Estimation device, estimation system, method for estimation, and program
The estimation device addresses the challenge of discriminating trace components and foreign substances by using a sensor with a sensitive film to analyze waveform features during state transitions, improving detection accuracy.
Patent Information
- Application Number
- JP2023219140
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-12-26
- Publication Date
- 2025-07-08
AI Technical Summary
Existing methods struggle to accurately discriminate trace components and foreign substances in objects due to interference from external factors like temperature and humidity.
An estimation device that uses a sensor with a sensitive film that physically changes in response to substances generated from an object, acquiring signals during state transitions between exposure to a reference and the object, and analyzing waveform features to estimate the object's state, including the presence of foreign substances and deterioration.
Enhances the ability to detect trace components and foreign substances by minimizing environmental interference, allowing for precise identification of object characteristics.
Smart Images
Figure 2025102001000001_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an estimation device, an estimation system, an estimation method, and a program.
Background Art
[0002] Patent Documents 1 to 3 describe outputting a quality value or fermentation status information of a solid fermented product based on an odor detection value of the solid fermented product. Patent Document 4 describes a monitoring system for monitoring whether a monitoring target substance is present in the air and a detection device used in the monitoring system. [Prior Art Documents] [Patent Documents] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2022-157743 [Patent Document 2] Japanese Unexamined Patent Application Publication No. 2022-157744 [Patent Document 3] Japanese Unexamined Patent Application Publication No. 2022-157745 [Patent Document 4] International Publication No. 2018 / 225776
Summary of the Invention
Problems to be Solved by the Invention
[0003] It is desired to more accurately grasp the characteristics of an object based on the detection results of at least one substance generated from the object. However, due to external interference factors such as temperature or humidity, it is often difficult to discriminate foreign substances caused by trace components contained in small amounts in the object, for example.
Means for Solving the Problems
[0004] An estimation device according to an aspect of the present invention may include an acquisition unit that acquires a signal output from a sensor having a sensitive part that physically changes in response to at least one substance generated from an object and outputs a signal corresponding to the physical change, during a detection period from a first state in which the sensor is exposed to a reference substance, after switching to a second state in which the sensor is exposed to the object, and until switching back to the first state from the second state again. The estimation device may include an estimation unit that estimates the state of the object based on a feature amount of the waveform of the signal.
[0005] In the estimation device, the reference substance may be a substance of the same type as the object that satisfies predetermined conditions, or water.
[0006] In any of the above-described estimation devices, the sensor may have a plurality of sensitive parts having different characteristics of physical changes corresponding to the at least one substance generated from the object. The acquisition unit may acquire a plurality of signals output from each of the plurality of sensitive parts during the detection period. The estimation unit may estimate the state of the object based on a combination of feature amounts of waveforms of at least two of the plurality of signals.
[0007] In any of the above-described estimation devices, the sensitive part may include a sensitive film that deforms when the at least one substance is adsorbed and diffuses therein.
[0008] In any of the above-described estimation devices, the sensitive film may include an organic-inorganic hybrid material.
[0009] In any of the above-described estimation devices, the feature amount of the waveform of the signal may include at least one of each amplitude of a plurality of divided waveforms obtained by dividing the waveform of the signal at predetermined intervals, the sum of the amplitudes of the plurality of divided waveforms, each change rate of the amplitudes of the plurality of divided waveforms, the sum of the change rates of the amplitudes of the plurality of divided waveforms, and the average value of the change rates of the amplitudes of the plurality of divided waveforms.
[0010] In any of the above-described estimation devices, the combination of the feature amounts of the waveforms of the at least two signals may include at least one of the ratios of the amplitudes of the plurality of divided waveforms of each of the at least two signals obtained by dividing the waveforms of the at least two signals at predetermined intervals, the ratios of the total values of the amplitudes of the plurality of divided waveforms of each of the at least two signals, the ratios of the change rates of the amplitudes of the plurality of divided waveforms of each of the at least two signals, the ratios of the total of the change rates of the amplitudes of the plurality of divided waveforms of each of the at least two signals, and the ratios of the average values of the amplitudes of the plurality of divided waveforms of each of the at least two signals.
[0011] In any of the above-described estimation devices, the object may be food.
[0012] In any of the above-described estimation devices, the object may be a liquid substance.
[0013] In any of the above-described estimation devices, the estimation unit may estimate whether or not a foreign object is mixed in the object as the state of the object.
[0014] In any of the above-described estimation devices, when the estimation unit estimates that a foreign object is mixed in the object, the estimation unit may further estimate the type of the foreign object based on the relationship information indicating the relationship between the feature amount of the signal waveform and the type of the foreign object and the signal.
[0015] In any of the above-described estimation devices, the estimation unit may estimate whether or not the object is in a deteriorated state as the state of the object.
[0016] In any of the above-described estimation devices, when the estimation unit estimates that the object is in a deteriorated state, the estimation unit may further estimate the degree of the deteriorated state of the object based on the relationship state indicating the relationship between the feature amount of the signal waveform and the degree of the deteriorated state of the object and the signal.
[0017] An estimation system according to one aspect of the present invention includes any one of the above-described estimation devices, the sensor, and a switching mechanism that communicates with the space inside the housing portion that houses the object and switches between the first state in which the sensor is exposed to the reference object and the second state in which the sensor is exposed to the object.
[0018] In the system, the switching mechanism may include a first pipe that communicates with the space inside the housing portion, a second pipe that communicates with the space outside the housing portion, and a third pipe that is switchably communicable with one of the first pipe and the second pipe. The sensor may be disposed inside the third pipe. When the third pipe communicates with the second pipe, the sensor may enter the first state, and when the third pipe communicates with the first pipe, the sensor may enter the second state.
[0019] An estimation method according to one aspect of the present invention may include a step of acquiring a signal output from a sensor having a sensitive portion that physically changes in response to at least one substance generated from an object and outputs a signal corresponding to the physical change, during a detection period from when the sensor switches from a first state of being exposed to a reference object to a second state of being exposed to the object and then switches back from the second state to the first state again. The estimation method may include a step of estimating the state of the object based on a feature amount of the waveform of the signal.
[0020] A program according to one aspect of the present invention, when executed by a computer, causes the computer to acquire a signal output from a sensor having a sensitive portion that physically changes in response to at least one substance generated from an object and outputs a signal corresponding to the physical change, during a detection period from when the sensor switches from a first state of being exposed to a reference object to a second state of being exposed to the object and then switches back from the second state to the first state again, and to estimate the state of the object based on a feature amount of the waveform of the signal.
[0021] Note that the above summary of the invention does not enumerate all the features of the present invention. Also, sub - combinations of these feature groups can also be inventions.
Brief Description of Drawings
[0022]
Figure 1
Figure 2
Figure 3A
Figure 3B
Figure 3C
Figure 3D
Figure 3E
Figure 3F
Figure 3G
Figure 3H
Figure 4A
Figure 4B
Figure 4C
Figure 4D
Figure 4E
Figure 4F
Figure 4G
Figure 4H
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11A
Figure 11B
Figure 12A
Figure 12B
Figure 12C
Figure 12D
Figure 12E
Figure 12F
Figure 12G
Figure 12H
Figure 13A
Figure 13B
Figure 13C
Figure 13D
Figure 13E
Figure 13F
Figure 13G
Figure 13H
Figure 14
Figure 15
Figure 16
Figure 17
Embodiments for Carrying Out the Invention
[0023] Hereinafter, the present invention will be described through embodiments of the invention. However, the following embodiments do not limit the invention according to the claims. Also, not all combinations of features described in the embodiments are essential for the solution means of the invention.
[0024] FIG. 1 is a diagram showing an example of a functional block of the overall configuration of the estimation system 10 according to the present embodiment. The estimation system 10 estimates the state of the object based on the detection results of at least one substance generated from the object. The object is, for example, food. The object may be a liquid substance. That is, the object may be a liquid. The estimation system 10 may estimate whether a foreign object is mixed in the object as the state of the object. The estimation system 10 may estimate whether the object is in a deteriorated state as the state of the object.
[0025] The estimation system 10 includes a container 20, a container 22, a switching mechanism 30, a sensor 50, and an estimation device 100. The container 20 houses an object such as a beverage or food. The container 22 houses a reference substance which is a substance detected by the sensor 50 for causing the sensor 50 to output a signal indicating a reference for estimating the state of the object by the sensor 50. The reference substance may be a substance that satisfies a predetermined condition that when exposed to the sensor 50 instead of the object, according to the characteristics of the object, the type of the object, whether the object is a gas, a liquid, or a solid, etc., there is no change in the environment where the sensor 50 is located, such as humidity or temperature, or the amount of change falls within a predetermined range. The reference substance may be, for example, water or a substance of the same type as the object in a normal state that satisfies a predetermined condition.
[0026] The switching mechanism 30 is connected to the container 20 and the container 22. The switching mechanism 30 has a pipe 31 communicating with the internal space of the container 20, a pipe 32 communicating with the internal space of the container 22, and a pipe 33 connected to the pipe 31 and the pipe 32 via a switching valve 34 and communicating with either one of the pipe 31 and the pipe 32. The switching valve 34 includes a valve that communicates between the pipe 33 and the container 20 or the container 22. The gas existing in the internal space of the container 20 passes through the pipe 31. The gas existing in the internal space of the container 22 passes through the pipe 32. The gas existing in the internal space of the container 20 contains at least one substance generated from the object. The gas existing in the internal space of the container 22 contains at least one substance generated from the reference substance.
[0027] The amount or type of the generated gas differs depending on the substance constituting the object. Therefore, it is possible to estimate the state of the object by specifying at least one of the type and amount of the generated gas.
[0028] The sensor 50 is provided inside the pipe 33 and detects the gas passing through the pipe 33. The sensor 50 may be a so-called olfactory sensor. The configuration of the switching mechanism 30 shown above is only an example. The switching mechanism 30 may be any mechanism as long as it can switch between a detection preparation state (first state) in which the sensor 50 is exposed to at least one substance generated from a reference substance and a detection state (second state) in which the sensor 50 is exposed to at least one substance generated from a target substance. The switching mechanism 30 may have an injection unit that purges the gas around the sensor 50 by injecting a gas containing at least one substance generated from the reference substance in the container 22 inside the pipe 33. The switching mechanism 30 may have a suction mechanism that sucks gas into the pipe 33 so that the gas present in the internal space of the container 20 or container 22 is efficiently supplied to the pipe 33.
[0029] The sensor 50 may be a film surface type stress sensor that functions as an olfactory sensor. The film surface type stress sensor has a sensitive part that physically changes in response to at least one substance generated from a reference substance or a target substance and outputs a signal corresponding to the physical change. The sensitive part includes a sensitive film that deforms when at least one substance adsorbs and diffuses inside. The sensitive film may include an organic-inorganic hybrid material.
[0030] The organic-inorganic hybrid includes a structure represented by RSiO 3 / 2 where "R" represents an organic functional group.
[0031] In this embodiment, it is preferable that the organic functional group contains one or more aromatic rings (aromatic ring structures). When an aromatic ring is included, the moisture resistance tends to be further improved. The aromatic ring is not particularly limited because it can be appropriately selected in consideration of the use of the sensor and the like. For example, aromatic hydrocarbon groups such as phenyl group, naphthyl group, p-tolyl group, biphenyl group, substituted aromatic hydrocarbon groups such as 4-chlorophenyl group, 4-methoxyphenyl group, 4-aminophenyl group, pentafluorophenyl group, heterocyclic hydrocarbon groups such as 3-furyl group, 3-thienyl group, 2-pyridyl group, 3-pyridyl group, 4-pyridyl group, and metallocenes such as ferrocenyl group can be mentioned. The organic functional group in this embodiment may contain one of the above-described aromatic rings alone, or may contain two or more thereof in combination.
[0032] An environmental sensor 40 may be provided in the containers 20 and 22. The environmental sensor 40 detects the environmental state in the container 20 or the container 22. The environmental sensor 40 may detect temperature and humidity as the environmental state in the container 20 or the container 22, and provide environmental information including temperature information and humidity information to the estimation device 100.
[0033] FIG. 2 shows an example of the functional blocks of the estimation device 100. The estimation device 100 includes a control unit 110 and a storage unit 120. The control unit 110 may be configured by a central processing unit (CPU).
[0034] The estimation device 100 may be configured by a computer. The computer may be a personal computer, a tablet computer, a smartphone, a workstation, a server computer, or a general-purpose computer, or may be a computer system to which a plurality of computers are connected. Such a computer system is also a computer in a broad sense. The computer may be a dedicated computer designed for the estimation process of the estimation device 100, or may be dedicated hardware realized by a dedicated circuit. The computer may be implemented in a virtual computer environment. When a computer is used, the estimation device 100 is realized by executing a program on the computer.
[0035] The control unit 110 includes an acquisition unit 112, an estimation unit 114, and a notification unit 116. The acquisition unit 112 acquires a signal output from the sensor 50 during a detection period from when the sensor 50 switches from a detection preparation state of being exposed to a reference object to a detection state of being exposed to an object, until it switches back from the detection state to the detection preparation state.
[0036] The sensor 50 may have a plurality of sensitive parts with different characteristics of physical changes corresponding to at least one substance generated from the reference object or the object. The plurality of sensitive parts may have a plurality of sensitive films with different characteristics. The sensor 50 may have a plurality of channels for outputting signals from each of the plurality of sensitive parts.
[0037] Depending on the state of the object, the amount and type of gas generated are different. Therefore, the signals output from the plurality of channels of the sensor 50 are also different. That is, depending on the state of the object, the feature amounts of the waveforms of the respective signals are different.
[0038] Here, in the detection preparation state, it is conceivable to expose the sensor 50 to air. However, the sensor 50 is sensitive to temperature or humidity, and even when exposed to the same substance, a change in temperature or humidity causes a change in the output signal. Therefore, for example, when the object is a liquid substance such as a beverage, when the sensor 50 switches from a state of being exposed to air to a state of being exposed to the object, the humidity in the space where the sensor 50 is located changes greatly, which may greatly affect the signal output from the sensor 50. In such a case, it may be difficult to discriminate foreign substances due to trace components contained in a small amount in the object from the signal output from the sensor 50.
[0039] Figures 3A to 3H show the waveforms of the signals of each channel when, using a film surface stress sensor composed of an array of 8 types (8 channels) of sensitive films as the sensor 50, the sensor 50 is switched from the detection preparation state exposed to air with the sensor 50 as a reference object to the detection state exposed to each of water, orange juice, and orange juice mixed with 100 ppm of ethanol (which may be referred to as "orange juice mixed with ethanol").
[0040] The waveforms of the signals shown in Figures 3A to 3H are the waveforms of the signals output by each channel of the sensor 50 when, after exposing the sensor 50 to air as a reference object from 0 seconds to 120 seconds in the detection preparation state, the sensor 50 is exposed to each of water, orange juice, and orange juice mixed with ethanol as the object to be detected from 120 seconds to 240 seconds in the detection state, repeated for 30 minutes in a 2-minute cycle, and normalized to 0 mV with the starting point at 0 seconds.
[0041] As shown in Figures 3A to 3H, there are channels where differences appear between the waveform of the water signal and the waveform of the orange juice signal. However, almost no difference appears between the waveform of the orange juice signal and the waveform of the orange juice mixed with ethanol signal. Therefore, when using air as a reference object, it is difficult to estimate from the signal of the sensor 50 whether trace amounts of ethanol are contained in the orange juice.
[0042] Figures 4A to 4H show the waveforms of the signals of each channel when, using a film surface stress sensor composed of an array of 8 types (8 channels) of sensitive films as the sensor 50, the state where the sensor 50 is exposed to orange juice mixed with ethanol as the object to be detected and the state where the sensor 50 is exposed to orange juice as a reference object are alternately switched.
[0043] The waveforms of the signals shown in FIGS. 4A to 4H are the waveforms of the signals output by each channel of the sensor 50 when the sensor 50 is exposed to orange juice mixed with ethanol from 0 seconds to 120 seconds, and then the state of the sensor 50 being exposed to orange juice is repeated for 30 minutes at a 2-minute cycle from 120 seconds to 240 seconds, and normalized to 0 mV with the 0-second point as the starting point.
[0044] In the waveforms of the signals shown in FIGS. 4A to 4H, when the sensor 50 is exposed to orange juice mixed with ethanol, the amplitudes of the respective signals tend to increase, and when the sensor 50 is exposed to orange juice, the amplitudes of the respective signals tend to decrease. That is, it can be seen that there is a clear difference in the signals output by each channel of the sensor 50 between the state where the sensor 50 is exposed to orange juice mixed with ethanol and the state where the sensor 50 is exposed to orange juice. That is, the difference from the signals of each channel of the sensor 50 when exposed to orange juice mixed with ethanol appears more easily when orange juice is used than when air is used as the reference substance.
[0045] FIG. 5 shows the difference between the amplitudes of the signals of each channel of the sensor 50 when exposed to each reference substance and the amplitudes of the signals of each channel of the sensor 50 when exposed to orange juice mixed with ethanol. As shown in FIG. 5, when the reference substance is air, the amplitude corresponding to ethanol becomes negative, and the mixing of ethanol cannot be detected from the signals of each channel of the sensor 50. This is due to a difference in the humidity of the space to which the sensor 50 is exposed between the case of air and the case of orange juice mixed with ethanol, and the influence appears in the signals of each channel of the sensor 50. On the other hand, when the reference substance is water or orange juice, which is the same substance as the object, there is little difference in the humidity of the space to which the sensor 50 is exposed, and the mixing of ethanol can be detected from the signals of each channel of the sensor 50 even if the amount is very small.
[0046] Therefore, in this embodiment, in order to suppress changes in humidity between the detection preparation state and the detection state, in the detection preparation state, the sensor 50 is exposed to a reference substance that is not air. The reference substance is water or a substance of the same type as the object in a normal state that satisfies predetermined conditions. If the predetermined conditions are that the object is food manufactured in a plurality of factories, the reference substance may be, for example, the same food as the object manufactured in a reference factory.
[0047] In this way, by suppressing changes in humidity between the detection preparation state and the detection state, the change in the signal of the sensor 50 caused by switching from the detection preparation state to the detection state is likely to be due to at least one substance generated from the object, rather than the influence of changes in humidity. That is, it becomes easier to determine the presence of foreign substances due to trace components contained in small amounts in the object from the signal output from the sensor 50. Also, it becomes easier to determine minute components generated from the object due to deterioration of the object from the signal output from the sensor 50. Note that even when there are changes in temperature between the detection preparation state and the detection state, a large difference occurs in the signal output from the sensor 50. Therefore, it is preferable to suppress changes in temperature between the detection preparation state and the detection state.
[0048] Therefore, the acquisition unit 112 acquires the signal output from the sensor 50 in the detection period from when the sensor 50 switches from the detection preparation state in which it is exposed to a reference substance that is a substance of the same type as the object that satisfies predetermined conditions or water, to the detection state in which it is exposed to the object, until it switches again from the detection preparation state to the detection state. The estimation unit 114 estimates the state of the object based on the feature amount of the waveform of the signal from the sensor 50.
[0049] The estimation unit 114 may estimate whether or not a foreign substance is mixed in the object as the state of the object. When the estimation unit 114 estimates that a foreign substance is mixed in the object, the type of the foreign substance may be further estimated based on the relationship information indicating the relationship between the feature amount of the waveform of the signal and the type of the foreign substance, and the signal from the sensor 50.
[0050] The estimation unit 114 may estimate, as the state of the object, whether the object is in a deteriorated state. When estimating that the object is in a deteriorated state, the estimation unit 114 may further estimate the degree of the deteriorated state of the object based on a relationship state indicating a relationship between the feature amount of the signal waveform and the degree of the deteriorated state of the object, and the signal from the sensor 50.
[0051] FIG. 6 shows an example of the waveform of a signal output from one channel (channel 1) of the sensor 50. The estimation unit 114 may derive the amplitudes s1 to s5 of the multiple divided waveforms by dividing the waveform of the signal S1 at a predetermined interval t. The estimation unit 114 may derive the sum (s1+s2+s3+s4+s5=s_total) of the amplitudes s1 to s5 of the multiple divided waveforms. The estimation unit 114 may derive the change rates (s2-s1) / t=Δs1, (s3-s2) / t=Δs2, (s4-s3) / t=Δs3, and (s5-s4)t=Δs4 of the amplitudes s1 to s5 of the multiple divided waveforms. The estimation unit 114 may derive the sum (Δs1+Δs2+Δs3+Δs4) of the amplitudes of the multiple divided waveforms. The estimation section 114 may derive the average value ((Δs1+Δs2+Δs3+Δs4) / 4) of the rate of change of each amplitude of the multiple divided waveforms.
[0052] The acquisition unit 112 may acquire a plurality of signals output from each of the plurality of sensory units during the detection period. The estimation unit 114 may estimate that a foreign object is mixed into the object or that the object is deteriorated when the maximum amplitude of each of the amplitudes s1 to s5 of the plurality of divided waveforms is equal to or greater than a predetermined threshold. The estimation unit 114 may estimate that a foreign object is mixed into the object or that the object is deteriorated when the sum of the amplitudes s1 to s5 of the plurality of divided waveforms is equal to or greater than a predetermined threshold. The estimation unit 114 may estimate that a foreign object is mixed into the object or that the object is deteriorated when the sum of the respective change rates of the respective amplitudes of the plurality of divided waveforms is equal to or greater than a predetermined threshold. The estimation unit 114 may estimate that a foreign object is mixed into the object or that the object is deteriorated when the average value of the respective change rates of the respective amplitudes of the plurality of divided waveforms is equal to or greater than a predetermined threshold.
[0053] The estimation unit 114 may estimate that a foreign object is present in the object or that the object is deteriorated, based on a combination of waveform feature amounts of at least two signals among the plurality of signals.
[0054] The combination of features of the waveforms of the at least two signals may include at least one of a ratio between the amplitudes of each of a plurality of divided waveforms of the at least two signals obtained by dividing the waveforms of the at least two signals at a predetermined interval t, a ratio between the sums of the amplitudes of each of the plurality of divided waveforms of the at least two signals, a ratio between the rates of change of the amplitudes of each of the plurality of divided waveforms of the at least two signals, a ratio between the sums of the rates of change of the amplitudes of each of the plurality of divided waveforms of the at least two signals, and a ratio between the average values of the amplitudes of each of the plurality of divided waveforms of the at least two signals.
[0055] The estimation unit 114 may estimate that a foreign matter is present in the object or that the object is deteriorated when at least one of the ratio between the amplitudes of each of the multiple divided waveforms of at least two signals obtained by dividing the waveforms of at least two signals at a predetermined interval t, the ratio between the sums of the amplitudes of each of the multiple divided waveforms of at least two signals, the ratio between the rates of change of each of the amplitudes of each of the multiple divided waveforms of at least two signals, the ratio between the sums of the rates of change of each of the amplitudes of each of the multiple divided waveforms of at least two signals, and the ratio between the average values of the amplitudes of each of the multiple divided waveforms of at least two signals falls within a predetermined numerical range.
[0056] The estimation unit 114 may estimate a substance contaminating the object based on the multiple signals and relationship information indicating a relationship between a combination of waveform feature amounts of at least two signals among the multiple signals and a substance corresponding to a foreign object. The estimation unit 114 may estimate a substance contaminating the object based on the multiple signals and relationship information indicating a relationship between a combination of waveform feature amounts of at least two signals among the multiple signals and a degree of a deterioration state of the object.
[0057] FIG. 7 shows an example of the waveform of the signal S2 output from another channel (channel 2) of the sensor 50. The estimation unit 114 may derive, for example, the ratios (s1 / m1, s2 / m2, s3 / m3, s4 / m4, s5 / m5) of the amplitudes s1 to s5 of the divided waveforms of the signal S1 in FIG. 6 and the amplitudes m1 to m5 of the divided waveforms of the signal S2 in FIG. 7. The estimation unit 114 may derive the ratio (s_total / m_total) of the sum (s1 + s2 + s3 + s4 + s5 = s_total) of the amplitudes s1 to s5 of the divided waveform of the signal S1 and the sum (m1 + m2 + m3 + m4 + m5 = m_total) of the amplitudes m1 to m5 of the divided waveform of the signal 2. The estimation unit 114 may derive the ratios (Δs1 / Δm1, Δs2 / Δm2, Δs3 / Δm3, Δs4 / Δm4) of the respective change rates (Δs1, Δs2, Δs3, Δs4) of the respective amplitudes s1 to s5 of the divided waveform of the signal S1 and the respective change rates (Δm1, Δm2, Δm3, Δm4) of the respective amplitudes m1 to m5 of the divided waveform of the signal S2. The estimation unit 114 may derive the ratio (s_total / m_total) of the average value (s_total / 4) of the amplitudes s1 to s5 of the divided waveform of the signal S1 and the average value (m_total / 4) of the respective amplitudes m1 to m5 of the divided waveform of the signal S2.
[0058] FIG. 8 is an example of the waveform of the signal S3 output from yet another channel (channel 3) of the sensor 50.
[0059] When the signal S3 is output from yet another channel (channel 3) of the sensor 50, in addition to the combination of the feature amounts of the waveform of the signal S1 and the feature amounts of the waveform of the signal S2, the estimation unit 114 may, for the combinations of the feature amounts of the waveform of the signal S1 and the feature amounts of the waveform of the signal S3 and the combinations of the feature amounts of the waveform of the signal S2 and the feature amounts of the waveform of the signal S3, derive respectively the ratios of the amplitudes to each other, the ratios of the sums of the amplitudes to each other, the ratios of the change rates of the amplitudes to each other, the ratios of the sums of the change rates of the amplitudes to each other, and the ratios of the average values of the amplitudes to each other.
[0060] The estimation unit 114 may estimate the type of foreign matter or the degree of deterioration of the target object based on the combination of a plurality of feature amounts of the waveform of one signal, the relationship information indicating the relationship between the type of foreign matter or the degree of deterioration of the target object, and the signal from the sensor 50. Alternatively, the estimation unit 114 may estimate the type of foreign matter or the degree of deterioration of the target object based on the combination of a plurality of feature amounts of the waveforms of at least two signals, the relationship information indicating the relationship between the type of foreign matter or the degree of deterioration of the target object, and at least two signals.
[0061] In this way, the estimation unit 114 may derive a plurality of parameters indicating the characteristics of the waveform of one or more signals, and estimate the type of foreign matter or the degree of deterioration of the target object based on the combination of the respective parameters. Thereby, the type of foreign matter or the degree of deterioration of the target object can be estimated with higher accuracy than estimating the type of foreign matter or the degree of deterioration of the target object based on one parameter for one signal.
[0062] The estimation unit 114 may estimate the state of the target object by using a learning model generated by supervised learning as the relationship information indicating the relationship between the feature amount of the waveform of the signal and the state of the target object, or the relationship information indicating the relationship between the combination of the feature amounts of the waveforms of at least two signals among the plurality of signals and the state of the target object. The estimation unit 114 may estimate the presence or absence of foreign matter in the target object by using a learning model generated by supervised learning as the relationship information indicating the relationship between the feature amount of the waveform of the signal and the mixing of foreign matter, or the relationship information indicating the relationship between the combination of the feature amounts of the waveforms of at least two signals among the plurality of signals and the mixing of foreign matter.
[0063] The estimation unit 114 performs machine learning according to a supervised learning algorithm using, as explanatory variables, the feature quantities of the signal waveform, at least two feature quantities of the signal waveform, or a plurality of combinations of the feature quantities of the waveforms of two signals, and using the state of the object as the objective variable, to generate a learned learning model for estimating the state of the object from the feature quantities of the signal waveform or a plurality of combinations of the feature quantities of the waveforms of two signals, and may store the model in the storage unit 120. The algorithm may be an algorithm of any method such as a neural network, a support vector machine, multiple regression analysis, a decision tree, a Gaussian process, etc. Further, the estimation unit 114 uses an anomaly detection algorithm such as principal component analysis, clustering, or OCSVM to show the difference between the waveform of the signal output from the sensor 50 when the sensor 50 is exposed to the reference object and the waveform of the signal output from the sensor 50 when the sensor 50 is exposed to the object, and may estimate the state of the object, for example, the degree of contamination of foreign matter in the object or the degree of deterioration of the object.
[0064] The notification unit 116 notifies the outside of the state of the object estimated by the estimation unit 114, for example, the degree of contamination of foreign matter in the object or the degree of deterioration of the object.
[0065] FIG. 9 is a flowchart showing an example of a procedure for estimating the contamination of foreign matter in an object. The acquisition unit 112 controls the switching valves 34 and 35 of the switching mechanism 30 to switch from a detection preparation state in which the sensor 50 is exposed to the internal space of the container 22 that houses the reference object to a detection state in which the sensor 50 is exposed to the internal space of the container 20 that houses the object (S100). The acquisition unit 112 controls the switching valves 34 and 35 so that only the pipe 31 that communicates with the container 20 that houses the object is in communication with the pipe 33, from the state in which only the pipe 32 that communicates with the container 22 that houses the reference object is in communication with the pipe 33.
[0066] The estimation device 100 starts detecting the substance in the internal space of the container 20 with the sensor 50 (S102). The acquisition unit 112 controls the switching valve 34 of the switching mechanism 30 to switch the sensor 50 from the detection state of being exposed to the internal air inside the container 20 to the detection preparation state of being exposed to the internal air inside the container 22 (S104). Next, the acquisition unit 112 acquires each signal detected in the detection period in the detection state from each channel of the sensor 50 (S106).
[0067] The estimation unit 114 derives the feature quantity of the waveform of each signal (S108). The estimation unit 114 estimates the presence or absence of foreign matter mixed into the object based on the feature quantity of the waveform of each signal (S110). If there is foreign matter mixed in, the estimation unit 114 estimates the type of foreign matter based on the feature quantity of the waveform of each signal and the relationship information indicating the relationship between the feature quantity of the waveform of each signal and the type of foreign matter (S112).
[0068] The notification unit 116 notifies the outside that it is an abnormal state in which foreign matter has been mixed into the object and the type of foreign matter (S114). When the estimation unit 114 estimates that no foreign matter has been mixed into the object, the notification unit 116 notifies the outside that the object is in a normal state (S116).
[0069] FIG. 10 is a flowchart showing an example of a procedure for estimating the deterioration state of an object. The acquisition unit 112 controls the switching valve 34 and the switching valve 35 of the switching mechanism 30 to switch the sensor 50 from the detection preparation state of being exposed to the internal space of the container 22 that houses the reference object to the detection state of being exposed to the internal space of the container 20 that houses the object (S100). The acquisition unit 112 controls the switching valve 34 and the switching valve 35 so that only the pipe 31 that communicates with the container 20 that houses the object communicates with the pipe 33, from the state where only the pipe 32 that communicates with the container 22 that houses the reference object communicates with the pipe 33.
[0070] The estimation device 100 starts detecting the substances in the internal space of the container 20 by the sensor 50 (S102). The acquisition unit 112 controls the switching valve 34 of the switching mechanism 30 to switch the sensor 50 from the detection state of being exposed to the internal air inside the container 20 to the detection preparation state of being exposed to the internal air inside the container 22 (S104). Next, the acquisition unit 112 acquires each signal detected during the detection period in the detection state from each channel of the sensor 50 (S106).
[0071] The estimation unit 114 derives the feature amounts of the waveforms of each signal (S108). The estimation unit 114 estimates whether the object is in a deteriorated state based on the feature amounts of the waveforms of each signal (S120). When the object is in a deteriorated state, the estimation unit 114 estimates the degree of deterioration of the object based on the feature amounts of the waveforms of each signal and the relationship information indicating the relationship between the feature amounts of the waveforms of each signal and the degree of deterioration of the object (S122).
[0072] The notification unit 116 notifies the outside that the object is in a deteriorated state and the degree of deterioration of the object (S124). When the estimation unit 114 estimates that the object is not in a deteriorated state, the notification unit 116 notifies the outside that the object is in a normal state (S126).
[0073] FIG. 11A and FIG. 11B are flowcharts showing an example of the procedure for deriving the feature amounts of each signal. The relationship information used by the estimation unit 114 for estimating the state of the object, such as the presence or absence of foreign matter mixed into the object and the deteriorated state of the object, may vary depending on the type of the object. Therefore, the estimation unit 114 specifies the types of the feature amounts of each waveform of each signal shown in association with the state of the object according to the type of the object for the relationship information used for estimating the state of the object. When the estimation unit 114 estimates the type of foreign matter, it is considered that the types of foreign matter that may be mixed in vary depending on the type of the object. Thus, the estimation unit 114 may specify the types of the feature amounts of each waveform of each signal shown in association with the type of foreign matter according to the type of the object for the relationship information used for estimating the type of foreign matter. The estimation unit 114 may specify the type of the object based on the information about the object input from the user.
[0074] The estimation unit 114 acquires each signal of each channel for the gas in the container 20 via the acquisition unit 112 (S200). The estimation unit 114 determines whether to derive each amplitude of each signal based on the type of feature amount of each specified signal (S202).
[0075] If each amplitude is to be derived, the estimation unit 114 generates a feature amount group yc1 by deriving each amplitude of each signal at each interval t (S204).
[0076] Next, the estimation unit 114 determines whether to derive the total value of each amplitude of each signal based on the type of feature amount of each specified signal (S206). If the total value of each amplitude is to be derived, the estimation unit 114 generates a feature amount group yc2 by deriving the total value of each amplitude of each signal at each interval t (S208).
[0077] Next, the estimation unit 114 determines whether to derive the inter-channel ratio of each amplitude of each signal based on the type of feature amount of each specified signal (S210). If the inter-channel ratio of each amplitude is to be derived, the estimation unit 114 generates a feature amount group yc3 by deriving the inter-channel ratio of each amplitude of each signal at each interval t (S212).
[0078] Next, based on the type of feature amount of each specified signal, it is determined whether to derive the inter-channel ratio of the total value of each amplitude of each signal (S214). If the inter-channel ratio of the total value of each amplitude is to be derived, the estimation unit 114 generates a feature amount group yc4 by deriving the inter-channel ratio of the total value of each amplitude of each signal at each interval t (S216).
[0079] Next, the estimation unit 114 determines whether to derive the change rate of each amplitude of each signal based on the type of feature amount of each specified signal (S218). If the change rate of each amplitude of each signal is to be derived, the estimation unit 114 generates a feature amount yc5 by deriving the change rate of each amplitude of each signal at each interval t (S220).
[0080] Next, the estimation unit 114 determines whether to derive the total value of the change rates of the amplitudes of each signal based on the type of the feature amount of each identified signal (S222). If it is to derive the total value of the change rates of the amplitudes of each signal, the estimation unit 114 generates a feature amount group yc6 by deriving the total value of the change rates of the amplitudes at each interval t (S224).
[0081] Next, the estimation unit 114 determines whether to derive the average value of the change rates of the amplitudes of each signal based on the type of the feature amount of each identified signal (S226). If it is to derive the average value of the change rates of the amplitudes of each signal, the estimation unit 114 generates a feature amount group yc7 by deriving the average value of the change rates at each interval t (S228).
[0082] Next, the estimation unit 114 determines whether to derive the average value and the standard deviation of the temperature based on the type of the feature amount of each identified signal (S230). If it is to derive the average value and the standard deviation of the temperature, the estimation unit 114 acquires temperature information indicating the temperature inside the container 20, and generates feature amount groups yt1, yt2, and yt3 by deriving the temperature, the average value of the temperature, and the standard deviation of the temperature at each interval t (S232).
[0083] Next, the estimation unit 114 determines whether to derive the average value and the standard deviation of the humidity based on the type of the feature amount of each identified signal (S234). If it is to derive the average value and the standard deviation of the humidity, the estimation unit 114 acquires humidity information indicating the humidity inside the container 20, and generates feature amount groups yh1, yh2, and yh3 by deriving the humidity, the average value of the humidity, and the standard deviation of the humidity at each interval t (S236).
[0084] From the above processing, the estimation unit 114 derives a feature amount group used for estimating the state of the object and the like. Note that the types and derivation order of the feature amount groups shown in FIGS. 11A and 11B are merely examples.
[0085] According to the estimation device 100 according to this embodiment, by suppressing changes such as humidity or temperature in the environment where the sensor 50 is placed, it is possible to suppress the influence of environmental changes on the waveform of the signal output from the sensor 50, and it is possible to estimate whether a trace amount of substance is contained in the object based on the signal output from the sensor 50.
[0086] Figs. 12A to 12H show the waveforms of the signals of each channel when, as the sensor 50, a film surface type stress sensor composed of 8 types (8 channels) of sensitive film arrays is used, and the sensor 50 is exposed to milk mixed with 38 ppm of acetone (sometimes referred to as "milk mixed with acetone") as the object, and the sensor 50 is exposed to normal milk without foreign matter mixed in as the reference object, and the states are alternately switched.
[0087] The waveforms of the signals shown in Figs. 12A to 12H are the waveforms of the signals output from each channel of the sensor 50 when, from 0 seconds to 120 seconds, the sensor 50 is exposed to milk mixed with acetone, and from 120 seconds to 240 seconds, the sensor 50 is exposed to normal milk, and this state is repeated for 30 minutes in a 2-minute cycle, and the waveforms are normalized to 0 mV with the starting point at 0 seconds.
[0088] In the waveforms of the signals shown in Figs. 12A to 12H, when the sensor 50 is exposed to milk mixed with acetone, the amplitudes of the respective signals tend to increase, and when the sensor 50 is exposed to milk, the amplitudes of the respective signals tend to decrease. That is, it can be seen that there is a difference in the signals output from each channel of the sensor 50 between the state where the sensor 50 is exposed to milk mixed with acetone and the state where the sensor 50 is exposed to milk. That is, the difference from the signals of each channel of the sensor 50 when exposed to milk mixed with acetone is more likely to appear when milk is used than when air is used as the reference object.
[0089] Figures 13A to 13H show the waveforms of the signals of each channel when, as the sensor 50, a film surface stress sensor composed of an array of 8 types (8 channels) of sensitive films is used, and the sensor 50 is exposed to water contaminated with tobacco (which may be referred to as "water contaminated with tobacco") as the object, and the sensor 50 is exposed to normal water without foreign matter as the reference object, and the states are alternately switched.
[0090] The waveforms of the signals shown in Figures 13A to 13H are the waveforms of each signal output by each channel of the sensor 50 when, after exposing the sensor 50 to water contaminated with tobacco from 0 seconds to 120 seconds, and then exposing the sensor 50 to normal water from 120 seconds to 240 seconds, the cycle is repeated every 2 minutes for 30 minutes, and the time point of 0 seconds is used as the starting point and normalized to 0 mV.
[0091] In the waveforms of the signals shown in Figures 13A to 13H, when the sensor 50 is exposed to water contaminated with tobacco, the amplitude of each signal has an upward trend, and when the sensor 50 is exposed to normal water, the amplitude of each signal has a downward trend. That is, it can be seen that there is a difference in the signals output by each channel of the sensor 50 between the state where the sensor 50 is exposed to water contaminated with tobacco and the state where the sensor 50 is exposed to normal water. That is, the difference from the signals of each channel of the sensor 50 when exposed to water contaminated with tobacco appears more easily when water is used than when air is used as the reference object.
[0092] Figure 14 shows the results of principal component analysis of the waveforms of the signals of each channel when, as the sensor 50, a film surface stress sensor composed of an array of 8 types (8 channels) of sensitive films is used, and the sensor 50 is exposed to the remaining water in the drink after taking a drink with the mouth on the drinking spout as the object, the sensor 50 is exposed to water intentionally mixed with saliva as the object, and the sensor 50 is exposed to normal water without foreign matter as the reference object, and the states are alternately switched every 1 minute for 15 minutes.
[0093] From the results of the principal component analysis shown in FIG. 14, it can be seen that water mixed with saliva and normal water without saliva can be distinguished. That is, the estimation unit 114 performs principal component analysis on the waveform of the signal output from the sensor 50 during the detection period from the detection preparation state exposed to the reference substance water to the detection state exposed to the target substance water, and then from the detection preparation state to the detection state again. Based on the results of the principal component analysis, it may be estimated whether saliva is mixed in the target substance water. The estimation unit 114 performs principal component analysis, which is a type of multivariate analysis, on a plurality of feature amounts of the signal waveform, using the plurality of feature amounts of the signal waveform as explanatory variables and the first principal component and the second principal component as target variables, and plots the target object on a two-dimensional plane with the first principal component as the X-axis and the second principal component as the Y-axis to derive the results of the principal component analysis obtained. The estimation unit 114 may estimate whether saliva is mixed in the target substance water based on the region in the two-dimensional plane where the plot of the target object exists.
[0094] FIG. 15 shows the results of principal component analysis of the waveforms of the signals of each channel when, as the sensor 50, a film surface type stress sensor composed of 8 types (8 channels) of sensitive film arrays is used, and the sensor 50 is exposed to milk one week after manufacture, the sensor 50 is exposed to milk three weeks after manufacture, and the sensor 50 is exposed to milk immediately after manufacture as a reference substance, and the exposure states are alternately switched every 1 minute for 15 minutes.
[0095] The estimation unit 114 can estimate the elapsed period from the manufacturing date of the target substance milk by specifying the position of the two-dimensional plot obtained as the result of the principal component analysis of the waveform of the signal output from the sensor 50.
[0096] FIG. 16 shows the results of principal component analysis of the waveforms of the signals of each channel when, as the sensor 50, a film surface type stress sensor composed of 8 types (8 channels) of sensitive film arrays is used, and the sensor 50 is exposed to Earl Grey tea, oolong tea, roasted green tea, green tea, barley tea, and dark green tea as the target substance, and the sensor 50 is exposed to water as the reference substance, and the exposure states are alternately switched every 1 minute for 15 minutes.
[0097] The estimation unit 114 can estimate the type of the object by identifying the position of the two-dimensional plot obtained as a result of the principal component analysis of the waveform of the signal output from the sensor 50.
[0098] FIG. 17 shows an example of a computer 1200 in which multiple aspects of the present invention may be embodied in whole or in part. A program installed in the computer 1200 can cause the computer 1200 to function as an operation associated with the apparatus according to an embodiment of the present invention or as one or more "parts" of the apparatus. Alternatively, the program can cause the computer 1200 to execute the operation or the one or more "parts". The program can cause the computer 1200 to execute a process according to an embodiment of the present invention or a stage of the process. Such a program may be executed by the CPU 1212 to cause the computer 1200 to execute certain operations associated with some or all of the blocks of the flowcharts and block diagrams described herein.
[0099] The computer 1200 according to the present embodiment includes a CPU 1212 and a RAM 1214, which are interconnected by a host controller 1210. The computer 1200 also includes a communication interface 1222 and an input / output unit, which are connected to the host controller 1210 via an input / output controller 1220. The computer 1200 also includes a ROM 1230. The CPU 1212 operates according to programs stored in the ROM 1230 and the RAM 1214, thereby controlling each unit.
[0100] The communication interface 1222 communicates with other electronic devices via a network. A hard disk drive may store programs and data used by the CPU 1212 in the computer 1200. The ROM 1230 stores therein a boot program etc. executed by the computer 1200 when activated, and / or a program dependent on the hardware of the computer 1200. Programs are provided via a computer-readable recording medium such as a CD-ROM, a USB memory or an IC card, or via a network. The programs are installed in the RAM 1214, which is also an example of a computer-readable recording medium, or in the ROM 1230, and executed by the CPU 1212. The information processing described in these programs is read by the computer 1200, resulting in cooperation between the programs and the various types of hardware resources described above. The device or method may be configured by realizing the operation or processing of information according to the use of the computer 1200.
[0101] For example, when communication is executed between the computer 1200 and an external device, the CPU 1212 may execute a communication program loaded in the RAM 1214 and instruct the communication interface 1222 to perform communication processing based on the processing described in the communication program. The communication interface 1222 reads the transmission data stored in the transmission buffer area provided in the RAM 1214 or in a recording medium such as a USB memory under the control of the CPU 1212, transmits the read transmission data to the network, or writes the received data received from the network to the reception buffer area etc. provided on the recording medium.
[0102] Also, the CPU 1212 may cause all or a necessary part of a file or database stored in an external recording medium such as a USB memory etc. to be read into the RAM 1214, and execute various types of processing on the data on the RAM 1214. The CPU 1212 may then write back the processed data to the external recording medium.
[0103] Various types of information such as various types of programs, data, tables, and databases may be stored in a recording medium and may undergo information processing. The CPU 1212 may perform various types of processing on the data read from the RAM 1214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, search / replacement of information, etc., described throughout this disclosure and specified by the instruction sequence of the program, and write back the results to the RAM 1214. Also, the CPU 1212 may search for information in files, databases, etc. within the recording medium. For example, when a plurality of entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored in the recording medium, the CPU 1212 searches for an entry that matches the condition where the attribute value of the first attribute is specified among the plurality of entries, reads the attribute value of the second attribute stored in the entry, and thereby may obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.
[0104] The program or software module described above may be stored in a computer-readable storage medium on or near the computer 1200. Also, a recording medium such as a hard disk or RAM provided within a server system connected to a dedicated communication network or the Internet can be used as a computer-readable storage medium, thereby providing the program to the computer 1200 via the network.
[0105] A computer-readable medium may include any tangible device that can store instructions executable by a suitable device. As a result, a computer-readable medium having instructions stored thereon will comprise a product that includes instructions that may be executed to create means for performing the operations specified in a flowchart or block diagram. Examples of computer-readable media may include electronic memory media, magnetic memory media, optical memory media, electromagnetic memory media, semiconductor memory media, and the like. More specific examples of computer-readable media may include floppy (registered trademark) disks, diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM (registered trademark)), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital versatile disc (DVD), Blu-ray (RTM) disc, memory stick, integrated circuit card, and the like.
[0106] Computer-readable instructions may include either source code or object code written in any combination of one or more programming languages. The source code or object code may include conventional procedural programming languages. Conventional procedural programming languages may include assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or object-oriented programming languages such as Smalltalk (registered trademark), JAVA (registered trademark), C++, and the like, and the "C" programming language or similar programming languages. The computer-readable instructions may be provided to a processor or programmable circuit of a programmable data processing device locally or via a wide area network (WAN) such as a local area network (LAN), the Internet, or the like. The processor or programmable circuit may execute the computer-readable instructions to create means for performing the operations specified in a flowchart or block diagram.
[0107] Here, the computer may be a computer such as a PC (personal computer), a tablet computer, a smartphone, a workstation, a server computer, or a general-purpose computer, or may be a computer system to which a plurality of computers are connected. Such a computer system to which a plurality of computers are connected is also called a distributed computing system and is a computer in a broad sense. In a distributed computing system, each of the plurality of computers executes a part of the program, and the plurality of computers collectively execute the program by transferring data during program execution between the computers as needed.
[0108] Examples of the processor include a computer processor, a central processing unit (CPU), a processing unit, a microprocessor, a digital signal processor, a controller, a microcontroller, and the like. The computer may include one processor or a plurality of processors. In a multiprocessor system including a plurality of processors, each processor executes a part of the program, and the plurality of processors collectively execute the program by transferring data during program execution between the processors as needed. For example, in the execution of multitasking, each of the plurality of processors may execute a part of each task in small pieces by switching tasks every time slice. In this case, which part of one program each processor executes changes dynamically. Also, which part of the program each of the plurality of processors executes may be statically determined by programming that takes into account the multiprocessor.
[0109] As described above, the present invention has been described using embodiments, but the technical scope of the present invention is not limited to the scope described in the above embodiments. It is obvious to those skilled in the art that various changes or improvements can be made to the above embodiments. It is clear from the description of the claims that forms with such changes or improvements can also be included in the technical scope of the present invention.
[0110] In the claims, the description, and the drawings, for the operations, procedures, steps, stages, and other processes in the apparatus, system, program, and method shown, the execution order of each process, such as the operation order, procedure, step, and stage, is not explicitly stated as "earlier" or "preceding" etc., and it should be noted that it can be realized in any order as long as the output of the previous process is not used in the subsequent process. Regarding the operation flow in the claims, the description, and the drawings, even if it is described using "first," "next," etc. for convenience, it does not mean that it is essential to implement in this order.
Explanation of Reference Numerals
[0111] 10 Estimation system 20, 22 Container 30 Switching mechanism 31, 32, 33 Pipe 34, 35 Switching valve 40 Environment sensor 50 Sensor 100 Estimation device 110 Control unit 112 Acquisition unit 114 Estimation unit 116 Notification unit 120 Storage unit 1200 Computer 1210 Host controller 1212 CPU 1214 RAM 1220 Input / output controller 1222 Communication interface 1230 ROM
Claims
1. A sensor having a sensing unit that physically changes in response to at least one substance generated from an object and outputs a signal corresponding to the physical change, from a first state in which it is exposed to a reference substance to a second state in which it is exposed to the object, and then again from the second state to the first state. An acquisition unit that acquires a signal output from the sensor during a detection period until the change; An estimation unit that estimates the state of the object based on a feature amount of the waveform of the signal. An estimation device comprising:
2. The estimation device according to claim 1, wherein the reference substance is a substance of the same type as the object that satisfies a predetermined condition, or water.
3. The sensor has a plurality of sensing units having different characteristics of physical changes corresponding to the at least one substance generated from the object. The acquisition unit acquires a plurality of signals output from each of the plurality of sensing units during the detection period. The estimation device according to claim 1, wherein the estimation unit estimates the state of the object based on a combination of feature amounts of waveforms of at least two of the plurality of signals.
4. The estimation device according to claim 1, wherein the sensing unit includes a sensing film that deforms when the at least one substance is adsorbed and diffuses therein.
5. The estimation device according to claim 4, wherein the sensing film includes an organic-inorganic hybrid material.
6. The feature amount of the waveform of the signal includes at least one of each amplitude of a plurality of divided waveforms obtained by dividing the waveform of the signal at predetermined intervals, the sum of the amplitudes of the plurality of divided waveforms, each change rate of the amplitudes of the plurality of divided waveforms, the sum of the change rates of the amplitudes of the plurality of divided waveforms, and the average value of the change rates of the amplitudes of the plurality of divided waveforms. The estimation device according to claim 1.
7. The combination of the feature amounts of the waveforms of the at least two signals includes at least one of the ratios of the amplitudes of the plurality of divided waveforms of each of the at least two signals obtained by dividing the waveforms of the at least two signals at predetermined intervals, the ratios of the total values of the amplitudes of the plurality of divided waveforms of each of the at least two signals, the ratios of the change rates of the amplitudes of the plurality of divided waveforms of each of the at least two signals, the ratios of the total of the change rates of the amplitudes of the plurality of divided waveforms of each of the at least two signals, and the ratios of the average values of the amplitudes of the plurality of divided waveforms of each of the at least two signals. The estimation device according to claim 3.
8. The estimation device according to claim 1, wherein the object is food.
9. The estimation device according to claim 1, wherein the object is a liquid substance.
10. The estimation device according to claim 1, wherein the estimation unit estimates whether or not a foreign object is mixed in the object as the state of the object.
11. The estimation device according to claim 10, wherein when the estimation unit estimates that a foreign object is mixed in the object, the estimation unit further estimates the type of the foreign object based on the relationship information indicating the relationship between the feature amount of the signal waveform and the type of the foreign object and the signal.
12. The estimation device according to claim 1, wherein the estimation unit estimates whether or not the object is in a deteriorated state as the state of the object.
13. The estimation device according to claim 12, wherein when the estimation unit estimates that the object is in a deteriorated state, the estimation unit further estimates the degree of deterioration of the object based on the relationship state indicating the relationship between the feature amount of the signal waveform and the degree of deterioration of the object and the signal.
14. The estimation device according to any one of claims 1 to 13, the sensor, a switching mechanism that communicates with the internal space of the housing unit that houses the object and switches between the first state in which the sensor is exposed to the reference object and the second state in which the sensor is exposed to the object An estimation system comprising.
15. The switching mechanism is a first pipe communicating with the internal space of the housing unit, a second pipe communicating with the external space of the housing unit, a third pipe that is switchably communicable with one of the first pipe and the second pipe Including, The sensor is disposed in the third pipe. The estimation system according to claim 14, wherein when the third pipe communicates with the second pipe, the sensor is in the first state, and when the third pipe communicates with the first pipe, the sensor is in the second state.
16. A step of obtaining a signal output from a sensor having a sensitive portion that physically changes in response to at least one substance generated from an object and outputs a signal corresponding to the physical change, from a first state in which the sensor is exposed to a reference substance to a second state in which the sensor is exposed to the object, and then from the second state to the first state again, during a detection period until the sensor switches back to the first state; A step of estimating the state of the object based on a feature amount of the waveform of the signal An estimation method comprising:
17. When executed by a computer, the computer Causes a sensor having a sensitive portion that physically changes in response to at least one substance generated from an object and outputs a signal corresponding to the physical change to obtain a signal output from the sensor during a detection period from a first state in which the sensor is exposed to a reference substance to a second state in which the sensor is exposed to the object, and then from the second state to the first state again, until the sensor switches back to the first state; A program for causing the state of the object to be estimated based on a feature amount of the waveform of the signal.