FIG. 5). The calculated noise level indicates the influence of the external factor on the target data, and is time-series data over a first period. The noise information calculation unit 351 is an example of a noise level calculation unit. Moreover, the noise level is an example of noise information indicating the degree of influence of noise on the target data. Moreover, the time-series data of the noise level is also referred to as noise level data.
[0078] The time delay correction unit 352 corrects the time delay between the target data and the external factor data. Even if the target measurement data and the external factor data are data measured during the same first period, it takes time due to the data acquisition cycle of each sensor, filter processing of the data, etc., and a time difference (delay) may occur between the two pieces of data. The time delay correction unit 352 corrects such a time difference based on data based on the target data (e.g., data for determination) and data based on the external factor data (e.g., noise level data). Note that the time delay correction unit 352 is not an essential component. Also, the time delay correction unit 352 is an example of a delay correction unit.
[0079] The accuracy information calculation unit 353 calculates the measurement accuracy of the target data based on the external factor data acquired by the second acquisition unit 340. The accuracy information calculation unit 353 calculates the measurement accuracy of the target data acquired by the first acquisition unit 320 from the noise level calculated by the noise information calculation unit 351 using a table in which noise levels and measurement accuracy are associated (for example, see the second conversion table shown in FIG. 8). The calculated measurement accuracy is time-series data over a first period. Information including the measurement accuracy and time information is also referred to as accuracy information. The time information here includes, for example, a time stamp corrected by the time delay correction unit 352.
[0080] The accuracy information integrating unit 360 generates judgment data by associating the target data with accuracy information including the measurement accuracy, and outputs the generated judgment data. The accuracy information integrating unit 360 generates judgment data by integrating the target data with the accuracy information, and outputs the generated judgment data. Integrating means generating one piece of data from two separately acquired pieces of data in a state in which the relationship between the two pieces of data can be understood. The accuracy information integrating unit 360 functions as a data output unit that outputs judgment data to the judgment unit 370. The accuracy information integrating unit 360 outputs the same judgment data to the first judgment unit 371, the second judgment unit 372, and the third judgment unit 373.
[0081] The determination unit 370 determines the state of the object based on the determination data output from the accuracy information integration unit 360. The determination unit 370 has a first determination unit 371, a second determination unit 372, and a third determination unit 373. The first determination unit 371, the second determination unit 372, and the third determination unit 373 determine different states of the object based on the determination data. Note that it is sufficient for the determination unit 370 to determine at least one state of the object, and it is sufficient for the determination unit 370 to have at least one of the first determination unit 371 to the third determination unit 373.
[0082] The first determination unit 371 determines a first state of the object. The first determination unit 371 extracts a first portion that satisfies the first accuracy from the object data based on a first accuracy corresponding to the determination of the first state, and determines the first state based on the extracted first portion.
[0083] For example, the first determination unit 371 may determine whether or not the driver has difficulty driving the vehicle as the first state. For example, the first determination unit 371 may perform a determination for deadman detection. Deadman detection includes detecting a sudden change in the driver's physical condition, a situation in which the driver has difficulty driving, and the like. Deadman detection is performed by detecting a decrease in the driver's consciousness from at least one of the heart rate, the blink interval, and the steering time. Deadman detection may be performed by detecting a sudden change in the heart rate, for example. By performing deadman detection, it becomes possible to automatically stop the vehicle safely when the driver has difficulty driving the vehicle.
[0084] The second determination unit 372 determines a second state different from the first state. The second determination unit 372 extracts a second portion that satisfies the second accuracy from the target data based on the second accuracy according to the determination of the second state, and determines the second state based on the extracted second portion.
[0085] For example, the second determination unit 372 may determine whether the driver is dozing or looking aside as the second state. For example, the second determination unit 372 may perform a determination for detecting whether the driver is dozing. The driver's dozing can be detected from at least one of the heart rate, the blink interval, the steering time, the number of yawns, and image analysis of the camera image (for example, the driver's torso behavior).
[0086] Further, the second determination unit 372 may determine, for example, the fatigue of the driver. The fatigue determination is performed in order to provide the driver with a guide to take a rest, etc. The second determination unit 372 may estimate the state of the function of the autonomic nervous system from the heart rate, and calculate the fatigue level from the estimation result.
[0087] The third determination unit 373 determines a third state different from the first state and the second state. The third determination unit 373 extracts a third portion that satisfies the third accuracy from the target data based on the third accuracy according to the determination of the third state, and determines the third state based on the extracted third portion.
[0088] The third determination unit 373 may determine, for example, the other of whether the driver is dozing or looking aside as the third state. The third determination unit 373 may perform a determination for detecting, for example, the driver's looking aside. Looking aside can be detected from the driver's line of sight, etc. Looking aside can be detected, for example, by monitoring the driver's line of sight while the vehicle is traveling with a camera, etc. By detecting the driver's looking aside, it becomes possible to make the driver stop looking aside by voice, etc. In addition, by setting a condition under which it is acceptable to look aside (for example, while the vehicle is stopped), it is possible to reduce erroneous determinations under that condition.
[0089] The determinations made by the first determination unit 371 to the third determination unit 373 are not limited to the above. For example, the first determination unit 371 may determine the driver's drowsiness, the second determination unit 372 may determine either the driver's emotion or whether the driver is ill, and the third determination unit 373 may determine the other of the driver's emotion and whether the driver is ill. The driver's emotion can be used to determine whether a dangerous situation is imminent during driving, and can be detected, for example, based on the driver's micro-expression in an image. In addition, whether the driver is ill is determined in order to present a likely illness, and it is possible to determine a tendency for facial paralysis in stroke, retinopathy of the eye due to diabetes, and the like from an image. In addition, it is possible to determine whether the driver is ill, such as high blood pressure and myocardial infarction, from an electrocardiogram.
[0090] The output unit 380 outputs the determination result of the determination unit 370. The output unit 380 outputs the determination result to an external device, for example, via communication. The output unit 380 is, for example, communicably connected to the notification device 400, and may output the determination result to the notification device 400.
[0091] As described above, since the judgment data includes accuracy information, each of the first judgment unit 371, the second judgment unit 372, and the third judgment unit 373 can extract a portion of the judgment data with a desired accuracy and judge the state of the object using the extracted portion. The first accuracy, the second accuracy, and the third accuracy may be stored in advance in the judgment unit 370.
[0092] [1-2. Operation of the data generation system] Next, the operation of the data generation system 1 configured as above will be described with reference to Figs. 3 to 12. Fig. 3 is a flowchart showing the operation of the data generation system 1 according to this embodiment. Steps S11 to S18 shown in Fig. 3 show the operation (data generation method) of the precision-added information generating unit 310 of the information processing device 300, and steps S19 and S20 show the operation of the determining unit 370 of the information processing device 300. The data generation method includes a method of generating determination data for determining the state of the object. It is assumed that the object measurement sensor 100 is a millimeter wave sensor, and the external factor measurement sensor 200 is an acceleration sensor.
[0093] 3, the first acquisition unit 320 acquires object data obtained by measuring an object from the object measurement sensor 100 (S11). The first acquisition unit 320 outputs the acquired object data to the analysis unit 330.
[0094] Next, the analysis unit 330 performs feature conversion of the target data (S12). The analysis unit 330 converts the target data into a format used by the determination unit 370. For example, the analysis unit 330 converts the target data acquired from a millimeter wave sensor or the like into a heart rate, which is an example of a feature.
[0095] FIG. 4 is a diagram showing the determination data and noise levels according to the present embodiment.
[0096] In (a) of Fig. 4, heart rate data obtained by converting the feature amount of the target data measured by the target measurement sensor 100 is shown as an example of the determination data. As shown in Fig. 4, the determination data includes noise whose magnitude varies depending on the time. Note that the horizontal axis of (a) of Fig. 4 indicates time, and the vertical axis indicates the intensity of the determination data. The time is, for example, a time specified by a time stamp, and is the acquisition time of the target data.
[0097] 3 again, the analysis unit 330 performs a filter process on the determination data (S13). The analysis unit 330 is configured to include, for example, a band-pass filter, and performs a filter process using the band-pass filter. The analysis unit 330 outputs the filtered determination data to the accuracy information generation unit 350 (specifically, the time delay correction unit 352) and the accuracy information integration unit 360.
[0098] Next, the second acquisition section 340 acquires external factor data in which the external factor is measured from the external factor measurement sensor 200 (S14). The second acquisition section 340 outputs the acquired external factor data to the accuracy information generation section 350.
[0099] Next, the noise information calculation unit 351 calculates the noise level of the external factor based on the external factor data (S15). The noise information calculation unit 351 converts the external factor data into a noise level using, for example, the first conversion table shown in FIG.
[0100] FIG. 5 is a diagram showing a first conversion table between external factor data and noise levels according to the present embodiment.
[0101] As shown in Fig. 5, the first conversion table is a table showing the correspondence between external factor data and noise level, with the horizontal axis showing the strength of the external factor data and the vertical axis showing the noise level. For example, when the external factor is vibration and the external factor data is the acceleration of the vehicle, the higher the acceleration, the further to the right the coordinate of the horizontal axis is located. For example, Fig. 5 shows an example in which when the strength of the external factor data is a1, the noise level is n1. The noise level is expressed, for example, by a numerical value, but is not limited thereto.
[0102] The first conversion table is created according to the type of external factor (e.g., acceleration, external light, etc.). The first conversion table is, for example, acquired in advance by an experiment or the like, and is stored in advance in a storage unit (not shown) of the information processing device 300. When the information processing device 300 stores a plurality of conversion tables, the noise information calculation unit 351 may select a desired first conversion table from the plurality of conversion tables based on the type of external factor, information identifying the external factor measuring sensor 200, and the like. The storage unit is realized, for example, by a semiconductor memory or the like, but is not limited thereto.
[0103] The noise information calculation unit 351 uses the first conversion table to convert the time series data of the external factor data into time series data of the noise level (noise level data).
[0104] (b) of Fig. 4 shows time series data of the noise level of the external factor converted by the noise information calculation unit 351. As shown in (b) of Fig. 4, the noise level may change over time. For example, the noise level may change due to a change in the acceleration of the vehicle.
[0105] The noise information calculation unit 351 outputs the converted external factor noise level to the time delay correction unit 352 .
[0106] Referring again to Fig. 3, the time delay correction unit 352 corrects the delay due to the time correlation between the target data and the noise level (S16). The operation of step S16 will be described with reference to Fig. 4 and Fig. 6. Fig. 6 is a flowchart showing the operation (S16) of correcting the delay due to the time correlation shown in Fig. 3.
[0107] As shown in (a) and (b) of FIG. 4, a delay time T may occur between the determination data based on the object data and the noise level of an external factor based on the external factor data, both of which are measured during the same period (for example, the first period). In addition, it is considered that there is a strong relationship between the object data obtained by measuring the object and the external factor data that may become noise when measuring the object. In other words, it is considered that there is a strong relationship between the determination data and the noise level data. Therefore, in this embodiment, the delay (delay time T) is corrected by calculating a correlation that indicates the strength of the relationship between the determination data and the noise level data. For example, the delay is corrected by the process shown in FIG. 6 below.
[0108] As shown in FIG. 6, the time delay correction unit 352 sets the time difference ΔT=-T1 (S21). As shown in FIG. 4(a), the time delay correction unit 352 uses time T0 as a reference and places a frame (dashed line frame) of time width W on the judgment data with the time T0 as the center. The time width from time T0 to the end of the frame is W / 2. The time width W is, for example, a predetermined bit width. The frame of the time width W is, for example, rectangular, but is not limited to this. In addition, the time delay correction unit 352 uses time T0 as a reference and places a frame of time width W (not shown) with the time T0 as the center on the noise level of the external factor (noise level data). The frame on the noise level data can be moved in the time axis direction. ΔT is the time difference between time T0 and the center time of the frame on the noise level data.
[0109] Next, the time delay correction unit 352 shifts the time series of the noise level data of the external factor by ΔT (S22). The time delay correction unit 352 shifts the frame arranged on the noise level data by T1 in the negative direction of the time axis (see the dashed frame shown in (b) of FIG. 4).
[0110] Next, the time delay correction unit 352 calculates a correlation coefficient between the determination data and the noise level of the external factor (S23). The time delay correction unit 352 calculates the correlation coefficient between the two data by, for example, performing a convolution operation on the data within the determination data frame (inside the dashed frame) and the data within the frame (inside the dashed frame) when the time difference of the noise level is ΔT=−T1.
[0111] In step S23, the time delay correction unit 352 extracts the determination data and the noise level of the external factor for a certain period (time width W), and calculates one correlation coefficient indicating the strength of correlation based on the two extracted data.
[0112] Next, the time delay correction unit 352 sets ΔT=ΔT+Δt (S24) and judges whether ΔT>T1 (S25). In step S24, the time delay correction unit 352 judges whether the position has moved to the position of the dashed line frame shown in FIG. 4B (the center position of the frame is the position of T0+T1).
[0113] If ΔT>T1 is not satisfied (No in S25), the time delay correction unit 352 returns to step S22 and continues the process. The time delay correction unit 352 uses ΔT calculated in step S24 to shift the time width W (extraction section) of the noise level data by ΔT without changing the time width W (extraction section) of the determination data, and performs the processes from step S22 onward.
[0114] In this way, the time delay correction unit 352 generates a graph (see FIG. 7 described later) of ΔT and the correlation coefficient by shifting the frame on the noise level data by Δt.
[0115] Furthermore, if ΔT>T1 (Yes in S25), the time delay correction unit 352 acquires the time difference at which the correlation coefficient is maximum based on a graph of ΔT and the correlation coefficient (S26). Fig. 7 is a diagram showing the relationship between ΔT and the correlation coefficient according to this embodiment.
[0116] As shown in Fig. 7, the time delay correction unit 352 obtains the time difference T2, which is the time difference at which the correlation coefficient is maximum. Since there is a strong relationship between the target data and the noise level data, it is considered that the correlation coefficient becomes smaller as ΔT becomes more distant from the actual delay. Therefore, the time difference ΔT with the largest correlation coefficient (time difference T2 in the example of Fig. 7) becomes the delay time T between the target data and the noise level data.
[0117] Next, the time delay correction unit 352 corrects the time series of the noise level data of the external factor based on the time difference T2 acquired in step S26 (S27). The time delay correction unit 352 shifts the time stamp of the noise level data by ΔT (here, the time difference T2). It can also be said that the time delay correction unit 352 corrects the time information corresponding to the noise level data based on the time difference T2. This makes it possible to match the time stamps of the target data and the noise level data.
[0118] The time delay correction unit 352 outputs the noise level and the corrected time stamp (time information) to the accuracy information calculation unit 353.
[0119] As a result, it is possible to more accurately correspond the time between the determination data and the accuracy information in the accuracy-information-attached data described later. In other words, the accuracy-attached information generating unit 310 can generate accuracy-attached data that can more accurately determine the target object.
[0120] It is to be noted that the process of step S16 may not be performed. For example, if it is known in advance that the time difference between the target data and the noise level is within a predetermined time, the process of step S16 may not be performed.
[0121] 3 again, the accuracy information calculation unit 353 calculates the accuracy of the target data based on the noise level data with the delay corrected (S17). The accuracy information calculation unit 353 converts the noise level into accuracy, for example, by using the second conversion table shown in FIG.
[0122] FIG. 8 is a diagram showing a second conversion table between noise level and accuracy according to the present embodiment.
[0123] As shown in Fig. 8, the second conversion table is a table showing the correspondence between the noise level and the accuracy of the judgment data for the noise level, with the horizontal axis showing the noise level and the vertical axis showing the accuracy of the judgment data. For example, Fig. 8 shows an example in which the accuracy is p1 when the noise level is n1. In other words, the accuracy of the target data measured when the external factor data is a1 is p1. The accuracy is expressed by, for example, a numerical value, but is not limited thereto.
[0124] The second conversion table is obtained in advance by, for example, an experiment or the like, and is stored in advance in a storage unit (not shown) included in the information processing device 300.
[0125] The accuracy information calculation unit 353 converts the time series data of the noise level into time series data of the accuracy using the second conversion table. Then, the accuracy information calculation unit 353 associates the converted accuracy with the corrected time information and outputs it to the accuracy information integration unit 360.
[0126] 3 again, next, the accuracy information integrating unit 360 generates data with accuracy information based on the target data acquired from the analyzing unit 330 and the accuracy information (accuracy and time information) acquired from the accuracy information calculating unit 353 (S18). The accuracy information-added data is data in which the target data and the accuracy information are associated with each other. The accuracy information integrating unit 360 generates the accuracy information-added data by temporally associating the target data with the accuracy based on the timestamp included in the target data and the corrected time information.
[0127] Fig. 9 is a diagram showing data with accuracy information according to the present embodiment, in which heart rate data with accuracy information is illustrated as an example of the data with accuracy information.
[0128] As shown in Fig. 9, the data with accuracy information is associated with target data and accuracy. In the example of Fig. 9, the accuracy of the target data between times T3 and T4 is 2, the accuracy of the target data between times T4 and T5 is 10, and the accuracy of the target data between times T5 and T6 is 7. Note that the higher the accuracy value, the higher the accuracy.
[0129] From this, it can be seen that the accuracy of the target data between times T3 and T4 is low, and the accuracy of the target data between times T4 and T5 is high. In this way, the accuracy information integration unit 360 according to this embodiment is characterized in that it adds information indicating the accuracy of the target data to the target data.
[0130] The accuracy information integrating unit 360 outputs the generated data with accuracy information to the determining unit 370. Specifically, the accuracy information integrating unit 360 outputs the same data with accuracy information to each of the first determining unit 371, the second determining unit 372, and the third determining unit 373. It can also be said that the accuracy information integrating unit 360 outputs data with accuracy information in which the time delay has been corrected and the target data and the accuracy information are associated with each other.
[0131] 3 again, the determination unit 370 performs a determination process to determine the state of the object based on the data with accuracy information acquired from the accuracy information integration unit 360 (S19). In step S19, each of the first determination unit 371, the second determination unit 372, and the third determination unit 373 extracts a portion that satisfies an accuracy corresponding to the determination content from the data with accuracy information, and performs a determination of the state of the object based on the extracted portion.
[0132] FIG. 10 is a flowchart showing the operation of the determination process shown in FIG.
[0133] 10, the first determination unit 371 determines whether or not there is data in the data with accuracy information, where the accuracy of the data with accuracy information is greater than P1 (S31). P1 is a threshold value of accuracy according to the determination (first determination) of the first state of the target object made by the first determination unit 371, and is an example of the first accuracy. The threshold value P1 is a necessary accuracy that the target data should have in order to accurately make the first determination.
[0134] If there is data where the accuracy of the data with accuracy information is greater than the threshold P1 (Yes in S31), the first judgment unit 371 extracts data whose accuracy is higher than the threshold P1 from the target data (measurement data) (S32). For example, if there is data where the accuracy of the data with accuracy information is greater than the threshold P1, the first judgment unit 371 extracts a portion of the judgment data whose accuracy is higher than the threshold P1 from the data with accuracy information. The extracted data is also referred to as data for the first judgment unit. Furthermore, if there is no data where the accuracy of the data with accuracy information is greater than the threshold P1 (No in S31), the first judgment unit 371 does not perform the first judgment and proceeds to step S36.
[0135] FIG. 11 is a diagram showing data for the first determination unit according to the present embodiment.
[0136] As shown in FIG. 11, when the threshold value P1 is 8 or more (when the required accuracy is 8 or more), only the judgment data between time T4 and time T5 where the accuracy is 8 or more is extracted, and the extracted judgment data is set as the first judgment data.
[0137] 10 again, the first judgment unit 371 performs a first judgment based on the extracted first judgment data (S33). In this way, the first judgment unit 371 extracts a portion of the judgment data that satisfies the desired accuracy from the accuracy-information-added data, and performs the first judgment using only the extracted portion, so that the first judgment can be performed more accurately than when all of the judgment data from time T3 to T6 is used.
[0138] Next, the first determination unit 371 determines whether or not notification of the first determination result is necessary (S34). The first determination unit 371 makes this determination based on, for example, a table in which the first determination result is associated with the necessity of notification. If notification of the first determination result is necessary (Yes in S34), the first determination unit 371 performs processing to cause the notification device 400 to notify the first determination result (S35). If Yes in S34, the first determination unit 371 outputs, for example, the first determination result to the output unit 380. If notification of the first determination result is not necessary (No in S34), the first determination unit 371 proceeds to step S36.
[0139] Next, the second determination unit 372 determines whether or not there is data in the accuracy-information-attached data where the accuracy of the accuracy-information-attached data is greater than P2 (S36). P2 is an accuracy threshold according to the determination (second determination) of the second state of the target object made by the second determination unit 372, and is an example of the second accuracy. The threshold P2 is the required accuracy that the target data should have in order to accurately make the second determination.
[0140] If there is data where the accuracy of the data with accuracy information is greater than the threshold P2 (Yes in S36), the second judgment unit 372 extracts data with accuracy higher than the threshold P2 from the target data (S37). If there is data where the accuracy of the data with accuracy information is greater than the threshold P2, the second judgment unit 372 extracts a portion of the judgment data with accuracy higher than the threshold P2 from the data with accuracy information. The extracted data is also referred to as data for the second judgment unit. If there is no data where the accuracy of the data with accuracy information is greater than the threshold P2 (No in S36), the second judgment unit 372 does not perform the second judgment and proceeds to step S41.
[0141] FIG. 12 is a diagram showing data for the second judging section according to the present embodiment.
[0142] 12, when the threshold P2 is 1 or more (when the required accuracy is 1 or more), the judgment data between time T3 and time T6 where the accuracy is 1 or more is extracted, and the extracted judgment data is set as the second judgment data. In other words, the second judgment data includes all of the judgment data included in the accuracy-information-added data.
[0143] 10 again, the second judgment unit 372 performs a second judgment based on the extracted second judgment data (S38). In this way, when the threshold P2 of the data used for judgment is low, the second judgment unit 372 performs the second judgment using all of the judgment data from time T3 to T6. By using all of the judgment data from time T3 to T6, it is possible to prevent the number of data used for judgment from decreasing, and therefore the second judgment unit 372 can prevent the judgment from becoming inaccurate due to the decrease in the number of data.
[0144] Next, the second determination unit 372 determines whether or not notification of the second determination result is necessary (S39). The second determination unit 372 makes this determination based on, for example, a table in which the second determination result is associated with the necessity of notification. If notification of the second determination result is necessary (Yes in S39), the second determination unit 372 performs processing for causing the notification device 400 to notify the second determination result (S40). If Yes in S39, the second determination unit 372 outputs, for example, the second determination result to the output unit 380. If notification of the second determination result is not necessary (No in S39), the second determination unit 372 proceeds to step S41.
[0145] Next, the third determination unit 373 determines whether or not there is data in the data with accuracy information where the accuracy of the data with accuracy information is greater than P3 (S41). P3 is an accuracy threshold according to the determination of the third state of the object (third determination) made by the third determination unit 373, and is an example of the third accuracy. The threshold P3 is the required accuracy that the target data should have in order to accurately make the third determination.
[0146] If there is data where the accuracy of the data with accuracy information is greater than the threshold P3 (Yes in S41), the third judgment unit 373 extracts data whose accuracy is higher than the threshold P3 from the target data (S42). If there is data where the accuracy of the data with accuracy information is greater than the threshold P3, the third judgment unit 373 extracts a portion of the judgment data whose accuracy is higher than the threshold P3 from the data with accuracy information. The extracted data is also referred to as data for the third judgment unit. Furthermore, if there is no data where the accuracy of the data with accuracy information is greater than the threshold P3 (No in S41), the third judgment unit 373 does not perform the third judgment and ends the judgment process.
[0147] The third determination unit 373 performs a third determination based on the extracted third determination data (S43). In this way, the third determination unit 373 extracts a portion of the determination data that satisfies the desired accuracy from the accuracy-information-added data, and performs the third determination using only the extracted portion, so that the third determination can be performed more accurately than when all of the target data from times T3 to T6 is used.
[0148] Next, the third determination unit 373 determines whether or not notification of the third determination result is necessary (S44). The third determination unit 373 makes this determination based on, for example, a table in which the third determination result is associated with whether or not notification is necessary. If notification of the third determination result is necessary (Yes in S44), the third determination unit 373 performs processing for causing the notification device 400 to notify the third determination result (S45). If Yes in S44, the third determination unit 373 outputs, for example, the third determination result to the output unit 380. If notification of the third determination result is not necessary (No in S44), the third determination unit 373 ends the determination processing.
[0149] The accuracy thresholds P1, P2, and P3 may be different values from each other. The thresholds P1, P2, and P3 are obtained in advance and stored in a storage unit (not shown) included in the information processing device 300.
[0150] It should be noted that the various processes in the first determination section 371, the second determination section 372, and the third determination section 373 may be performed in parallel, for example.
[0151] 3 again, output unit 380 outputs the determination result acquired from determination unit 370 (S20). Output unit 380 outputs the determination result to notification device 400, for example.
[0152] The notification device 400 notifies the user of the acquired determination result, thereby making it possible to notify the user of only the desired determination result.
[0153] (Embodiment 2) The data generation system according to this embodiment will be described below with reference to Fig. 13 to Fig. 18. Note that the following description will focus on differences from the first embodiment, and descriptions of contents that are the same as or similar to the first embodiment will be omitted or simplified.
[0154] [2-1. Data generation system configuration] First, the configuration of the data generation system according to this embodiment will be described with reference to FIG. 13. FIG. 13 is a block diagram showing the functional configuration of an information processing device 300a according to this embodiment. The information processing device 300a according to this embodiment differs from the information processing device 300 according to the first embodiment in that it has a third acquisition unit 341. In this embodiment, the external factor data acquired by the second acquisition unit 340 is described as the first external factor data in order to distinguish it from the second external factor data acquired by the third acquisition unit 341, but the data acquired by the second acquisition unit 340 is the same as in the first embodiment.
[0155] For example, when the object measurement sensor 100 is a camera or the like, the vibration of the vehicle and the change in external light can be external factors of the object data. In this embodiment, a method of generating data with accuracy information using a plurality of external factors will be described.
[0156] As shown in FIG. 13, the information processing device 300a includes a precision-added information generating unit 310a, a determining unit 370, and an output unit 380.
[0157] The third acquisition unit 341 acquires second external factor data, in which an external factor that may become noise in the target data is measured, from another external factor measurement sensor (not shown) different from the object measurement sensor 100 and the external factor measurement sensor 200. The third acquisition unit 341 is configured to include, for example, a communication interface for communicating with the other external factor measurement sensor. The second external factor data is time-series data in which an external factor is measured over a second period at least partly overlapping with the first period, and includes time information indicating the time when the external factor was measured. For example, the first period and the second period may be the same period. The second external factor data is data in which an external factor different from the first external factor data is measured.
[0158] In this embodiment, the data generating system 1 further includes the other external factor measuring sensor. The other external factor measuring sensor is an example of a third sensor. The third sensor may be, for example, a brightness sensor.
[0159] The accuracy information calculation unit 353 calculates the measurement accuracy based on the second external factor data in addition to the first external factor data. The accuracy information calculation unit 353 calculates one accuracy based on the first external factor data and the second external factor data.
[0160] [2-2. Operation of the data generation system] Next, the operation of the data generating system 1 configured as above will be described with reference to Figs. 14 to 18. Fig. 14 is a flowchart showing the operation of the data generating system 1 according to this embodiment. Fig. 15 is a diagram showing a third conversion table between the first external factor data and the noise level according to this embodiment. Fig. 16 is a diagram showing a fourth conversion table between the second external factor data and the noise level according to this embodiment.
[0161] The flowchart shown in Fig. 14 includes steps S51 and S52 in addition to the flowchart shown in Fig. 3, and includes steps S16a and S17a instead of steps S16 and S17. Also, the third conversion table shown in Fig. 15 is a table similar to the first conversion table according to embodiment 1. In step S15 shown in Fig. 14, the noise level of the first external factor is calculated using the third conversion table shown in Fig. 15, but since this is the same process as step S15 shown in Fig. 3, a description thereof will be omitted.
[0162] 14, the third acquisition unit 341 acquires second external factor data in which an external factor is measured from another external factor measurement sensor (S51). The third acquisition unit 341 outputs the acquired second external factor data to the accuracy information generation unit 350.
[0163] The order in which the target data, the first external factor data, and the second external factor data are acquired is not limited to the order shown in FIG.
[0164] Next, the noise information calculation unit 351 calculates the noise level of the second external factor based on the second external factor data (S52). The noise information calculation unit 351 converts the second external factor data into a noise level using, for example, the fourth conversion table shown in FIG.
[0165] As shown in FIG. 16, the fourth conversion table is a table showing the correspondence between the second external factor data and the noise level, with the horizontal axis showing the strength of the second external factor data and the vertical axis showing the noise level. For example, if the second external factor is light (external light) and the second external factor data is the luminance of the light incident on the vehicle, the higher the luminance, the further to the right the coordinate of the horizontal axis is located. For example, FIG. 16 shows an example in which when the strength of the second external factor data is b1, the noise level is n2. The noise level is expressed, for example, by a numerical value, but is not limited thereto.
[0166] The fourth conversion table is created according to the type of the second external factor (for example, acceleration, external light). The fourth conversion table is obtained in advance, for example, by an experiment or the like, and is stored in advance in a storage unit (not shown) of the information processing device 300a. As shown in Figs. 15 and 16, the two conversion tables are defined by different functions.
[0167] The noise information calculation unit 351 converts the time series data of the second external factor data into time series data of the noise level using the fourth conversion table. Then, the noise information calculation unit 351 outputs the converted noise level of the second external factor to the time delay correction unit 352.
[0168] Referring again to Fig. 14, the time delay correction unit 352 corrects the delay due to the time correlation between the target data and the noise level (S16a). The operation of step S16a will be described with reference to Fig. 17. Fig. 17 is a flowchart showing the operation (S16a) of correcting the delay due to the time correlation shown in Fig. 14. The operations of steps S21 to S27 shown in Fig. 17 are similar to steps S21 to S27 shown in Fig. 6, and therefore will not be described.
[0169] As shown in Fig. 17, the time delay correction unit 352 according to this embodiment performs the operations of steps S61 to S67 in addition to those of the first embodiment. Steps S61 to S67 are operations for correcting a time delay for the noise level data of the second external factor. It can also be said that steps S61 to S67 are operations for correcting a delay due to a time correlation between the target data and the noise level of the second external factor. The operations of steps S61 to S67 are similar to those of steps S21 to S27, and therefore a description thereof will be omitted.
[0170] In this way, the time delay correction unit 352 according to this embodiment corrects the time delay between the target data and the first external factor data, as well as the time delay between the target data and the second external factor data.
[0171] In step S61, an example is shown in which ΔT=−T1, that is, ΔT is the same as that in step S21, but this is not limited to this, and ΔT in step S21 and ΔT in step S61 may be different values.
[0172] 14 again, next, the accuracy information calculation unit 353 calculates the accuracy of the target data based on the noise level data of the first external factor data with the delay corrected and the noise level data of the second external factor data with the delay corrected (S17a). The accuracy information calculation unit 353 converts the two noise levels into one accuracy using, for example, the fifth conversion table shown in FIG.
[0173] FIG. 18 is a diagram showing a fifth conversion table between noise level and accuracy according to the present embodiment.
[0174] As shown in Fig. 18, the fifth conversion table is a table showing a correspondence relationship between a first noise level indicating the noise level of a first external factor, a second noise level indicating the noise level of a second external factor, and the accuracy of the determination data for the two noise levels. In Fig. 18, the table is shown in a three-dimensional orthogonal coordinate system, and each axis indicates the first noise level, the second noise level, and the accuracy of the determination data. This makes it possible to obtain one accuracy from the first noise level and the second noise level.
[0175] The fifth conversion table is obtained in advance, for example, through experiments or the like, and is stored in advance in a storage unit (not shown) included in the information processing device 300a.
[0176] The accuracy information calculation unit 353 converts the time series data of the first noise level and the second noise level into time series data of accuracy using the fifth conversion table. Then, the accuracy information calculation unit 353 associates the converted accuracy with the corrected time information, and outputs the converted accuracy to the accuracy information integration unit 360.
[0177] (Embodiment 3) The data generation system according to this embodiment will be described below with reference to Fig. 19 to Fig. 26. Note that the following description will focus on differences from the first embodiment, and descriptions of contents that are the same as or similar to the first embodiment will be omitted or simplified.
[0178] [3-1. Data generation system configuration] First, the configuration of the data generation system according to the present embodiment will be described with reference to Fig. 19. Fig. 19 is a diagram showing the configuration of a data generation system 1b according to the present embodiment. The data generation system 1b according to the present embodiment differs from the data generation system 1 according to the first embodiment in that it includes a server device 500, a terminal device 600, an input device 700, and a display device 800, and in that it includes an information processing device 300b instead of the information processing device 300.
[0179] As shown in FIG. 19, the data generation system 1b includes an object measurement sensor 100, an external factor measurement sensor 200, an information processing device 300b, a notification device 400, a server device 500, a terminal device 600, an input device 700, and a display device 800. The data generation system 1b updates at least one of the determination method and threshold value (determination criterion) used in the information processing device 300b based on the authentication result of the object. The data generation system 1b further updates the conversion data used in the information processing device 300b. In the following, an example will be described in which the object is a driver and the object measurement sensor 100 and the external factor measurement sensor 200 are mounted on a vehicle.
[0180] The information processing device 300b acquires at least one of a determination method and a threshold value corresponding to the driver from the server device 500 based on the driver authentication result in the input information acquired by the input device 700. Then, the information processing device 300b performs various processes using at least one of the acquired determination method and threshold value. In addition, the information processing device 300b performs various conversions using the conversion data acquired from the server device 500.
[0181] The server device 500 is communicatively connected to the information processing device 300b, and performs processing for outputting to the information processing device 300b the authentication result of the driver authentication, as well as at least one of the conversion data for updating, the determination method, and the threshold value.
[0182] Although the server device 500 and the information processing device 300b are separate devices, they may be realized as, for example, an integrated device.
[0183] The terminal device 600 is communicably connected to the server device 500 and presents information from the server device 500. The terminal device 600 may be a device owned by the driver, or may be a preset device to which authentication results and the like are transmitted. The terminal device 600 may be, for example, a mobile terminal such as a smartphone, or a stationary device such as a PC (personal computer).
[0184] The input device 700 receives input from the driver. The input device 700 is realized by, for example, a button, a touch panel, a mouse, etc., but may also be realized by a device (for example, a microphone, a camera, etc.) that receives input from the driver by, for example, voice, gesture, etc.
[0185] The input device 700 may be configured to be capable of acquiring biometric information specific to the driver. The input device 700 may be configured to include a reader for reading, for example, a fingerprint, an iris, or the like.
[0186] The display device 800 is communicably connected to the information processing device 300b, and displays information from the server device 500 to the driver. The display device 800 may display, for example, an authentication result of the server device 500. The display device 800 is provided, for example, in an object (for example, a vehicle) on which the object measurement sensor 100 and the external factor measurement sensor 200 are mounted. The display device 800 is realized, for example, by a liquid crystal display device or the like.
[0187] In addition, when the object on which the object measuring sensor 100 and the external factor measuring sensor 200 are mounted is a vehicle, the input device 700 and the display device 800 may be realized by, for example, a car navigation system.
[0188] Here, details of the information processing device 300b and the server device 500 will be described with reference to Fig. 20. Fig. 20 is a block diagram showing functional configurations of the information processing device 300b and the server device 500 according to this embodiment.
[0189] 20, the information processing device 300b includes a fourth acquisition unit 342 and an input information processing unit 390 in addition to the information processing device 300 according to the first embodiment, and has an accuracy information generation unit 350b instead of the accuracy information generation unit 350. The information processing device 300b includes a precision-added information generation unit 310b, a determination unit 370, and an output unit 380.
[0190] The fourth acquisition unit 342 acquires input information indicating an input by the driver from the input device 700. The input information includes information for authenticating the driver acquired from the driver. The input information includes, but is not limited to, an ID and a password (authentication information), an image of a face or the like, biometric information, and the like. The fourth acquisition unit 342 is configured to include, for example, a communication interface for communicating with the input device 700. The biometric information is, for example, a fingerprint, an iris, voice information, and the like.
[0191] The input information processing unit 390 performs a predetermined process on the input information. In the present embodiment, the input information processing unit 390 performs a process of outputting the input information acquired by the fourth acquisition unit 342 to the server device 500. In this case, the input information processing unit 390 is configured to include a communication interface for communicating with the server device 500. Note that the input information processing unit 390 may also be connected to, for example, the display device 800 so as to be able to communicate with it.
[0192] The accuracy information generating unit 350b includes a coefficient updating unit 354 in addition to the accuracy information generating unit 350 according to the first embodiment.
[0193] The coefficient update unit 354 updates various information used for the operation of the noise information calculation unit 351 and the accuracy information calculation unit 353 based on information from the server device 500. The coefficient update unit 354 updates the first conversion table and the second conversion table. For example, the coefficient update unit 354 may access a server device that manages the first conversion table and the second conversion table, and when the first conversion table and the second conversion table are updated, acquire the updated first conversion table and the second conversion table, and store the acquired first conversion table and the second conversion table in a storage unit (not shown). The coefficient update unit 354 is an example of a first update unit.
[0194] The server device 500 includes a personal authentication unit 510 and a determination method update unit 520 .
[0195] The personal authentication unit 510 authenticates the driver based on the input information of the driver acquired via the input device 700. The personal authentication unit 510 may identify the driver by, for example, comparing the input information, such as an ID, a password, and biometric information, acquired via the input device 700, with a pre-registered ID, a password, and biometric information. The personal authentication unit 510 may identify the driver by, for example, fingerprint authentication, iris authentication or face authentication, voice authentication, or the like. Note that the method by which the personal authentication unit 510 authenticates the driver is not limited to the above, and any known technology may be used.
[0196] The determination method update unit 520 performs processing for updating at least one of the determination method and the threshold value based on the authentication result of the personal authentication unit 510 to at least one of the determination method and the threshold value according to the driver.
[0197] At least the input device 700 and the display device 800 are mounted on an object (for example, a vehicle) on which the object measuring sensor 100 and the external factor measuring sensor 200 are mounted.
[0198] The input information processing unit 390 may perform the process performed by the personal authentication unit 510. In other words, the input information processing unit 390 may perform a process of outputting an authentication result based on the input information. In this case, the server device 500 may not have the personal authentication unit 510.
[0199] [3-2. Operation of the data generation system] Next, the operation of the data generation system 1b configured as above will be described with reference to Fig. 21 to Fig. 26. Fig. 21 is a flowchart showing the operation of the data generation system 1b according to this embodiment. Fig. 21 shows the operation of the information processing device 300b.
[0200] The flowchart shown in FIG. 21 includes steps S71 and S72 in addition to the components in the flowchart shown in FIG. 3, and also includes steps S15b, S17b, and S19b instead of steps S15, S17, and S19.
[0201] As shown in Fig. 21, the input information processing unit 390 performs processing for personal authentication based on the input information for authenticating the driver acquired from the driver by the fourth acquisition unit 342 (S71). The input information processing unit 390 performs processing to output the input information to the server device 500, for example. Fig. 22 is a flowchart showing the operation (S71) of personal authentication shown in Fig. 21. Fig. 22 shows a case where the input information is an ID and a password.
[0202] 22, the input information processing unit 390 determines whether or not an individual has not been authenticated (S81). The input information processing unit 390 acquires information such as the driver's name via the fourth acquisition unit 342, and determines whether or not authentication processing of the acquired driver has been performed. For example, the input information processing unit 390 may perform the determination of step S81 every time the ignition of the vehicle is turned on, or may perform the determination of step S81 at predetermined time intervals.
[0203] If the individual has not been authenticated (Yes in S81), the input information processing unit 390 requests the input of an ID and a password (S82). The input information processing unit 390, for example, causes the display device 800 to display a screen for inputting an ID and a password.
[0204] Next, when the input information processing unit 390 acquires an input of an ID and a password from the input device 700 via the fourth acquisition unit 342 (S83), it performs a login process (S84). For example, as the login process, the input information processing unit 390 outputs the ID and password to the server device 500 and acquires an authentication result from the server device 500. Note that the personal authentication unit 510 of the server device 500 determines whether or not the ID and password from the input information processing unit 390 match the registered ID and password.
[0205] Next, the input information processing unit 390 judges whether the login has been successful or not (S85). The input information processing unit 390 may, for example, acquire information indicating whether the login has been successful or not based on a judgment result indicating whether the ID and password match from the server device 500, and judge whether the login has been successful or not based on the acquired information. For example, the input information processing unit 390 may acquire an authentication result in the server device 500. The input information processing unit 390 functions as an authentication result acquisition unit.
[0206] If the login is successful (Yes in S85), the input information processing unit 390 ends the personal authentication operation, and if the login is unsuccessful (No in S85), it performs error processing (S86). As the error processing, the input information processing unit 390 may, for example, cause the display device 800 to display information indicating that the login has failed (personal authentication has failed). Then, the input information processing unit 390 may return to step S82 and request the input of the ID and password again.
[0207] If the individual has already been authenticated (No in S81), the input information processing unit 390 determines whether or not there is a change input (S87). The input information processing unit 390 may make the determination in step S87 based on whether or not an input indicating a change to at least one of the ID and the password has been received from the driver, for example.
[0208] If there is any change input (Yes in S87), input information processing unit 390 proceeds to step S82 and requests input of the changed ID and password, and if there is no change input (No in S87), ends the personal authentication operation.
[0209] Here, the operation of the server device 500 after successful login will be described with reference to Fig. 23. Fig. 23 is a flowchart showing the operation of the server device 500 according to this embodiment.
[0210] 23, the determination method update unit 520 determines whether or not the vehicle is starting up (S91). The determination method update unit 520 may determine whether or not the vehicle is starting up from, for example, a driving history of the vehicle. The term "not starting up the vehicle" includes, for example, a predetermined time or more has passed since the vehicle was started up.
[0211] When the vehicle is started (Yes in S91), the determination method update unit 520 checks whether the cloud data has been updated (S92). The determination method update unit 520 checks whether the information of the driver has been updated in another server device that manages the information of the driver. The information of the driver may be, for example, the driver's medical history, a disease to be determined, sleep time, etc., or other information about the driver.
[0212] Next, if there is an update to the cloud data, the determination method update unit 520 acquires the updated cloud data (S93), and determines a determination method and threshold value according to the individual (driver) based on the acquired cloud data (S94).
[0213] For example, in a case where the first judgment unit 371 judges whether the driver is dozing, the second judgment unit 372 judges the driver's emotions, and the third judgment unit 373 judges whether the driver is sick, the judgment method update unit 520 may update at least one of the judgment methods of dozing, emotions, and illness, and at least one of the threshold values.
[0214] Taking an example of updating at least one of the determination method and the threshold value in determining an illness, the determination method update unit 520 acquires, for example, data on the driver's underlying illness, chronic illness, most recent diagnosis result, etc., as cloud data. The determination method update unit 520 may also acquire, for example, information on illnesses registered by the driver that the driver wishes to detect while the driver is driving the vehicle, as cloud data.
[0215] In step S94, the determination method update unit 520 may determine a determination method and a threshold value that can effectively detect a disease according to a disease included in the most recent diagnosis result or a disease to be detected. In addition, in step S94, the determination method update unit 520 may determine a determination method and a threshold value that prevent the driver's underlying disease or chronic disease from being detected repeatedly.
[0216] Next, the determination method update unit 520 outputs the updated determination method and threshold (update data) to the information processing device 300b (S95).
[0217] If the vehicle is not started (No in S91), the determination method update unit 520 further determines whether or not the personal authentication has occurred immediately after the personal authentication (S96). If the personal authentication has occurred immediately after the personal authentication (Yes in S96), the determination method update unit 520 proceeds to step S92, and if the personal authentication has not occurred immediately after the personal authentication (No in S96), the determination method update unit 520 ends the process.
[0218] The determination method update unit 520 may update at least one of the determination method and threshold value of dozing and emotion according to cloud data related to illness determination.
[0219] 21 again, the determination unit 370 acquires the update data from the server device 500 (S72). The determination unit 370 may store the update data in a storage unit.
[0220] Further, the noise information calculation unit 351 calculates the noise level of the external factor based on the external factor data (S15b). Fig. 24 is a flowchart showing the operation of calculating the noise level of the external factor shown in Fig. 21.
[0221] As shown in FIG. 24, the coefficient update unit 354 determines whether or not there is an update to the first conversion table (S101). The coefficient update unit 354 may, for example, access a server device that manages the first conversion table to determine whether or not there is an update. When there is an update to the first conversion table, the coefficient update unit 354 acquires the updated first conversion table and updates the coefficient table (for example, the first conversion table) based on the acquired first conversion table (S102). For example, the coefficient update unit 354 replaces the old first conversion table with the updated first conversion table.
[0222] Next, the noise information calculation unit 351 calculates a noise level based on the updated coefficient table (the updated first conversion table) (S103). Step S103 is the same process as step S15 shown in FIG. 3, and the description will be omitted.
[0223] Furthermore, the accuracy information calculation unit 353 calculates the accuracy of the target data based on the noise level data in which the delay has been corrected (S17b). Fig. 25 is a flowchart showing the accuracy calculation operation (17b) shown in Fig. 21.
[0224] As shown in FIG. 25, the coefficient update unit 354 2 The coefficient update unit 354 determines whether there is an update to the second conversion table (S111). The coefficient update unit 354 may access a server device and determine whether or not there is an update. When the second conversion table is updated, the coefficient update unit 354 acquires the updated second conversion table and updates the coefficient table (e.g., the second conversion table) based on the acquired second conversion table (S112). For example, the coefficient update unit 354 replaces the old second conversion table with the updated second conversion table.
[0225] Next, the accuracy information calculation unit 353 converts the noise level into accuracy based on the updated coefficient table (updated second conversion table) (S113). Step S113 is the same process as step S17 shown in FIG. 3, and the description will be omitted.
[0226] Referring again to Fig. 21, the determination unit 370 performs a determination process for determining the state of the target data based on the noise level data with the delay corrected (S19b). Fig. 26 is a flowchart showing the operation (S19b) of the determination process shown in Fig. 21.
[0227] The flowchart shown in FIG. 26 includes steps S121 and S122 in addition to the flowchart shown in FIG. 6 of the first embodiment.
[0228] First, the determination unit 370 determines whether or not the determination method and the threshold have been updated (S121). The determination unit 370 may perform the determination in step S121 based on whether or not the determination method and the threshold have been acquired from the determination method update unit 520 after the previous determination process.
[0229] If there is an update to the judgment method and thresholds (e.g., thresholds P1 to P3) (Yes in S121), the judgment unit 370 updates the judgment method and thresholds P1 to P3 (S122). The judgment unit 370 replaces the judgment method and thresholds P1 to P3 used in the judgments of steps S31 to S45 with the judgment method and thresholds P1 to P3 acquired from the judgment method update unit 520. Then, the judgment unit 370 performs the processes of step S31 and thereafter.
[0230] Furthermore, if the determination method and thresholds P1 to P3 have not been updated (No in S121), the determination unit 370 proceeds to step S31. In this case, the processes from step S31 onwards are performed using the determination method and thresholds P1 to P3 used in the previous determination process.
[0231] The determination unit 370 only needs to update at least one of the determination method and the threshold determined based on the authentication result. The determination unit 370 functions as a second update unit.
[0232] (Embodiment 4) The data generation system according to this embodiment will be described below with reference to Fig. 27 to Fig. 29. Note that the following description will focus on differences from the third embodiment, and descriptions of the same or similar contents as the third embodiment will be omitted or simplified.
[0233] [4-1. Data generation system configuration] First, the configuration of the data generation system according to the present embodiment will be described with reference to Fig. 27. Fig. 27 is a block diagram showing the functional configuration of an information processing device 300c and a server device 500c according to the present embodiment. The data generation system according to the present embodiment differs from the data generation system 1b according to the third embodiment in that the server device 500 includes a determination unit 370.
[0234] 27, the data generation system according to the present embodiment includes an information processing device 300c and a server device 500c. The object measurement sensor 100, the external factor measurement sensor 200, the notification device 400, the terminal device 600, the input device 700, and the display device 800 are the same as those in the third embodiment, and are not shown in the figure.
[0235] 27, the information processing device 300c has a configuration in which the determination unit 370 and the like are removed from the information processing device 300b according to embodiment 3. The server device 500c further includes the determination unit 370 from the server device 500 according to embodiment 3. The configuration and functions of the determination unit 370 of the server device 500c are similar to those of the determination unit 370 according to embodiment 3, and therefore description thereof will be omitted.
[0236] As described above, the information processing device 300c only needs to have a configuration for generating data with accuracy information.
[0237] [4-2. Operation of the data generation system] Next, the operation of the data generation system configured as above will be described with reference to Fig. 28 and Fig. 29. Fig. 28 shows the operation of the information processing device 300c according to this embodiment. Note that steps S71 to S18 shown in Fig. 28 are similar to steps S71 to S18 shown in Fig. 21, and description thereof will be omitted. Fig. 29 is a flowchart showing the operation of the server device 500c according to this embodiment. Fig. 29 shows the operation of the determination unit 370 of the server device 500c.
[0238] As shown in FIG. 28, when the accuracy information integration unit 360 generates the data with accuracy information in step S18, it outputs the generated data with accuracy information to the server device 500c (S131).
[0239] As shown in FIG. 29, the determination unit 370 of the server device 500c acquires data with accuracy information from the information processing device 300c (S141).
[0240] Next, the determination unit 370 performs a determination process based on the acquired data with accuracy information (S142). The determination process may be the same as the process shown in FIG.
[0241] Next, the determination unit 370 outputs the determination result determined in step S142 (S143). The determination unit 370 outputs the determination result to, for example, the information processing device 300c. When the input information processing unit 390 of the information processing device 300c acquires the determination result (input information) from the determination unit 370 of the server device 500c, the input information processing unit 390 causes, for example, the display device 800 to display the determination result.
[0242] The determination unit 370 may directly output the determination result to the display device 800. The determination unit 370 may also output the determination result to the terminal device 600. The determination unit 370 may further determine a health improvement suggestion to be proposed to the driver based on the determination result, and output the determined health improvement suggestion to the terminal device 600, for example.
[0243] (Embodiment 5) The data generation system according to this embodiment will be described below with reference to Fig. 30 and Fig. 31. Note that the following description will focus on differences from the third embodiment, and descriptions of contents that are the same as or similar to the third embodiment will be omitted or simplified.
[0244] [5-1. Data generation system configuration] First, the configuration of the data generation system according to the present embodiment will be described with reference to Fig. 30. Fig. 30 is a diagram showing the configuration of a data generation system 1d according to the present embodiment. The data generation system 1d according to the present embodiment differs from the data generation system 1b according to the third embodiment in that the data generation system 1d according to the present embodiment includes a voice device 410 and an emergency call device 420 as an example of a notification device 400.
[0245] 30, the data generation system 1d includes an object measurement sensor 100, an external factor measurement sensor 200, an information processing device 300b, an audio device 410, an emergency call device 420, a server device 500, a terminal device 600, an input device 700, and a display device 800. The data generation system 1d is characterized in that the method of notifying the determination result of the state of the object in the information processing device 300b is changed depending on the determination result.
[0246] The audio device 410 is connected to the information processing device 300b and outputs audio corresponding to the determination result of the determination unit 370 of the information processing device 300b. The audio device 410 is mounted on, for example, a vehicle and outputs audio corresponding to the determination result to the driver. The audio device 410 is realized by, for example, a speaker. Note that the data generation system 1d may have, instead of or together with the audio device 410, a vibration device that stimulates the driver by vibration, a fragrance generating device that stimulates the driver by fragrance, and a light emitting device that stimulates the driver by light. The audio device 410 is an example of a second notification unit.
[0247] The emergency call device 420 is connected to the information processing device 300b, and makes an emergency call according to the determination result of the determination unit 370 of the information processing device 300b. The emergency call device 420 is a device for a so-called help net, and makes an emergency call to, for example, a specialized operator. For example, when the driver has difficulty driving, the emergency call device 420 automatically notifies the operator of information indicating that the driver has difficulty driving. The emergency call device 420 may also automatically notify the operator of information such as the position of the vehicle. The emergency call device 420 is an example of a first notification unit.
[0248] The voice device 410 and the emergency call device 420 may be included in the information processing device 300b. Although the data generation system 1d has been described as having two types of notification devices, it may have three or more types of notification devices. The data generation system 1d may have different notification devices for each type of determination made by the determination unit 370.
[0249] [5-2. Operation of the data generation system] Next, the operation of the data generating system 1d configured as above will be described with reference to FIG. 31. FIG. 31 is a flowchart showing the operation of the data generating system 1d according to this embodiment. In FIG. 31, the first judgment unit 371 performs a deadman judgment of the driver as the first judgment, the second judgment unit 372 performs a fatigue judgment of the driver as the second judgment, and the third judgment unit 373 performs a disease judgment of the driver as the third judgment. In addition, whether or not the driver is deadman is an example of the first state of the driver, the state of fatigue is an example of the second state, and whether or not the driver is sick is an example of the third state. For example, the second state and the third state are states that are less dangerous to the driver than the first state.
[0250] Operations that are the same as or similar to those in FIG. 26 are given the same reference numerals as in FIG. 26, and descriptions thereof will be omitted or simplified.
[0251] 31, when there is data where the accuracy of the data with accuracy information is greater than the threshold value P1 (Yes in S31), the first determination unit 371 extracts data having an accuracy higher than the threshold value P1 from the target data (measurement data) (S32) and performs a deadman determination on the driver (S33d). For example, the first determination unit 371 may calculate the driver's heart rate based on the extracted portion of the determination data, and perform the determination in step S33d based on the calculated heart rate.
[0252] Next, if there is something wrong with the driver (Yes in S34d), the first determination unit 371 performs emergency call processing (S35d). The emergency call processing is, for example, processing of operating at least the emergency call device 420 out of the audio device 410 and the emergency call device 420 to give a notification. For example, the first determination unit 371 outputs information indicating that there is something wrong with the driver to the emergency call device 420.
[0253] The emergency call device 420 makes an emergency call when the first determination unit 371 determines that the first state is abnormal. The emergency call device 420, for example, automatically notifies an operator that the first state is abnormal.
[0254] Furthermore, if there is no abnormality in the driver (No in S34d), the first determination unit 371 proceeds to step S36. Note that, if Yes in step S34d, the processes in and after step S36 may not be performed.
[0255] Next, if there is data where the accuracy of the data with accuracy information is greater than the threshold value P2 (Yes in S36), the second determination unit 372 extracts data with accuracy higher than the threshold value P2 from the target data (measurement data) (S37) and performs a fatigue determination for the driver (S38d). The second determination unit 372 may, for example, calculate the driver's heart rate based on the extracted portion of the determination data, and perform the determination in step S38d based on the calculated heart rate.
[0256] Next, if the driver is fatigued (Yes in S39d), the second determination unit 372 performs voice processing (S40d). The voice processing is, for example, processing for reducing the driver's fatigue by operating at least the voice device 410 out of the voice device 410 and the emergency call device 420. The second determination unit 372 outputs, for example, information indicating that the driver is fatigued to the voice device 410.
[0257] For example, when the second determination unit 372 determines that the second state is abnormal, the audio device 410 outputs a sound. For example, the audio device 410 may output a sound encouraging the driver to take a rest. Furthermore, when the data generation system 1d includes a vibration device, in step S40d, vibrations that reduce fatigue may be applied to the driver.
[0258] Moreover, if the driver is not fatigued (No in S39d), the second determination unit 372 proceeds to step S41.
[0259] Next, if there is data where the accuracy of the data with accuracy information is greater than the threshold value P3 (Yes in S41), the third determination unit 373 extracts data with accuracy higher than the threshold value P3 from the target data (S42) and performs illness determination for the driver (S43d). The second determination unit 372 may, for example, determine whether the driver has a predetermined illness (for example, a highly urgent illness such as high blood pressure or myocardial infarction) based on the extracted part of the determination data.
[0260] Next, the third determination unit 373 performs voice processing (S45d) when the driver (chauffeur) has a highly urgent illness (Yes in S44d). The voice processing is, for example, processing for notifying the driver of a highly urgent illness by operating at least the voice device 410 out of the voice device 410 and the emergency call device 420. The third determination unit 373 outputs, for example, information indicating that the driver has a highly urgent illness to the voice device 410.
[0261] For example, when the third state is determined to be abnormal by the third determination unit 373, the audio device 410 outputs a sound. For example, the audio device 410 may output a sound to prompt the user to stop the vehicle or to go to a hospital.
[0262] Moreover, if the driver does not have an urgent illness (No in S44d), the third determination unit 373 ends the determination process.
[0263] As described above, the method of notifying the determination result may be appropriately selected depending on the contents of the determination made by the determination unit 370 regarding the state of the object.
[0264] Also, as shown in FIG. 31, the information processing device 300b can make a judgment on the state of a plurality of different types of drivers by only generating one piece of data with accuracy information (for example, data with accuracy information generated in S18 shown in FIG. 21). The information processing device 300b can make a judgment on the state of a plurality of types of drivers with different data accuracy (for example, thresholds P1, P2, and P3) required for judgment, for example, from one piece of data with accuracy information. For example, the information processing device 300b can make both a judgment that can be made even if the data accuracy is low, such as the presence or absence of a heartbeat for detecting dead man, and a judgment that can be made only if the data accuracy is high, such as whether or not the driver is ill, including a symptom of heart failure, based on one piece of data with accuracy information. In other words, the accuracy-attached information generating unit 310b only needs to generate one piece of data with accuracy information to make a judgment on the state of a plurality of types of drivers. This can reduce the amount of processing in the accuracy-attached information generating unit 310b, compared to, for example, a case in which data corresponding to each judgment is generated for each of a plurality of types of judgments.
[0265] (Other embodiments) Although the data generating device according to one or more aspects has been described based on each embodiment, the present disclosure is not limited to each of these embodiments. As long as it does not deviate from the gist of the present disclosure, various modifications conceived by a person skilled in the art to the present embodiment and forms constructed by combining components in different embodiments may also be included in the present disclosure.
[0266] For example, the object in each of the above embodiments is not limited to a driver, and may be any object. The object may be a person, a specific device, or the like. The object may be a moving object, or a stationary device, or the like. Furthermore, the object on which at least a part of the data generation system is mounted is not limited to a vehicle, and may be an air vehicle such as a drone, or a robot capable of autonomous driving. Furthermore, the object may be a moving object, or a stationary object. Note that the vehicle includes an automobile, a train, and the like.
[0267] In the above embodiment, the object measurement sensor and the external factor measurement sensor are non-contact sensors, but the present invention is not limited to this. At least one of the object measurement sensor and the external factor measurement sensor may be a contact sensor. The object measurement sensor may be, for example, a wearable terminal worn by the driver.
[0268] In addition, in the above embodiments, examples have been described in which the noise level and accuracy are expressed by numerical values, but this is not limited to this, and they may also be expressed by stages such as "high," "medium," and "low."
[0269] In addition, in the above embodiment, the third sensor is exemplified as a luminance sensor, but is not limited to having a luminance sensor. When the object measurement sensor is a camera, the information processing device may, for example, perform luminance conversion based on an image captured by the object measurement sensor to obtain the intensity of external light (intensity of an external factor).
[0270] In the above embodiment and the like, the first sensor and the second sensor are described as being mounted on a vehicle, but the present invention is not limited thereto. The first sensor and the second sensor may be mounted on an aircraft such as a drone, or may be mounted indoors. The first sensor and the second sensor may be mounted on a home appliance, for example. The first sensor and the second sensor may be worn by a person, for example.
[0271] In the above embodiment, the time delay correction unit corrects the time delay of the noise level data of the external factor, but the time delay of the external factor data may be corrected. Then, the noise information calculation unit may convert the external factor data with the time delay corrected into a noise level.
[0272] In the above embodiments, each component may be implemented by dedicated hardware or by executing a software program suitable for each component. Each component may be implemented by a program execution unit such as a CPU or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory.
[0273] In addition, the order in which each step is performed in the flowchart is merely an example for specifically explaining the present disclosure, and an order other than the above may be used. In addition, some of the steps may be performed simultaneously (in parallel) with other steps, or some of the steps may not be performed.
[0274] In addition, the division of functional blocks in the block diagram is an example, and multiple functional blocks may be realized as one functional block, one functional block may be divided into multiple blocks, or some functions may be transferred to another functional block. Furthermore, the functions of multiple functional blocks having similar functions may be processed in parallel or in a time-sharing manner by a single piece of hardware or software.
[0275] Furthermore, the data generation system or data generation device according to the above-mentioned embodiments may be realized as a single device or may be realized by multiple devices. When the data generation system or data generation device is realized by multiple devices, each component of the data generation system or data generation device may be distributed in any manner among the multiple devices. When the data generation system or data generation device is realized by multiple devices, the communication method between the multiple devices is not particularly limited, and may be wireless communication or wired communication. Furthermore, wireless communication and wired communication may be combined between the devices.
[0276] Moreover, each of the components described in the above embodiments may be realized as software, or may be realized as an LSI, which is typically an integrated circuit. These may be individually integrated into one chip, or may be integrated into one chip to include some or all of them. Here, LSI is used, but it may be called IC, system LSI, super LSI, or ultra LSI depending on the degree of integration. Furthermore, the method of integration is not limited to LSI, and may be realized by a dedicated circuit or a general-purpose processor. After LSI manufacture, a programmable FPGA (Field Programmable Gate Array) or a reconfigurable processor that can reconfigure the connection or setting of the circuit cells inside the LSI may be used. Furthermore, if an integrated circuit technology that replaces LSI appears due to the progress of semiconductor technology or a different derived technology, it is natural that the integration of the components may be performed using that technology.
[0277] A system LSI is an ultra-multifunctional LSI manufactured by integrating multiple processing units on a single chip, and specifically, is a computer system that includes a microprocessor, a ROM (Read Only Memory), a RAM (Random Access Memory), etc. Computer programs are stored in the ROM. The system LSI achieves its functions by the microprocessor operating in accordance with the computer program.
[0278] Another aspect of the present disclosure may be a computer program causing a computer to execute each of the characteristic steps included in the data generation method shown in any of Figures 3, 6, 10, 14, 17, 21, 22, 24 to 26, 28, and 31. Another aspect of the present disclosure may be a computer program causing a computer to execute each of the characteristic steps included in the information processing method of the server device shown in Figure 23.
[0279] Also, for example, the program may be a program to be executed by a computer. Also, one aspect of the present disclosure may be a computer-readable non-transitory recording medium on which such a program is recorded. For example, such a program may be recorded on a recording medium and distributed or circulated. For example, the distributed program may be installed in a device having another processor, and the program may be executed by the processor, thereby making it possible to cause the device to perform each of the above processes. [Industrial Applicability]
[0280] The present disclosure is useful for devices and the like that generate data that measure an object. [Explanation of symbols]
[0281] 1, 1b, 1d Data Generation System 100 Object measurement sensor (first sensor) 200 Sensor for measuring external factors (second sensor) 300, 300a, 300b, 300c Information processing device 310, 310a, 310b Precision information generating unit (data generating device) 320 First acquisition part 330 Analysis Department 340 Second Acquisition Department 341 Third Acquisition Department 342 4th Acquisition Department 350, 350b Accuracy information generation section (accuracy calculation section) 351 Noise information calculation unit (noise level calculation unit) 352 Time Delay Correction Unit (Delay Correction Unit) 353 Accuracy Information Calculation Department 354 Coefficient update unit (first update unit) 360 Accuracy Information Integration Unit (Data Output Unit) 370 Judgment unit (data output unit, second update unit) 371 1st Judgment Department 372 Second Judgment Department 373 Third Judgment Department 380 Output section 390 Input information processing unit (authentication result acquisition unit) 400 Notification device 410 Audio device (second notification unit) 420 Emergency Call Device (First Notification Section) 500, 500c server equipment 510 Personal authentication section 520 Judgment method update section 600 Terminal Equipment 700 Input Device 800 display device
Claims
1. A data generating device that generates judgment data for judging a state of an object, a first acquisition unit that acquires measurement data obtained by measuring the object from a first sensor; a second acquisition unit that acquires first factor data in which a first noise factor that may become noise in the measurement data is measured from a second sensor different from the first sensor; an accuracy calculation unit that calculates a measurement accuracy of the measurement data based on the first factor data; a data output unit that outputs the judgment data in which the measurement data is associated with accuracy information indicating the measurement accuracy; a noise level calculation unit that calculates the noise level from the first factor data acquired by the second acquisition unit based on a first table in which the first factor data and a noise level that the first factor data imparts to the measurement data are associated with each other; a determination unit that determines the state of the object based on the determination data output from the data output unit, the accuracy calculation unit calculates the measurement accuracy of the measurement data acquired by the first acquisition unit from the noise level calculated by the noise level calculation unit, using a second table in which the noise level and the measurement accuracy are associated with each other; the determination unit includes a first determination unit that determines a first state of the object based on a first accuracy, and a second determination unit that determines a second state different from the first state based on a second accuracy, the data output unit generates and outputs the same judgment data to the first judgment unit and the second judgment unit; the second state being a state in which the danger to the object is less than that of the first state, The second accuracy is less accurate than the first accuracy. Data generation device.
2. the measurement data and the first factor data are time-series data measured over a first period of time; a delay correction unit that corrects a time delay between the measurement data and the first factor data, The data output unit outputs the determination data in which the time delay has been corrected and the measurement data and the accuracy information are associated with each other. The data generating device according to claim 1 .
3. a third acquisition unit that acquires second factor data, in which a second noise factor that may become noise in the measurement data is measured, from a third sensor different from the first sensor and the second sensor; The accuracy calculation unit further calculates the measurement accuracy based on the second factor data.
3. The data generating device according to claim 1 or 2.
4. the second factor data is time-series data measured over a second period at least a portion of which overlaps with the first period; The delay correction unit further corrects a time delay between the measurement data and the second factor data. A data generating device according to claim 3 dependent on claim 2.
5. The method further includes the steps of: updating at least one of the first table and the second table; The data generating device according to any one of claims 1 to 4.
6. the object is a person, The apparatus further includes an authentication result acquisition unit that acquires an authentication result of the person. The data generating device according to any one of claims 1 to 5.
7. The method further includes updating at least one of a determination method of the determination unit and a threshold value used for the determination, the method being determined based on the authentication result. The data generating device according to claim 6.
8. The first determination unit extracts a first portion that satisfies the first accuracy from the measurement data based on the first accuracy corresponding to the determination of the first state, and determines the first state based on the extracted first portion; The second determination unit extracts a second portion that satisfies the second accuracy from the measurement data based on the second accuracy corresponding to the determination of the second state, and determines the second state based on the extracted second portion. The data generating device according to any one of claims 1 to 7.
9. A first notification unit that issues an emergency call when the first determination unit determines that the first state is abnormal; a second notification unit that outputs a sound when the second determination unit determines that the second state is abnormal. The data generating device according to any one of claims 1 to 8.
10. The measurement data and the first factor data are time-series data measured over a first period of time, The calculated measurement accuracy is time-series data over the first period including multiple levels of accuracy. The data generating device according to any one of claims 1 to 9.
11. The data generating device is mounted on a vehicle, the object is a driver of the vehicle; The first determination unit determines whether or not the driver has difficulty driving the vehicle, The second determination unit determines whether the driver is dozing off or looking away. The data generating device according to any one of claims 1 to 10.
12. The data generating device is mounted on a vehicle, the object is a driver of the vehicle; The first determination unit determines a drowsiness of the driver, The second determination unit determines an emotion of the driver or whether the driver is ill. The data generating device according to any one of claims 1 to 10.
13. the first sensor is a millimeter wave sensor or an infrared camera; The second sensor is an acceleration sensor that measures the acceleration of the vehicle.
13. The data generating device according to claim 11 or 12.
14. The data generating device is mounted on a vehicle, the first sensor is an infrared camera; the second sensor is an acceleration sensor that measures an acceleration of the vehicle; The third sensor is a luminance sensor that measures the luminance around the vehicle. The data generating device according to claim 3.
15. Obtaining information identifying the first noise factor from the first sensor. The data generating device according to any one of claims 1 to 14.
16. A data generation method for generating judgment data for judging a state of an object, comprising the steps of: Acquiring measurement data of the object from a first sensor; acquiring first factor data in which a first noise factor that may become noise in the measurement data is measured from a second sensor different from the first sensor; Calculating a measurement accuracy of the measurement data based on the first factor data; outputting the judgment data in which the measurement data is associated with accuracy information indicating the measurement accuracy; calculating the noise level from the acquired first factor data based on a first table in which the first factor data and a noise level that the first factor data affects the measurement data are associated with each other; determining the state of the object based on the output determination data; In the calculation of the measurement accuracy, a second table in which the noise level and the measurement accuracy are associated with each other is used to calculate the measurement accuracy of the measurement data acquired from the calculated noise level; Determining the state of the object includes determining a first state of the object based on a first accuracy and determining a second state of the object different from the first state based on a second accuracy; In the output of the judgment data, the same judgment data is generated and output to a first judgment unit that judges a first state of the object and a second judgment unit that judges a second state different from the first state, the second state being a state in which the danger to the object is less than that of the first state, The second accuracy is less accurate than the first accuracy. Data generation method.
17. A program causing a computer to execute the data generating method according to claim 16.
Citation Information
Patent Citations
Photoelectric biometric apparatus
JP2010252875A
Biological signal detection device and biological signal detection method
JP2019170967A
Information processing device, moving apparatus, method, and program
WO2019188398A1
Biological measurement device and biological measurement method
WO2020044854A1
Pulse wave detection device and pulse wave detection program
WO2020203914A1
