Odor analyzer, odor analysis method, and odor analysis program

Abstract

An odor analyzing device according to one aspect of the present invention is provided with: a measuring unit including n (where n is an integer at least equal to 1) odor sensors; an acquiring unit for acquiring measured values measured by the n odor sensors of the measuring unit; an extracting unit for extracting m (where m is an integer at least equal to 1, and n≥m) main components from among n-dimensional data having the measured value from one odor sensor, acquired by the acquiring unit, as one dimension; a projecting unit for projecting the n-dimensional data in m dimensions on the basis of the main components extracted by the extracting unit; and a rendering unit for rendering the m-dimensional projection data projected by the projecting unit.

Description

Technical Field

[0001] The present invention relates to an odor analyzer, an odor analysis method, and an odor analysis program.

Background Art

[0002] Conventionally, there has been an odor sensor that detects odor components by depositing a thin film on a crystal oscillator and detecting changes in frequency. For example, Patent Document 1 discloses a multi-head type odor sensor that includes a plurality of heads each provided with an organic thin film that selectively adsorbs an odor substance via an electrode on at least one surface of a plurality of crystal oscillators and surface acoustic wave elements, and determines an odor substance and an odor quality from changes in the natural vibration of each head.

Prior Art Documents

Patent Documents

[0003] <000°017>

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the conventional technology, although a plurality of measurement values measured by a plurality of odor sensors are displayed, it is sometimes difficult to recognize the characteristics of the odor to be measured because it is impossible to determine which measurement value of which odor sensor should be focused on.

[0005] In view of the above circumstances, the present invention has been made, and one object thereof is to provide an odor analyzer, an odor analysis method, and an odor analysis program that can make it easier to recognize the characteristics of an odor.

Means for Solving the Problems

[0006] (1) To solve the above problems, the odor analyzer comprises: a measurement unit having n odor sensors (where n is an integer of 1 or more); an acquisition unit that acquires measurement values ​​measured by the n odor sensors of the measurement unit; an extraction unit that extracts m principal components (where m is an integer of 1 or more, and n≧m) from n-dimensional data in which the measurement value of one odor sensor acquired by the acquisition unit is one-dimensional; a projection unit that projects the n-dimensional data into m dimensions based on the principal components extracted by the extraction unit; and a drawing unit that draws the m-dimensional projection data projected by the projection unit.

[0007] (2) In addition, in the odor analyzer of the embodiment, the odor sensor may change its natural frequency when odorous substances are adsorbed onto the adsorption part, and the measurement part may measure the natural frequency.

[0008] (3) In addition, in the odor analyzer of the embodiment, the projection unit may project n-dimensional data into m-dimensional space by projecting it onto a space represented by the principal component vectors of m principal components.

[0009] (4) In addition, in the odor analyzer of the embodiment, the extraction unit may extract the main components such that the difference between the projection data of the first odor sample and the projection data of the second odor sample is maximized.

[0010] (5) In addition, in the odor analyzer of the embodiment, the drawing unit may draw the projection data by changing the drawing scale for each dimension.

[0011] (6) In addition, in the odor analyzer of the embodiment, the drawing unit may draw the projection data of a reference odor sample and the projection data of an odor sample having a predetermined difference in a distinguishable manner.

[0012] (7) In order to solve the above problems, the odor analysis method includes a measurement step of measuring odor with n odor sensors (where n is an integer of 1 or more), an acquisition step of acquiring the measured values ​​obtained by the n odor sensors in the measurement step, an extraction step of extracting m principal components (where m is an integer of 1 or more) from n-dimensional data in which the measured value of one odor sensor acquired in the acquisition step is one-dimensional, a projection step of projecting the n-dimensional data into m dimensions based on the principal components extracted in the extraction step, and a drawing step of drawing the m-dimensional projected data projected in the projection step.

[0013] (8) To solve the above problems, the odor analysis program causes the computer to perform the following: a measurement process that measures odor using n odor sensors (where n is an integer greater than or equal to 1); an acquisition process that obtains the measured values ​​from the n odor sensors in the measurement process; an extraction process that extracts m principal components (where m is an integer greater than or equal to 1) from n-dimensional data, where the measured value from one odor sensor obtained in the acquisition process is one-dimensional; a projection process that projects the n-dimensional data onto m dimensions based on the principal components extracted in the extraction process; and a drawing process that draws the m-dimensional projected data obtained in the projection process. [Effects of the Invention]

[0014] According to one embodiment of the present invention, by having n odor sensors (where n is an integer of 1 or more), acquiring measurement values ​​from the n odor sensors, extracting m principal components (where m is an integer of 1 or more) from n-dimensional data in which the measurement value of one odor sensor is one-dimensional, projecting the n-dimensional data into m dimensions based on the extracted principal components, and plotting the projected m-dimensional data, the characteristics of the odor can be easily recognized. [Brief explanation of the drawing]

[0015] [Figure 1] This block diagram shows an example of the configuration of an odor analyzer in an embodiment. [Figure 2]It is a diagram showing an example of the appearance of the odor analyzer in the embodiment. [Figure 3] It is a diagram showing a first display example of the odor analyzer in the embodiment. [Figure 4] It is a diagram showing a second display example of the odor analyzer in the embodiment. [Figure 5] It is a block diagram showing an example of the hardware configuration of the odor analyzer in the embodiment. [Figure 6] It is a flowchart showing a first example of the operation of the odor analyzer in the embodiment.

Mode for Carrying Out the Invention

[0016] Hereinafter, the odor analyzer, the odor analysis method, and the odor analysis program in an embodiment of the present invention will be described in detail with reference to the drawings.

[0017] First, the configuration of the odor analyzer will be described using FIG. 1. FIG. 1 is a block diagram showing an example of the configuration of the odor analyzer in the embodiment.

[0018] In FIG. 1, the odor analyzer 1 has functional units of a measurement unit 11, an acquisition unit 12, an extraction unit 13, a projection unit 14, and a drawing unit 15. Each functional unit of the odor analyzer 1 will be described as being realized by software. The odor analyzer 1 is, for example, a desktop PC, a notebook PC, a smartphone, a tablet PC, or a head-mounted, glasses-type, or wristwatch-type device. The odor analyzer 1 is communicably connected to the server 2 via the network 9.

[0019] The measurement unit 11 has n odor sensors for measuring odor (where n is an integer of 1 or more). As the odor sensors, various types can be used, such as semiconductor types like oxide semiconductor type and organic semiconductor type, and crystal oscillator type using epoxy resin film, vinyl acetate resin film, Langmuir - Blodgett film, etc. as the sensitive film, and those using surface acoustic wave (SAW) filters or film bulk acoustic resonator (FBAR) filters. Hereinafter, the case where there are n odor sensors for measuring the natural frequency that changes when an odor substance is adsorbed to the adsorption part will be described.

[0020] The adsorption part is the part that adsorbs the odor substance in the odor sensor. The adsorption part can be implemented, for example, by providing a thin film on a vibrating body such as a crystal oscillator or a piezoelectric element. The thin film can be formed, for example, by thin film forming means such as vapor deposition or sputtering.

[0021] The material used for the thin film can be composed of, for example, at least one material selected from amino acids, saccharides, bases, aromatic ketones, dihydric alcohols or phenols.

[0022] For amino acids, for example, valine, isoleucine, leucine, methionine, lysine, phenylalanine, tryptophan, threonine, histidine, arginine, glycine, alanine, serine, tyrosine, cysteine, asparagine, glutamine, proline, aspartic acid, glutamic acid, etc. can be used.

[0023] Also, for saccharides, for example, α - glucose (glucose), β - glucose (glucose), fructose, galactose, sucrose, lactose, maltose, cellobiose amylose (starch), cellulose, etc. can be used.

[0024] Examples of bases that can be used include adenine, guanine, cytosine, uracil, and thymine.

[0025] Aromatic ketones include, for example, 3,4-methylenedioxypropiophenone, 3-acetyl-6-methoxybenzaldehyde, acridone, 2-acetyl-5-methylfuran, acetylpyrazine, acetylpyridine, 2-acetylpyridine, 2-acetylfuran, acetanisole, acetophenone, acetohexamide, acepromazine, amiodarone, alectinib, anthrone, isomaltol, indigocarmine, indigo, ebastine, eperisone, cathinone, galacetophenone, chalcone, xanthon, kynurenine, chloroacetophenone, chlorofacinone, ketanserine, zaltoprofen, diazonaphthoquinone, divapron, dihydromenaquinone, suprofen, celiprolol, tamsulosin, thiaprofen You can use benzoyl acid, desaspirin, tonalide, trichostatin A, droperidol, nitisinone, valerophenone, picein, piceol, vitamin K, pifisrin, hypericin, phantolide, phylloquinone, phenylglyoxal, phenylglyoxylic acid, bupropion, flubendazole, protopine, propafenone, butyrophenone, propiophenone, phlorizin, floretin, pungenin, versalid, benziodarone, benzyl, benzantrone, benzbromarone, benzoyl group, benzoin, benzophenone, aromatic polyether ketone, N-formyl kynurenine, musk ketone, metyrapone, methylmenaquinone, methcathinone, mebendazole, ramosetron, raloxifene, roberine, etc.

[0026] In addition, dihydric alcohols such as 1,2-ethanediol, ethylene glycol, 1,2-propanediol, 1,3-propanediol, 2-chloro-1,3-propanediol, 3-chloro-1,2-propanediol, 1,2-butanediol, 1,3-butanediol, 1,4-butanediol, 2,3-butanediol, 2-methyl-1,2-propanediol, 1,5-pentanediol, 2-methyl-2,3-butanediol, 1,6-hexanediol, 2,5-hexanediol, 2-methyl-2,4-pentanediol, 2,3-dimethyl-2,3-butanediol, and 2-butyne-1,4-diol can be used.

[0027] In addition, phenols such as phenol, dibutylhydroxytoluene, bisphenol A (BPA), cresol, estragyl, eugenol, guaiacol, picric acid, phenolphthalein, serotonin, dopamine, adrenaline, noradrenaline, thymol, and tyrosine can be used.

[0028] An adsorption section, where a thin film is provided on a vibrating body, has its own natural frequency. When odor substances are adsorbed onto the thin film on the vibrating body, the natural frequency of the adsorption section changes. The change in natural frequency becomes larger as the amount of odor substances adsorbed onto the thin film increases. By outputting the natural frequency corresponding to the amount of odor substances adsorbed, the odor sensor can measure the amount of odor substances adsorbed.

[0029] The measurement unit 11 measures the values ​​of n odor sensors, where n is an integer greater than or equal to 1. In this embodiment, the case where there are nine odor sensors will be described later as an example, but the number of odor sensors is arbitrary. In this example, the measurement unit 11 is shown as being implemented in software, so the measurement unit 11 does not include hardware odor sensors. However, the measurement unit 11 may be implemented as a sensor unit that includes odor sensors.

[0030] The acquisition unit 12 acquires the measured values ​​from the n odor sensors of the measurement unit 11. For example, if there are 9 odor sensors, the acquisition unit 12 will acquire the measured values ​​from 9 odor sensors. The acquisition unit 12 may also acquire measured values ​​at multiple measurement intervals. For example, if odor is measured every 10 seconds by 9 odor sensors, the acquisition unit 12 will acquire 9 × 6 = 54 measured values ​​per minute. In this embodiment, "acquisition" can refer to either pull-type reception or push-type reception. Similarly, "providing" can refer to either push-type transmission or pull-type transmission.

[0031] The extraction unit 13 extracts m principal components from n-dimensional data, where the measurement value of one odor sensor acquired by the acquisition unit 12 is one-dimensional (where m is an integer greater than or equal to 1, and n ≥ m). For example, if there are 9 odor sensors, n = 9, and the extraction unit 13 can extract 9-dimensional data. The extraction unit 13 extracts the measurement values ​​of m (for example, 3) odor sensors from the 9 sensors as principal components.

[0032] Here, the principal component is the component that characterizes the odor of the object being measured, and is the measured value obtained by a specific odor sensor. For example, when comparing the odors of sample A and sample B, the principal component is the component that characterizes and distinguishes the odor of sample A from that of sample B. For example, when distinguishing between the odor of oranges and lemons using nine odor sensors, suppose seven of the odor sensors output approximate measured values ​​for the same citrus odor. On the other hand, suppose two odor sensors distinguish between the odor of oranges and lemons and output measured values. In this case, the principal component would be the measured values ​​from the two odor sensors mentioned above. By extracting these two measured values ​​as the principal component, it becomes possible to characterize and distinguish between the odors of oranges and lemons. By extracting the principal component, the characteristics of the odor can be made easier to recognize.

[0033] The extraction of principal components is performed by selecting m measurement values ​​as principal components from n measurement values. The extraction of principal components may be performed based on predetermined settings (settings for which measurement values ​​to extract). For example, if the sample to be measured is determined, the principal components can be predetermined. For example, when measuring the change in odor of a sample over time, the measurement values ​​of an odor sensor that easily detects the change in odor over time can be predetermined as the principal components. Also, when distinguishing between genuine perfume and counterfeit perfume, the measurement values ​​of an odor sensor that easily shows differences between genuine and counterfeit perfumes can be predetermined as the principal components. The extraction unit 13 can perform rapid measurement by extracting the predetermined m principal components.

[0034] On the other hand, the extraction of principal components may be dynamically modified. For example, if the odor of a sample cannot be predicted, it is not possible to predetermine which odor sensor's measurement value will be used as the principal component. For example, the extraction unit 13 may dynamically determine this based on projection data, which will be described later. Alternatively, the extraction unit 13 may infer the principal components from the learning results of machine learning based on past data of sample types and suitable principal components. For example, when measuring the odor of spoiled food, the principal components may be predicted from the components of the food being measured by using machine learning with previously measured food components and suitable principal components as training data.

[0035] The projection unit 14 projects n-dimensional data into m-dimensional data based on the principal components extracted by the extraction unit 13 (where n ≥ m). In this embodiment, principal component analysis (PCA) may be performed based on the principal components extracted by the extraction unit 13.

[0036] Principal component analysis (PCA) is a multivariate analysis that synthesizes a small number of uncorrelated principal components from a large number of correlated variables to best represent the overall variability. By performing PCA, n-dimensional measurement data can be reduced to m-dimensional data, making it easier to understand the characteristics of the data.

[0037] The projection unit 14 projects the measured values ​​in an n-dimensional space onto a vector space formed by the m-dimensional principal component vectors (principal component vectors) extracted by the extraction unit 13, generating projected data represented by the vector data of the principal component vectors. For example, when n=9 and m=3, the 9-dimensional measured values ​​are projected onto a 3-dimensional vector space, allowing the projected data to be visualized. When m=2, the 9-dimensional measured values ​​are projected onto a 2-dimensional vector plane, allowing the projected data to be visualized. When m=1, the 9-dimensional measured values ​​are projected onto a 1-dimensional vector line, allowing the projected data to be visualized. By projecting the measured values ​​of an n-dimensional odor sensor into 1 to 3 dimensions to generate projected data, the measured data becomes visible, making it easier to recognize the characteristics of the odor.

[0038] The extraction unit 13 extracts principal components such that the difference between the projection data of the first odor sample and the projection data of the second odor sample is maximized. As described above, the projection data is obtained by projecting measurement values ​​other than the principal components onto an m-dimensional principal component vector, and is therefore m-dimensional vector data. The difference in projection data is the difference between two m-dimensional vector data, and can be calculated by subtracting the m-dimensional matrix. Since the projection data represents the characteristics of the odor, maximizing the difference with the projection data means maximizing the difference in the characteristic odor. In other words, the extraction unit 13 extracts principal components such that the difference between the odor of the first odor sample and the odor of the second odor sample is recognized as the maximum. For example, if n=9 and m=3, there are 9 × 8 × 7 = 504 possible combinations of principal components. The extraction unit 13 can calculate the difference in projection data for each combination and extract the principal components of the combination with the maximum difference. Furthermore, the combination of principal components may be prioritized based on past calculation results, and the difference may be calculated accordingly.

[0039] The drawing unit 15 draws the m-dimensional projection data projected by the projection unit 14. Drawing means rendering an image based on the projection data. For example, if m=1, the drawing unit 15 performs one-dimensional rendering. If m=2, the drawing unit 15 performs two-dimensional rendering. If m=3, the drawing unit 15 performs three-dimensional rendering. If m is 4 or greater, the drawing unit 15 performs rendering using a predetermined drawing method. A predetermined drawing method is, for example, a method for drawing shapes using tables, figures, symbols, or coloring. For example, the drawing unit 15 may represent four-dimensional projection data by changing the size of points mapped to a three-dimensional graph. Similarly, the drawing unit 15 may represent five-dimensional projection data by changing the size and color of points mapped to a three-dimensional graph. Furthermore, the drawing unit 15 may plot shapes (for example, circles, squares, triangles, stars, or characters) or symbols (for example, JIS symbols) on the graph instead of points. The rendering of projection data may also include rendering of video. For example, the drawing unit 15 may render a video that dynamically changes the shape or color of the plotted points. Additionally, the rendering of projection data may include rendering of audio. For example, the drawing unit 15 may render audio that changes the type and volume of sound according to the projection data.

[0040] The plotting unit 15 plots the projected data by changing the plotting scale for each dimension. Changing the plotting scale refers to, for example, the scale of the graph axes. For example, if it is desired to recognize in detail the changes in measured values ​​in a particular principal component vector, the plotting unit 15 may enlarge the scale of the graph axes of that principal component vector to make it easier to recognize smaller changes in measured values. Alternatively, the plotting unit 15 may increase the amount of change in the magnitude of points related to a particular principal component vector to make it easier to recognize smaller changes in measured values.

[0041] The drawing unit 15 may be configured to distinguish between the projection data of a reference odor sample and the projection data of an odor sample that has a predetermined difference. For example, when the odor analyzer 1 is used to distinguish between genuine and counterfeit perfumes, the drawing unit may distinguish between the projection data of a genuine perfume sample and an odor sample whose projection data has a predetermined difference and display it as a counterfeit. For example, when the drawing unit 15 determines that an odor sample is counterfeit, it may represent the measured value in a specific color or sound an alarm from a speaker (not shown).

[0042] The setting provision unit 21 provides setting information related to the analysis to the odor analyzer 1. The setting information related to the analysis includes, for example, information related to the method of extracting the principal components (for example, which sensor's measurement value will be used as the principal component, or whether to prioritize calculations using the principal component). The setting information related to the analysis may also include the method of plotting the projection data as described above (for example, the plotting angle of the graph, the method of changing the plotting scale, the coloring method, etc.).

[0043] The above-described functional units of the odor analyzer 1 are merely examples of functions and do not limit the functions of the odor analyzer 1. For example, the odor analyzer 1 does not need to have all of the above functional units, and may have only some of them. Furthermore, the odor analyzer 1 may have other functions not described above. For example, the functional units of the odor analyzer 1 may be implemented in the server 2.

[0044] Furthermore, as described above, each of the above functional units has been explained as being implemented by software. However, at least one of the above functional units may be implemented by hardware.

[0045] Furthermore, any of the above functional units may be implemented by dividing one functional unit into multiple functional units. Alternatively, any two or more of the above functional units may be combined into a single functional unit. Figure 1 represents the functions of the odor analyzer 1 using functional blocks, and does not indicate, for example, that each functional unit is composed of separate program files or the like.

[0046] Furthermore, the odor analyzer 1 may be a device implemented in a single housing, or it may be a system implemented from multiple devices connected via a network or the like. For example, the odor analyzer 1 may implement some or all of its functions using other virtual devices, such as cloud services provided by a cloud computing system. In other words, the odor analyzer 1 may implement at least one of the above-mentioned functional units using other devices.

[0047] Next, using Figure 2, we will explain an example of the display on the display device of the odor analyzer 1.

[0048] Figure 2 shows an example of the appearance of the odor analyzer 1 in an embodiment.

[0049] In Figure 2, the odor analyzer 1 has an odor sensor mounting section 16. The odor sensor mounting section 16 shown in Figure 2 illustrates a case where nine odor sensors 161 are mounted. In Figure 9, a code is assigned to only one sensor, and the assignment of codes to the other sensors is omitted. The number of odor sensors n mounted on the odor analyzer 1 is any integer greater than or equal to 1, for example, 32 sensors may be mounted. Although Figure 2 illustrates an odor analyzer 1 in which odor sensors and other components are mounted on a single substrate, the odor analyzer 1 may have the odor sensor mounting section 16 independently on a separate substrate.

[0050] The odor sensors 161 mounted on the odor sensor mounting section 16 are sensors each equipped with a different type of thin film. However, the odor sensor mounting section 16 may be equipped with multiple odor sensors 161 each equipped with the same thin film. For example, in an odor sensor mounting section 16 with n=8, if four odor sensors 161 each having the same type of thin film are mounted, the number of mounted odor sensors 161 will be 8 × 4 = 32. For example, by obtaining the measurement value by taking the average value of the measurement values ​​of odor sensors equipped with the same thin film, the measurement error due to individual differences in the odor sensors 161 can be reduced. In addition, by mounting multiple odor sensors 161 equipped with the same thin film, the measurement error due to the mounting position of the odor sensor 161 (the position where the odor is measured) can be reduced.

[0051] Next, we will explain an example of displaying the projection data drawn by the odor analyzer 1 using Figures 3 and 4. Figure 3 is a diagram showing a first display example of the odor analyzer in the embodiment.

[0052] In Figure 3, display screen 1000 displays a graph of two-dimensional projected data, where the number m of the principal components is set to 2 (m=2), and the measured values ​​are projected onto the principal component vectors of the two axes. Display screen 1000 is an example of a display where the plotting scale of the projected data is changed for each dimension. The plotting scale is the display magnification of the graph axis, and display screen 1000 displays a magnified view of a specific part of the graph axis. For example, in Figure 3, the projected data for the principal components is distributed between approximately -2 and 2 (not shown). Display screen 1000 magnifies the portion of the plotting scale of the horizontal graph axis between -0.75 and 1.5. Also, display screen 1000 magnifies the portion of the plotting scale of the vertical graph axis between -0.75 and 1.25. By changing the plotting scale of the projected data for each dimension, it becomes possible to display the projected data while focusing on the parts that show odor characteristics, making the odor characteristics easier to visualize.

[0053] The display screen 1000 shows the measured values ​​of odor sample A, odor sample B, and odor sample C. Odor sample A is a genuine perfume that is legally manufactured and sold. Odor samples B and C are perfumes to be judged as to whether they are genuine or counterfeit. The measured values ​​of odor sample A are distributed within the range of measurement value group 1002, centered at the center point 1001. In this embodiment, the method for calculating the center point uses the arithmetic mean calculated from the average of the simple sums of the measured values. However, the method for calculating the center point may also use a weighted average, geometric mean, or harmonic mean. Furthermore, the center point may be calculated by weighting the measured values ​​of a specific sensor among multiple sensors. For example, the center point may be calculated by weighting the measured values ​​of three specific odor sensors among 19 odor sensors by twice the weight of the measured values ​​of the other odor sensors. The diameter of the center point 1001 indicates the magnitude of the variation. The measured values ​​for odor sample B are distributed approximately within the range of measurement group 1004, centered at center point 1003. The diameter of center point 1003 indicates the magnitude of the variation. Similarly, the measured values ​​for odor sample C are distributed approximately within the range of measurement group 1010, centered at center point 1009. The diameter of center point 1010 indicates the magnitude of the variation.

[0054] The display screen 1000 clearly displays the projection data of a reference odor sample A and the projection data of an odor sample B that has a predetermined difference. For example, since measurement group 1002 and measurement group 1004 do not overlap, the user can recognize that there is a difference between the measurement values. Also, since the magnitude of the variability in measurement group 1002 and the magnitude of the variability in measurement group 1004 are different, the user can recognize that there is a difference between the measurement values. Furthermore, the user can recognize that there is a difference between the measurement values ​​from the distance between center point 1001 and center point 1003. Differences in measurement values, such as the degree of overlap between measurement groups, the magnitude of variability, or the distance between center points, can be calculated as numerical values. For example, the degree of overlap between measurement groups can be calculated using a statistical value called the effect size, which is obtained by dividing the difference in the mean values ​​of each measurement by the magnitude of variability (standard deviation, etc.). Similarly, the display screen 1000 displays the projection data of a reference odor sample A and the projection data of an odor sample C that has a predetermined difference, in a way that allows for identification.

[0055] The display screen 1000 may display a warning message 1005 if the difference in the measured values ​​described above is greater than (or exceeds) a predetermined value. The warning message 1005 is, for example, a warning message indicating that the item being judged is counterfeit. The content of the warning message 1005 may change depending on the magnitude of the difference in the measured values. For example, if the difference in the measured values ​​is close to a predetermined value where it is difficult to determine whether it is genuine or counterfeit, the warning message 1005 may prompt remeasurement or display a message such as "identical with a probability of aa%" or "effect size bb". Users can recognize that there is a difference between the measured values ​​by the warning message. The display screen 1000 may also display a message such as "identical", "genuine", "identical with a probability of cc%" or "effect size dd" if the difference in the measured values ​​described above is less than (or less than) a predetermined value.

[0056] Furthermore, the display screen 1000 can display projection data of multiple odor samples in a way that allows for relative comparison. For example, if there is variation in the odor of genuine perfume between manufacturing lots, or if the odor changes over time, the projection data of the standard genuine product will change, making it impossible to determine whether or not it is genuine based solely on the difference (absolute value) between the projection data mentioned above. The display screen 1000 can display the projection data of odor sample A, the projection data of odor sample B, and the projection data of odor sample C, allowing the user to recognize which odor sample's projection data is relatively close to the projection data of odor sample A on the graph.

[0057] For example, display screen 1000 can show that the distance between center point 1001 and center point 1003 is relatively larger than the distance between center point 1001 and center point 1009. The user can recognize the relative comparison between the distance between center point 1001 and center point 1003 and the distance between center point 1001 and center point 1009, rather than the absolute value of the distance between center point 1001 and center point 1003. As a result, the user can recognize that the difference in measured values ​​between odor sample A and odor sample B is relatively larger than the difference in measured values ​​between odor sample A and odor sample C. Similarly, display screen 1000 can show that the distance between measured value group 1002 and measured value group 1004 is relatively larger than the distance between measured value group 1002 and measured value group 1010. As a result, the user can recognize that the difference in measured values ​​between odor sample A and odor sample B is relatively larger than the difference in measured values ​​between odor sample A and odor sample C. This allows the user to recognize that odor sample B is counterfeit and that odor sample C is genuine. Furthermore, if it is known from the outset that odor sample C is a genuine product with a different manufacturing lot or manufacturing date, it becomes possible to judge odor sample B while taking into account variations in manufacturing lots, etc. In addition, display screen 1000 can display projection data even if the number of odor samples is four or more.

[0058] The display screen 1000 also has buttons 1006, 1007, and 1008. Button 1006 is for displaying one-dimensional projected data (data distribution on a straight line) obtained by projecting the measured values ​​onto a principal component vector of one axis, with the number of principal components m set to 1 (m=1). Button 1007 is for displaying two-dimensional projected data (data distribution on a plane) obtained by projecting the measured values ​​onto a principal component vector of two axes, with the number of principal components m set to 2 (m=2). Button 1008 is for displaying three-dimensional projected data (data distribution in space) obtained by projecting the measured values ​​onto a principal component vector of three axes, with the number of principal components m set to 3 (m=3). Figure 3 shows that button 1007 is pressed.

[0059] In Figure 3, an example is shown of a method for comparing multiple odor samples by calculating the average value of the measured values ​​and comparing the distances between the center points. However, odor samples may be compared using calculation methods other than calculating the average value. For example, the sum of the measured values ​​may be calculated and the sums of these sums may be compared. The sum may be calculated, for example, by simple addition, or by weighted addition of the measured values ​​from specific odor sensors.

[0060] Figure 4 shows a second display example of the odor analyzer in the embodiment.

[0061] Figure 4, similar to Figure 3, displays a graph of two-dimensional projected data where the number of principal components m is set to 2 (m=2), and the measured values ​​are projected onto the principal component vectors of the two axes. In addition, display screen 2000 magnifies the portion of the graph where the horizontal axis plotting scale is -1.0 to 1.2, and the vertical axis plotting scale is -0.75 to 1.1.

[0062] Display screen 2000 shows the measured values ​​indicating the change in odor over time for odor sample A and odor sample B. Odor sample A is a food product without additives, while odor sample B is the same food product with additives added. In other words, display screen 2000 shows the difference in the measured values ​​indicating the change in odor due to the presence or absence of additives.

[0063] Furthermore, the display screen 2000 has buttons 2003, 2004, and 2005. The functions of buttons 2003, 2004, and 2005 are the same as those of buttons 1005, 1006, and 1007, respectively, as described in Figure 3, so their descriptions are omitted.

[0064] Odor sample A shows variability in measured values ​​within the range of measurement value group 2001. Similarly, odor sample B shows variability in measured values ​​within the range of measurement value group 2002. Display screen 2000 shows that there is no overlap between measurement value group 2001 and measurement value group 2002, and that the variability is different. From the display on display screen 2000, users can visually confirm that there are differences in the change in odor caused by adding additives to food.

[0065] Next, the hardware configuration of the odor analyzer 1 will be explained using Figure 5. Figure 4 is a block diagram showing an example of the hardware configuration of the odor analyzer 1 in the embodiment. Note that the hardware configuration of the server 2 is the same as that of the odor analyzer 1, and therefore its explanation will be omitted.

[0066] The odor analyzer 1 has a CPU (Central Processing Unit) 101, RAM (Random Access Memory) 102, ROM (Read Only Memory) 103, I / O device 104, and communication I / F (Interface) 105. The odor analyzer 1 is a device that executes the information processing program described in Figure 1.

[0067] The CPU 101 controls the user terminal by executing information processing programs stored in RAM 102 or ROM 103. The information processing programs are obtained, for example, from a recording medium on which the programs are stored, or from a program distribution server via a network, installed in ROM 103, read by the CPU 101, and executed.

[0068] The I / O device 104 has an operation input function and a display function (operation display function). The I / O device 104 is, for example, a touch panel. The touch panel enables users of the information processing terminal 10 to perform operation input using their fingertips or a stylus. The I / O device 104 may be an integrated unit of a display device having a display function and an operation input device having an operation input function, or it may consist of a display device having a display function and an operation input device having an operation input function separately. The display screen of the touch panel can be performed as the display screen of the display device, and the operation of the touch panel can be performed as the operation of the operation input device. The I / O device 104 may be implemented in various forms such as a head-mounted type, glasses type, or wristwatch type display.

[0069] Communication I / F 105 is a communication interface. Communication I / F 105 performs short-range wireless communication such as wireless LAN, wired LAN, and infrared. The diagram shows only communication I / F 105 as a communication interface, but the information processing terminal 10 may have multiple communication interfaces for each communication method.

[0070] Next, the operation of the odor analyzer 1 will be explained using Figure 6. Figure 6 is a flowchart illustrating the operation of the odor analyzer 1 in this embodiment.

[0071] In Figure 6, the odor analyzer 1 starts measurement (step S11). Measurement is started, for example, when the user performs a measurement start operation on the odor analyzer 1 (for example, by pressing the start button).

[0072] After performing the process in step S11, the odor analyzer 1 acquires measurement values ​​(step S12). The acquisition of measurement values ​​may be performed according to the number of odor samples and a predetermined number of measurements. For example, if there are 2 odor samples, 3 measurements, and 9 sensors (n=9), then in the process of step S12, 2 × 3 × 9 = 42 measurement values ​​will be acquired.

[0073] After performing the process in step S12, the odor analyzer 1 extracts the main components. The extraction of the main components can be performed in a predetermined number of main components (m).

[0074] After executing the process in step S12, the odor analyzer 1 projects the n-dimensional data onto an m-dimensional principal component vector based on the m-dimensional principal components extracted in the extraction unit (step S14). The odor analyzer 1 may also extract principal components such that the difference in projected data between the odor samples being compared is maximized, and then project the measured values. In that case, the processes in steps S13 to S14 are repeated until the maximum value of the difference in projected data is calculated.

[0075] After performing the process in step S14, the odor analyzer 1 plots the m-dimensional projection data projected in step S14 (step S15). The plotting of the projection data may be performed in such a way that the difference in the projection data is maximized. As mentioned above, the difference in the projection data varies depending on which measurement values ​​are extracted as principal components. In the process of step S15, the projection data may be plotted according to the combination of extracted principal components.

[0076] After executing the process in step S15, the odor analyzer 1 displays the projection data drawn in step S15 on its display device or the like (step S16). In the process of step S16, the odor analyzer 1 may display the difference in projection data calculated in the process of step S14. Alternatively, the odor analyzer 1 may display the projection data corresponding to the combination of principal components drawn in the process of step S15. The display of projection data may be performed, for example, on a terminal (not shown) connected via the network 9. After executing the process of step S16, the odor analyzer 1 terminates the operation shown in the flowchart.

[0077] The flowchart shown is an example of the operation and does not limit the possible operations.

[0078] Furthermore, the various processes described above in this embodiment may be performed by recording a program for realizing the functions of the device described in this embodiment onto a computer-readable recording medium, loading the program recorded on the recording medium into a computer system, and executing it. The term "computer system" here may include hardware such as an operating system and peripheral devices. Also, if a WWW system is being used, the "computer system" shall also include the homepage provisioning environment (or display environment). Furthermore, "computer-readable recording medium" refers to a storage device such as a flexible disk, magneto-optical disk, ROM, flash memory, portable media such as a CD-ROM, or a hard disk built into a computer system.

[0079] Furthermore, "computer-readable recording media" includes volatile memory (e.g., DRAM (Dynamic Random Access Memory)) within computer systems that act as servers or clients when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, which retains the program for a certain period of time. In addition, the above program may be transmitted from the computer system that stores the program in a memory device, etc., to another computer system via a transmission medium or by transmission waves within the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium that has the function of transmitting information, such as a network such as the Internet or a communication line such as a telephone line. Furthermore, the above program may be for the purpose of realizing a part of the above-mentioned functions. In addition, it may be a so-called differential file (differential program) that realizes the above-mentioned functions in combination with a program already recorded in the computer system.

[0080] While embodiments of the present invention have been described above with reference to the drawings, the specific configuration is not limited to these embodiments, and various modifications are possible without departing from the spirit of the present invention. [Explanation of Symbols]

[0081] 1. Odor analyzer 11 Measurement Unit 12 Acquisition Department 13 Extraction part 14 Projection section 15 Drawing section 16 Odor sensor mounting section 2 servers 21 Settings provider 9 Network 101 CPU 102 RAM 103 ROM 104 I / O equipment 105 Communication I / F 1000 display screen 1001 center point 1002 Measurement Value Group 1003 Center point 1004 Measurement Value Group 1005 Warning display 1006 Buttons 1007 Buttons 1008 buttons 1009 Center point 1010 Measurement Value Group 2000 display screen 2001 Measurement Group 2002 Measurement Group 2003 Button 2004 Button 2005 Button

Claims

1. A measuring unit having n odor sensors (where n is an integer of 1 or more), An acquisition unit that acquires measured values ​​measured by the n odor sensors of the measurement unit, An extraction unit extracts m odor components (where m is an integer greater than or equal to 1, and n ≥ m) as main components from n odor components contained in n-dimensional data, which is one-dimensional in which the measurement value of one of the odor sensors acquired by the acquisition unit, A projection unit projects n-dimensional data into m-dimensional data based on the principal components extracted in the extraction unit, A drawing unit that draws the m-dimensional projection data projected in the projection unit. An odor analyzer equipped with [specific features / equipment].

2. The odor sensor changes its natural frequency when odorous substances are adsorbed onto the adsorption part. The odor analyzer according to claim 1, wherein the measurement unit measures the natural frequency.

3. The odor analyzer according to claim 1 or 2, wherein the projection unit projects the n-dimensional data into m dimensions by projecting it into a space represented by the principal component vectors of the m principal components.

4. The odor analyzer according to any one of claims 1 to 3, wherein the extraction unit extracts the main components such that the difference between the projection data in the first odor sample and the projection data in the second odor sample is maximized.

5. The odor analyzer according to any one of claims 1 to 4, wherein the drawing unit draws the projection data by changing the drawing scale for each dimension.

6. The odor analyzer according to any one of claims 1 to 4, wherein the drawing unit draws in a manner that allows for the identification of projection data of a reference odor sample and projection data of an odor sample having a predetermined difference.

7. A measurement step in which odor is measured using n odor sensors (where n is an integer greater than or equal to 1), The measurement step includes an acquisition step in which the measured values ​​measured by n odor sensors are acquired, An extraction step is performed to extract m odor components (where m is an integer greater than or equal to 1) as main components from the n odor components contained in the n-dimensional data, which is one-dimensional in which the measurement value of one of the odor sensors acquired in the acquisition step is one-dimensional. A projection step in which n-dimensional data is projected onto m-dimensional data based on the principal components extracted in the extraction step, A drawing step is performed to draw the m-dimensional projection data projected in the projection step. A method for analyzing odors, including [specific details omitted].

8. On the computer, A measurement process that measures odor using n odor sensors (where n is an integer greater than or equal to 1), The measurement process includes an acquisition process to acquire the measured values ​​measured by n odor sensors, An extraction process is performed to extract m odor components (where m is an integer greater than or equal to 1) as principal components from the n odor components contained in the n-dimensional data, which is one-dimensional in which the measurement value of one of the odor sensors acquired in the acquisition process. A projection process is performed to project n-dimensional data into m-dimensional data based on the principal components extracted in the extraction process. A drawing process for drawing the m-dimensional projection data projected in the aforementioned projection process. A scent analysis program to be run on a computer.

Images