Functional component measuring method, functional component measuring unit and functional component measuring program

Abstract

To provide a functional component measuring method capable of quick and accurate measurement when measuring a functional component and sugar content of vegetable and / or vegetable processed product.SOLUTION: A functional component measuring method for measuring a content of a functional component and sugar content of vegetable and / or a vegetable processed product comprises: a spectroscopy absorption spectrum acquiring step of applying light-source light to vegetable and / or vegetable processed product to be measured, and acquiring a spectroscopy absorption spectrum of visible light and near infrared light by detecting diffuse reflectance by interactance measurement; and an estimation model generating step of performing multivariable analysis on the spectroscopy absorption spectrum of the visible and near infrared light wavelength regions acquired in the spectroscopy absorption spectrum acquiring step, thereby generating an estimation model for estimating a content of a functional component and an estimation model for estimating a sugar content.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a functional component measuring method, a functional component measuring device, and a functional component measuring program. [Background technology]

[0002] Recently, lycopene contained in tomatoes and processed tomato products has been attracting attention as a functional ingredient. Under the functional food labeling system that began in April 2015, tomato juice and tomato fruit containing high concentrations of lycopene have been notified as functional food. Under the functional food labeling system, not only are notifiers required to set the ingredient content per recommended daily intake of a food, but also to check (monitor) the amount of the ingredient contained in the food even after notification. In addition, the sugar content of the raw material (tomato fruit) is checked at tomato processing sites.

[0003] However, lycopene is evaluated by measuring light, and sugar content is evaluated by measuring refractive index, so each method requires dedicated analytical equipment and labor.

[0004] Measurements using light allow for rapid measurements because the speed of light is fast. When performing rapid optical measurements of the content of quality components in agricultural products and foods, the use of visible light, near-infrared light, infrared light, etc. can be considered, taking into account factors such as cost. Since lycopene is a pigment, it is possible to use near-infrared light and infrared light in addition to visible light, but the concentration of lycopene in tomatoes is at its highest, at 17 mg / 100 g, and in processed tomatoes at around 35 mg / 100 g.

[0005] Therefore, even if it is possible to measure with near-infrared or infrared light, the information on the lycopene concentration may be small. A rapid quantitative method for lycopene in crushed tomato fruit and processed tomato products using infrared spectroscopy has been reported (Non-Patent Documents 1 and 2), but since the weak absorption of lycopene overlaps with the strong absorption of sugar, it may be an indirect measurement method that captures the internal correlation between lycopene and sugar (sugar content). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent No. 4621876

[0007] [Non-Patent Document 1] M. Baranska, W. Schutze, H. Schulz: Anal. Chem., 78, 8456 (2006). [Non-Patent Document 2] Y. Halim, S. J. Schwartz, D. Francis, NA Baldauf, LE Rodriguez-Saona: J. AOAC International, 89, 1257 (2006). [Non-Patent Document 3] Hidekazu Ito, Tatsuyuki Hannaka, Analytical Chemistry, 68, pp513-516 (2019). Summary of the Invention [Problem to be solved by the invention]

[0008] The present invention has been made in consideration of the above, and aims to provide a functional component measuring method, a functional component measuring device, and a functional component measuring program that are capable of measuring functional components and sugar content in fresh produce and / or processed fresh produce products quickly and with high accuracy. [Means for solving the problem]

[0009] In order to solve the above-mentioned problems and achieve the object, the present invention provides a functional ingredient measuring method for measuring the content and sugar content of functional ingredients in fresh produce and / or processed fresh produce products, comprising a spectral absorption spectrum acquiring step of irradiating the fresh produce and / or processed fresh produce products to be measured with light from a light source and detecting the diffuse reflected light by an interactance measurement method to acquire a spectral absorption spectrum of visible and near-infrared light, and an estimation model creating step of creating an estimation model for estimating the content of functional ingredients and an estimation model for estimating the sugar content by performing multivariate analysis on the spectral absorption spectrum in the visible and near-infrared wavelength range acquired in the spectral absorption spectrum acquiring step.

[0010] In the spectral absorption spectrum acquisition process, the fresh produce and / or processed fruit or vegetable products may be placed directly on a sample stage, or may be placed in a sample cell and the sample cell may be placed on the sample stage, and the light source light may be irradiated from the sample stage directly onto the fresh produce and / or processed fruit or vegetable products, or onto the sample cell.

[0011] According to another aspect of the present invention, in the estimation model creating step, a PLS regression analysis or a multiple regression analysis may be performed as the multivariate analysis to create a regression equation as the estimation model.

[0012] According to one aspect of the present invention, in the estimation model creation step, an estimation model for estimating the content of the functional ingredient may be created using a spectral absorption spectrum in the visible light wavelength range or the visible / near-infrared wavelength range among the spectral absorption spectra in the visible / near-infrared wavelength range acquired in the spectral absorption spectrum acquisition step, and an estimation model for estimating the sugar content may be created using a spectral absorption spectrum in the near-infrared wavelength range.

[0013] According to one aspect of the present invention, the method may further include a data pre-processing step of performing one or more of the following on the spectroscopic absorption spectrum: centering, standardization, normalization, differentiation, baseline correction, and smoothing.

[0014] Furthermore, according to one aspect of the present invention, the method may further include an estimation step of applying the spectral absorption spectrum in the visible / near-infrared wavelength range acquired in the spectral absorption spectrum acquisition step to an estimation model for estimating the content of the functional ingredient and an estimation model for estimating the sugar content, to estimate the content and sugar content of the functional ingredient in the fruit or vegetable and / or processed fruit or vegetable product to be measured.

[0015] Furthermore, according to one aspect of the present invention, in the estimation step, of the spectral absorption spectra in the visible and near-infrared wavelength range acquired in the spectral absorption spectrum acquisition step, a spectral absorption spectrum in the visible wavelength range or the visible and near-infrared wavelength range may be applied to an estimation model for estimating the content of the functional ingredient to estimate the content of the functional ingredient in the fruit or vegetable and / or processed fruit or vegetable product being measured, and a spectral absorption spectrum in the near-infrared wavelength range may be applied to an estimation model for estimating the sugar content to estimate the sugar content in the fruit or vegetable and / or processed fruit or vegetable product being measured.

[0016] According to one embodiment of the present invention, the functional ingredient may include lycopene, a carotenoid, or anthocyanin (polyphenol).

[0017] According to another aspect of the present invention, the fruit or vegetable may be a tomato fruit or a tomato liquid, the processed fruit or vegetable product may be a tomato processed product, and the functional ingredient may be lycopene.

[0018] According to one aspect of the present invention, the fruit or vegetable may be a citrus fruit or a citrus liquid, the processed fruit or vegetable product may be a citrus processed product, and the functional ingredient may be a carotenoid.

[0019] According to another aspect of the present invention, the fruit or vegetable may be a melon, and the functional ingredient may be β-carotene.

[0020] According to another aspect of the present invention, the fruit or vegetable may be a tomato fruit, and the functional ingredient may be lycopene.

[0021] According to another aspect of the present invention, the fruit or vegetable may be grapes, and the functional ingredient may be anthocyanin.

[0022] In addition, in order to solve the above-mentioned problems and achieve the object, the present invention provides a functional ingredient measuring device that measures the content and sugar content of functional ingredients in fresh produce and / or processed fresh produce products, and is characterized by including a spectral absorption spectrum acquisition means that irradiates the fresh produce and / or processed fresh produce products to be measured with light from a light source and detects the diffuse reflected light by an interactance measurement method to acquire a visible / near-infrared spectral absorption spectrum, and an estimation model creation means that creates an estimation model for estimating the content of functional ingredients and an estimation model for estimating the sugar content by performing multivariate analysis on the spectral absorption spectrum in the visible / near-infrared wavelength range acquired by the spectral absorption spectrum acquisition means.

[0023] Furthermore, according to one aspect of the present invention, the spectral absorption spectrum acquisition means may place the fresh produce and / or processed fruit or vegetable products directly on a sample stage, or may fill a sample cell and place the sample cell on the sample stage, and irradiate the light source light from the sample stage directly onto the fresh produce and / or processed fruit or vegetable products, or onto the sample cell.

[0024] According to another aspect of the present invention, the estimation model creation means may perform a PLS regression analysis or a multiple regression analysis as the multivariate analysis to create a regression equation as the estimation model.

[0025] According to one aspect of the present invention, the estimation model creation means creates an estimation model for estimating the content of the functional ingredient by using a spectral absorption spectrum in the visible light wavelength range or the visible / near-infrared wavelength range among the spectral absorption spectra in the visible / near-infrared wavelength range acquired by the spectral absorption spectrum acquisition means, and may also create an estimation model for estimating the sugar content by using a spectral absorption spectrum in the near-infrared wavelength range.

[0026] According to one aspect of the present invention, the apparatus may further include a data pre-processing means for performing one or more of the following processes on the spectroscopic absorption spectrum: centering, standardization, normalization, differentiation, baseline correction, and smoothing.

[0027] Furthermore, according to one aspect of the present invention, the device may further include an estimation means for estimating the content and sugar content of the functional ingredient in the fruit or vegetable and / or processed fruit or vegetable product to be measured by applying the spectral absorption spectrum in the visible / near-infrared wavelength range acquired by the spectral absorption spectrum acquisition means to an estimation model for estimating the content of the functional ingredient and an estimation model for estimating the sugar content.

[0028] According to one aspect of the present invention, the estimation means estimates the content of the functional ingredient in the fruit or vegetable and / or processed fruit product to be measured by applying a spectral absorption spectrum in the visible light wavelength range or the visible light wavelength range to the near-infrared light wavelength range among the spectral absorption spectra in the visible and near-infrared light wavelength range acquired by the spectral absorption spectrum acquisition means to an estimation model for estimating the content of the functional ingredient, and estimates the sugar content in the fruit or vegetable and / or processed fruit product to be measured by applying a spectral absorption spectrum in the near-infrared light wavelength range to an estimation model for estimating the sugar content. Good too.

[0029] According to one embodiment of the present invention, the functional ingredient may include lycopene, a carotenoid, or anthocyanin (polyphenol).

[0030] According to another aspect of the present invention, the fruit or vegetable may be a tomato fruit or a tomato liquid, the processed fruit or vegetable product may be a tomato processed product, and the functional ingredient may be lycopene.

[0031] According to one aspect of the present invention, the fruit or vegetable may be a citrus fruit or a citrus liquid, the processed fruit or vegetable product may be a citrus processed product, and the functional ingredient may be a carotenoid.

[0032] According to another aspect of the present invention, the fruit or vegetable may be a melon, and the functional ingredient may be β-carotene.

[0033] According to another aspect of the present invention, the fruit or vegetable may be a tomato fruit, and the functional ingredient may be lycopene.

[0034] According to another aspect of the present invention, the fruit or vegetable may be grapes, and the functional ingredient may be anthocyanin.

[0035] In addition, in order to solve the above-mentioned problems and achieve the object, the present invention is a functional ingredient measurement program for measuring the content and sugar content of functional ingredients in fresh produce and / or processed fresh produce products, which is a functional ingredient measurement method that causes a computer to execute a spectral absorption spectrum acquisition step of detecting diffuse reflected light from the fresh produce and / or processed fresh produce products to be measured in response to light irradiated from a light source using an interactance measurement method and acquiring the spectral absorption spectrum, and an estimation model creation step of creating an estimation model for estimating the content of functional ingredients and an estimation model for estimating the sugar content by performing multivariate analysis on the spectral absorption spectrum in the visible / near-infrared light wavelength range acquired in the spectral absorption spectrum acquisition step.

[0036] Furthermore, according to one aspect of the present invention, in the spectral absorption spectrum acquisition process, the fresh produce and / or processed fruit or vegetable products may be placed directly on a sample stage, or may be placed in a sample cell and the sample cell may be placed on the sample stage, and the light source light may be irradiated from the sample stage directly onto the fresh produce and / or processed fruit or vegetable products, or onto the sample cell.

[0037] According to another aspect of the present invention, in the estimation model creating step, a PLS regression analysis or a multiple regression analysis may be performed as the multivariate analysis to create a regression equation as the estimation model.

[0038] According to one aspect of the present invention, in the estimation model creation step, an estimation model for estimating the content of the functional ingredient may be created using a spectral absorption spectrum in the visible light wavelength range or the visible / near-infrared wavelength range among the spectral absorption spectra in the visible / near-infrared wavelength range acquired in the spectral absorption spectrum acquisition step, and an estimation model for estimating the sugar content may be created using a spectral absorption spectrum in the near-infrared wavelength range.

[0039] According to one aspect of the present invention, the computer may further perform a data pre-processing step of performing one or more of the following on the spectroscopic absorption spectrum: centering, standardization, normalization, differentiation, baseline correction, and smoothing.

[0040] According to another aspect of the present invention, a computer is further configured to execute the spectroscopic absorption spectrum acquisition. The spectral absorption spectrum in the visible / near-infrared wavelength range acquired in the obtaining step may be applied to an estimation model for estimating the content of the functional ingredient and an estimation model for estimating the sugar content, thereby executing an estimation step of estimating the content and sugar content of the functional ingredient in the fruit or vegetable and / or processed fruit or vegetable product to be measured.

[0041] Furthermore, according to one aspect of the present invention, in the estimation step, of the spectral absorption spectra in the visible and near-infrared wavelength range acquired in the spectral absorption spectrum acquisition step, a spectral absorption spectrum in the visible wavelength range or the visible and near-infrared wavelength range may be applied to an estimation model for estimating the content of the functional ingredient to estimate the content of the functional ingredient in the fruit or vegetable and / or processed fruit or vegetable product being measured, and a spectral absorption spectrum in the near-infrared wavelength range may be applied to an estimation model for estimating the sugar content to estimate the sugar content in the fruit or vegetable and / or processed fruit or vegetable product being measured.

[0042] According to one embodiment of the present invention, the functional ingredient may include lycopene, a carotenoid, or anthocyanin (polyphenol).

[0043] According to another aspect of the present invention, the fruit or vegetable may be a tomato fruit or a tomato liquid, the processed fruit or vegetable product may be a tomato processed product, and the functional ingredient may be lycopene.

[0044] According to one aspect of the present invention, the fruit or vegetable may be a citrus fruit or a citrus liquid, the processed fruit or vegetable product may be a citrus processed product, and the functional ingredient may be a carotenoid.

[0045] According to another aspect of the present invention, the fruit or vegetable may be a melon, and the functional ingredient may be β-carotene.

[0046] According to another aspect of the present invention, the fruit or vegetable may be a tomato fruit, and the functional ingredient may be lycopene.

[0047] According to another aspect of the present invention, the fruit or vegetable may be grapes, and the functional ingredient may be anthocyanin. Effect of the Invention

[0048] According to the present invention, when measuring functional components and sugar content in fruits and vegetables and / or processed fruits and vegetables products, it is possible to perform the measurement quickly and with high accuracy. [Brief description of the drawings]

[0049] [Figure 1] FIG. 1 is an explanatory diagram for explaining an outline of the present invention. [Diagram 2] FIG. 2 is a flowchart for explaining the functional component measuring method according to the present embodiment. [Diagram 3] FIG. 3 is a diagram showing an example of the external configuration of the functional component measuring device according to the present embodiment. [Figure 4] FIG. 4 is a block diagram showing an example of the configuration of a functional component measuring device according to this embodiment. [Diagram 5] FIG. 5 is a diagram showing an example of the configuration of the sample stage. [Figure 6] FIG. 6 is a diagram showing an example of the configuration of the main body of the sample cell. [Figure 7] FIG. 7 is a diagram showing an example of the configuration of a cover of a sample cell. [Figure 8] FIG. 8 is a diagram showing the visible / near-infrared spectroscopic absorption spectra (500 nm to 1000 nm) of a tomato fluid and a tomato processed product. [Figure 9]FIG. 9 is a diagram showing the correlation (PLS regression analysis) between the measured values ​​and the estimated values ​​of the lycopene content in tomato fluids and tomato processed products. [Figure 10] FIG. 10 is a diagram showing the correlation (PLS regression analysis) between the actually measured and estimated sugar contents of tomato fluids and tomato processed products. [Figure 11] FIG. 11 is a diagram showing the visible / near-infrared spectroscopic absorption spectra (500 nm to 1000 nm) of citrus fluids and citrus processed products. [Figure 12] FIG. 12 is a diagram showing the correlation (PLS regression analysis) between the measured values ​​and the estimated values ​​of the carotenoid contents in citrus fluids and citrus processed products. [Figure 13] FIG. 13 is a diagram showing the correlation (PLS regression analysis) between the measured and estimated sugar contents of citrus fluids and citrus processed products. [Figure 14] FIG. 14 is a diagram showing the visible / near-infrared spectroscopic absorption spectrum (500 nm to 1000 nm) of citrus fruits. [Figure 15] FIG. 15 is a diagram showing the correlation (PLS regression analysis) between the actually measured values ​​and the non-destructively measured values ​​(estimated values) of the carotenoid contents in citrus fruits. [Figure 16] FIG. 16 is a diagram showing the correlation (PLS regression analysis) between the actually measured sugar content values ​​and the non-destructively measured values ​​(estimated values) of citrus fruits. [Figure 17] FIG. 17 is a diagram showing the visible / near-infrared spectroscopic absorption spectrum (500 nm to 992 nm) of melon fruit (red flesh). [Figure 18] FIG. 18 is a graph showing the correlation (PLS regression analysis) between the actually measured values ​​and the non-destructively measured values ​​(estimated values) of the β-carotene content in melon fruit (red flesh). [Figure 19] FIG. 19 is a diagram showing the visible / near-infrared spectroscopic absorption spectrum (500 nm to 1000 nm) of a tomato fruit. [Figure 20] FIG. 20 is a diagram showing the correlation (PLS regression analysis) between the actually measured values ​​and the non-destructively measured values ​​(estimated values) of the lycopene content in tomato fruits. [Figure 21] FIG. 21 is a diagram showing the correlation (PLS regression analysis) between the actually measured values ​​and non-destructively measured values ​​(estimated values) of sugar content in tomato fruits. [Figure 22] FIG. 22 is a diagram showing the visible / near-infrared spectroscopic absorption spectrum (500 nm to 1000 nm) of grape berries. [Diagram 23] FIG. 23 is a graph showing the correlation (PLS regression analysis) between the actually measured values ​​and the non-destructively measured values ​​(estimated values) of the anthocyanin (converted into cyanidin) content in grape berries. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0050] Preferred embodiments of the functional component measuring method, functional component measuring device, and functional component measuring program according to the present invention will be described in detail below with reference to Fig. 1 to Fig. 23. Note that the present invention is not limited to these embodiments.

[0051] [Outline of the present invention] First, an outline of the present invention will be described with reference to Fig. 1. Fig. 1 is an explanatory diagram for explaining an outline of the present invention.

[0052] The present invention provides a functional component measurement method, a functional component measurement device, and a functional component measurement program that enable rapid and highly accurate measurement of functional components and sugar content in fresh produce and / or processed fresh produce products.

[0053] In the present invention, for example, since the functional ingredient lycopene is a pigment, visible spectroscopy or visible / near infrared spectroscopy, which has high sensitivity and low cost, was selected as the optimal spectroscopic method. On the other hand, for the sugar content in fruit, the near infrared light range (750 nm to 1100 nm) adjacent to the visible light range was selected.

[0054] Methods for measuring the spectroscopic absorption spectrum of a sample include the transmitted light measurement method (Non-Patent Document 3) and the diffuse reflected light measurement method. The former requires highly sensitive measurements, which tends to make the equipment expensive. On the other hand, the diffuse reflected light measurement method is suitable for low-cost small equipment, as it is easy to obtain the amount of received light. In the present invention, a method for simultaneous and rapid measurement of lycopene and sugar content in, for example, crushed tomato fruit (tomato fluid) and tomato processed products has been developed by measuring the diffuse reflection of visible light and near-infrared light. That is, in order to simultaneously measure the functional components, which are pigments such as lycopene, and the sugar content, the present invention proposes a method for simultaneously and rapidly measuring the functional components and sugar content in fruits and vegetables such as tomatoes and / or their processed products by simultaneously irradiating them with visible light and near-infrared light, measuring the diffuse reflection light using a non-contact diffuse reflection measurement method to obtain the visible and near-infrared light spectrum.

[0055] Examples of functional components that can be measured by the present invention include lycopene, carotenoids, anthocyanins (polyphenols), and the like, and the measurement targets can include fruits and vegetables containing these functional components and their processed products (juice, jelly, etc.). For example, fruits and vegetables containing lycopene include tomatoes, watermelons, carrots, persimmons, pink grapefruit, and rose hips. For example, fruits and vegetables containing carotenoids include citrus fruits and melons. For example, fruits and vegetables containing anthocyanins include grapes.

[0056] In the following description, the measurement of lycopene and sugar content in tomatoes and tomato products will be described as an example, and citrus fruits, melons, and grapes will be described as other examples.

[0057] The present invention includes an estimation model creation step (S1) of creating an estimation model for estimating the lycopene content and sugar content from the spectral absorption spectrum in the visible and near-infrared wavelength range of a target fresh produce or fresh produce processed product (e.g., a tomato or a tomato processed product), and a measurement step (S2) of measuring the spectral absorption spectrum of the target fresh produce or fresh produce processed product (e.g., a tomato or a tomato processed product) and applying the measured spectral absorption spectrum in the visible and near-infrared wavelength range to the estimation model to estimate the lycopene content and sugar content.

[0058] In the estimation model creation process (S1), the spectral absorption spectrum in the visible and near-infrared wavelength range of the target fresh produce and fresh produce processed products is measured, and the measured spectral absorption spectrum in the visible and near-infrared wavelength range is subjected to multivariate analysis to create an estimation model for estimating the lycopene content and sugar content in the fresh produce and fresh produce processed products. The measured spectral absorption spectrum is subjected to data preprocessing as necessary. In the multivariate analysis, the spectral absorption spectrum in the visible and near-infrared wavelength range of fresh produce and fresh produce processed products with known lycopene content and sugar content is measured, and a multivariate analysis is performed on the measured spectral absorption spectrum in the visible and near-infrared wavelength range to create an estimation model.

[0059] In the measurement process (S2), the spectral absorption spectrum in the visible and near-infrared wavelength range is obtained for the fresh produce or fresh produce processed product for which the lycopene content and sugar content are to be measured. Data pre-processing is performed on the obtained spectral absorption spectrum in the visible and near-infrared wavelength range, if necessary. The obtained spectral absorption spectrum in the visible and near-infrared wavelength range is applied to an estimation model to estimate the lycopene content and sugar content. In this way, once an estimation model is created, it is possible to simultaneously estimate (measure) the lycopene content and sugar content with high accuracy simply by obtaining the spectral absorption spectrum in the visible and near-infrared wavelength range of the fresh produce or fresh produce processed product to be measured.

[0060] [Method of measuring functional ingredients] The functional component measuring method according to the present embodiment will be described with reference to Fig. 2. Fig. 2 is a flow chart for explaining the functional component measuring method according to the present embodiment.

[0061] As shown in FIG. 2, the functional component measuring method according to the present embodiment includes an estimation model creating step (S1) of performing multivariate analysis on the spectral absorption spectrum in the visible and near-infrared wavelength range of the target fresh produce or fresh produce processed product to create an estimation model for estimating the lycopene content and sugar content, and a step of measuring the spectral absorption spectrum in the visible and near-infrared wavelength range of the target fresh produce or fresh produce processed product, and applying the measured spectral absorption spectrum in the visible and near-infrared wavelength range to the estimation model to estimate the lycopene content. and a measurement process (S2) for estimating the sugar content.

[0062] Multivariate analysis includes regression analysis. Regression analysis creates a regression equation (calibration curve) that estimates a specific value, and can be used to estimate the lycopene content and sugar content. Regression analysis includes PLS regression analysis and multiple regression analysis, and can be used to estimate the values ​​of the lycopene content and sugar content. In the multivariate analysis of this embodiment, a case where regression analysis is used will be described, but the present invention is not limited to this, and discriminant analysis, principal component analysis, cluster analysis, etc. may also be used. Note that a model may be created by machine learning (e.g., deep learning) instead of multivariate analysis, or by incorporating machine learning into multivariate analysis.

[0063] The fruits and vegetables and processed fruit products to be measured include, for example, various fruits and vegetables and processed products containing lycopene, such as tomatoes, watermelons, carrots, persimmons, pink grapefruit, and rose hips; various fruits and vegetables and processed products containing carotenoids, such as citrus fruits and melons; and various fruits and vegetables and processed products containing anthocyanins, such as grapes.

[0064] In the estimation model creation process, first, the fruits and vegetables and processed fruits and vegetables to be measured are prepared in order to create the estimation model (step S11). The fruits and vegetables and processed fruits and vegetables to be measured are filled into a sample cell, the sample cell is set on the sample stage, and the fruits and vegetables and processed fruits and vegetables are irradiated with visible and near-infrared light from a light source. The diffuse reflected light is detected by an interaction measurement method, and the spectroscopic absorption spectrum in the visible and near-infrared wavelength range is obtained by a spectroscopic detection device (step S13). When measuring the fruits and vegetables as they are (non-destructive measurement), they are set directly on the sample stage (if they are not liquid, direct measurement is possible without using a sample cell). Here, the diffuse reflected light may include both light reflected on the sample surface and light diffusely reflected inside the sample.

[0065] Next, data pre-processing is performed on the acquired spectroscopic absorption spectrum in the visible / near-infrared wavelength range as necessary (step S14). In the data pre-processing, signal processing operations such as centering, standardization, normalization, second derivative, baseline correction, and smoothing are performed one or a combination of them.

[0066] On the other hand, actual measured values ​​of lycopene content and sugar content are obtained for the fruits and vegetables and processed fruits and vegetables products to be measured (step S12). In step S15, the pre-processed data is subjected to multivariate analysis of the spectral absorption spectrum in the visible and near-infrared wavelength range to create estimation models for estimating the lycopene content and sugar content in the fruits and vegetables and processed fruits and vegetables. Here, the estimation model for estimating the lycopene content may be created using a spectral absorption spectrum in the visible wavelength range or the visible and near-infrared wavelength range. Also, the estimation model for estimating the sugar content may be created using a spectral absorption spectrum in the near-infrared wavelength range.

[0067] When regression analysis is used as the multivariate analysis, for example, a regression equation (calibration curve) for estimating the lycopene content and sugar content from the spectroscopic absorption spectrum is created as an estimation model. For example, PLS regression analysis or multiple regression analysis can be used as the regression analysis.

[0068] Next, in the measurement process, the fruits and vegetables or processed fruits and vegetables products (measurement objects) for which the lycopene content and sugar content are to be estimated are prepared (Step S21). The measurement object fruits and vegetables or processed fruits and vegetables products are placed in a sample cell, and light including visible and near-infrared wavelength ranges is irradiated onto the fruits and vegetables or processed fruits and vegetables products from a light source, and the spectral absorption spectrum in the visible and near-infrared wavelength ranges is measured and obtained using a spectroscopic detection device that measures diffuse reflectance (Step S22).

[0069] Next, data pre-processing is performed on the acquired spectroscopic absorption spectrum in the visible and near infrared wavelength range as necessary (step S23). The data pre-processing is the same as step S14. The pre-processed spectral absorption spectrum in the visible and near-infrared wavelength range is applied to the estimation model for estimating the lycopene content and sugar content created in step S15 to estimate the lycopene content and sugar content (step S24). When estimating the lycopene content, the spectral absorption spectrum in the visible wavelength range or the visible and near-infrared wavelength range among the acquired spectral absorption spectra in the visible and near-infrared wavelength range may be applied to the estimation model for estimating the lycopene content. When estimating the sugar content, the spectral absorption spectrum in the near-infrared wavelength range among the acquired spectral absorption spectra in the visible and near-infrared wavelength range may be applied to the estimation model for estimating the sugar content.

[0070] [Functional ingredient measuring device] Next, the configuration of the functional component measuring device of the present invention will be described by taking an embodiment as an example with reference to Figures 3 and 4. Note that the functional component measuring device according to this embodiment can be suitably used in the functional component measuring method described above, but the device used in the functional component measuring method according to this embodiment is not limited to this.

[0071] Here, Fig. 3 is a schematic diagram showing an example of the external configuration of a functional component measuring device according to this embodiment. Fig. 4 is a block diagram showing an example of the configuration of a functional component measuring device according to this embodiment, and conceptually shows only parts of the configuration related to the present invention. Fig. 5 is a diagram showing an example of the configuration of a sample stage. Fig. 6 is a diagram showing an example of the configuration of a body of a sample cell. Fig. 7 is a diagram showing an example of the configuration of a lid of a sample cell. As a sample cell, for example, a CR-A50 manufactured by Konica Minolta can be used.

[0072] As shown in Figures 3 and 4, the functional ingredient measuring device 1 includes a spectroscopic absorption spectrum acquiring device 10 and a data processing device 20. The spectroscopic absorption spectrum acquiring device 10 is a device that acquires a spectroscopic absorption spectrum, and includes a light source 11, a spectroscopic detection device 12, a sample stage 13, and a sample cell 14. The data processing device 20 is a device that measures the lycopene content and sugar content of the target fruit or vegetable / fruit or vegetable processed product 16 from the spectroscopic absorption spectrum acquired by the spectroscopic absorption spectrum acquiring device 10. The data processing device 20 includes a memory 21, a control unit 23, and a calculation processing unit 24, and the measurer inputs measurement conditions and the like into the data processing device 20 using a keyboard and mouse 22.

[0073] The spectroscopic absorption spectrum acquisition device 10 includes a light source 11, a spectroscopic detection device 12, a sample stage 13, and a sample cell 14. The spectroscopic absorption spectrum acquisition device 10 is connected to the sample stage 13 via an optical fiber 15. The optical fiber 15 includes a port for irradiating light and a port for detecting light. The light source 11 is connected to the sample stage 13 via the optical fiber 15. The light source 11 is a device that irradiates light of a predetermined wavelength (for example, a wavelength including a visible / near infrared wavelength range) to the fluid of the fruit or vegetable product to be measured, which is filled in the sample cell 14, via the optical fiber 15 and the sample stage 13. As the light source 11 and the spectroscopic detection device 12, for example, a spectrophotometer K-BA100R (using an interaction measurement method) manufactured by Kubota Corporation, which is integrated with these, can be used.

[0074] The spectroscopic detection device 12 is connected to the sample stage 13 via an optical fiber 15. The spectroscopic detection device 12 is a device that receives diffuse reflected light from the fluid of the fruit or vegetable or processed fruit or vegetable product to be measured, which is input via the sample stage 13 and the optical fiber 15, acquires a spectroscopic absorption spectrum in the visible and near-infrared wavelength range, and transmits it to the data processing device 20. Note that, although the measurement is performed using the optical fiber 15 here, the present invention is not limited to this, and it is also possible to perform the measurement without using an optical fiber.

[0075] The sample stage 13 irradiates the sample cell 14 with light from the light source 11 via an optical fiber 15, and detects the flow of the fruit or vegetable products or processed fruit or vegetable products to be measured that are filled in the sample cell 14. The diffusely reflected light from the object is collected and sent to the spectroscopic detector 12 via optical fiber 15 .

[0076] For example, as shown in FIG. 5, the sample stage 13 includes a light collecting section 13-1 for collecting diffuse reflected light from the measurement object of the sample cell 14, a ring light 13-2 for irradiating the measurement object of the sample cell 14 with light from a light source, a light shielding jig 13-3, and a disturbance light removing cushion 13-4. The sample cell 14 is set on the sample stage 13, and as shown in FIG. 6, the bottom surface of the sample cell 14 includes a measurement surface of a glass plate 14-1. The measurement surface of the glass plate 14-1 can transmit visible light and near infrared light. The measurement object of the sample cell 14 is set on the measurement surface, and a lid for blocking external light as shown in FIG. 7 is placed on the sample cell 14. The measurement surface of the sample cell 14 is irradiated with light from the ring light of the sample stage 13, and the diffuse reflected light of the measurement object is collected by the light collecting section 13-1 of the sample stage 13, and the diffuse reflected light is detected. A light reflecting material (such as a mirror) may be placed on the back of the lid of the sample cell 14. Furthermore, the light shielding jig 13-3 does not necessarily have to be provided between the sample stage 13 and the sample cell 14.

[0077] Returning to Figures 3 and 4, the data processing device 20 can be configured, for example, as a personal computer, and is equipped with a memory 21, a control unit 23, and a calculation processing unit 24, and is connected to a keyboard / mouse 22, an I / O port (e.g., a USB port, etc.) 26, a display 30, etc.

[0078] The memory 21 stores the spectral absorption spectra transferred from the spectroscopic detection device 12 to the data processing device 20 and acquired by the spectral absorption spectrum acquisition unit 24-1 of the calculation processing unit 24, the actual measured values ​​of the lycopene content and sugar content of fruits and vegetables and processed fruits and vegetable products used when creating an estimation model in the estimation model creation unit 24-2 of the calculation processing unit 24, and the created estimation model 27. The actual measured values ​​of the lycopene content and sugar content of fruits and vegetables and processed fruits and vegetable products used when creating the estimation model 27 can be inputted from the keyboard and mouse 22 or the I / O port 26.

[0079] The control unit 23 can perform control operations such as turning on / off the light irradiation of the light source 11 and instructing the spectroscopic detection device 12 to start spectral detection in response to the operator's operation of the keyboard and mouse 22, and can also command the calculation processing unit 24 to perform processing.

[0080] The spectral absorption spectra of the vegetables transferred from the spectral absorption spectrum acquisition device 10 are stored in the memory 21 of the data processing device 20. When the person making the measurement issues a command to the calculation processing unit 24 to perform processing via the keyboard and mouse 22, first, the spectral absorption spectrum acquisition unit 24-1 of the calculation processing unit 24 extracts the spectral absorption spectra in the visible and near-infrared wavelength range required for analysis from the spectral absorption spectra stored in the memory 21 by data preprocessing.

[0081] The estimation model creation unit 24-2 of the calculation processing unit 24 performs multivariate analysis on the extracted spectroscopic absorption spectrum in the visible / near-infrared light wavelength range to create an estimation model. In this embodiment, the estimation model creation unit 24-2 of the calculation processing unit 24 may create, as the estimation model, a regression equation (calibration curve) created by, for example, PLS regression analysis or multiple regression analysis as the multivariate analysis.

[0082] The estimation model creation unit 24-2 of the calculation processing unit 24 may store the results of the multivariate analysis (estimation model 27) in the memory 21, output them on the display 30, or print them out via a printer (not shown).

[0083] The estimation unit 24-3 of the calculation processing unit 24 estimates the lycopene content and sugar content in fruits and vegetables and processed fruits and vegetable products by applying the extracted spectral absorption spectrum in the visible and near-infrared wavelength range to the estimation model 27 stored in the memory 21. By storing estimation models 27 for a plurality of types of fruits and vegetables and processed fruit and vegetable products in the memory 21, it becomes possible to measure the lycopene content and sugar content of a plurality of types of fruits and vegetables and processed fruit and vegetable products with a single measuring device.

[0084] [Example] An example will be described with reference to Fig. 8 to Fig. 18. In this example, "tomato fluids and processed products thereof", "citrus fluids and processed products thereof", "citrus fruits", "melon fruit (red flesh)", "tomato fruit", and "grape fruit" were used as samples. The spectroscopic absorption spectra in the visible to near-infrared wavelength range of "tomato fluids and processed products thereof", "citrus fluids and processed products thereof", "citrus fruits", "melon fruit (red flesh)", "tomato fruit", and "grape fruit" were acquired by processing in the spectroscopic absorption spectrum acquisition unit 24-1.

[0085] Then, by processing the estimation model creation unit 24-2, "tomato fluids and processed products thereof", By performing PLS regression analysis on the spectral absorption spectra in the visible and near-infrared wavelength range of "citrus fluids and its processed products," "citrus fruits," "melon fruits (red flesh)," "tomato fruits," and "grape fruits," regression equations were created for quickly estimating "lycopene content and sugar content," "carotenoid content and sugar content," "carotenoid content and sugar content," "β-carotene content," "lycopene content and sugar content," and "anthocyanin content (cyanidin equivalent)." Furthermore, the estimation unit 24-3 used the created regression equations to quickly estimate "lycopene content and sugar content" and non-destructively estimate "carotenoid content and sugar content" from the spectral absorption spectra in the visible and near-infrared wavelength range of "tomato fluids and its processed products" and "citrus fruits."

[0086] 1. Tomatoes and tomato products (1-1. Preparation of the measurement object) As samples, 44 samples of tomatoes and tomato processed products (juice, ketchup, paste) were used. Tomato fruits were crushed (tomato liquid) using a crush mill manufactured by Iwatani Corporation, and 10 g of the tomato processed products (juice, ketchup, paste) were filled directly into the sample cell 14.

[0087] (1-2. Acquisition of spectroscopic absorption spectrum) As described above, each of the prepared measurement objects was filled into the sample cell 14, the sample cell 14 was placed on the sample stage 13, and the spectral absorption spectrum in the visible / near-infrared wavelength range of 500 nm to 1000 nm (750 nm to 1000 nm is defined as the near-infrared wavelength range) was measured using the spectral absorption spectrum acquisition device 10, and the spectral absorption spectrum of each sample (measurement object) was acquired by processing in the spectral absorption spectrum acquisition unit 24-1.

[0088] (1-3. Lycopene content and sugar content measured values) The lycopene content of each sample was quantified according to the method of Ito and Horie (2009) (Reports of the Vegetable and Tea Research Institute (8), 165-173, 2009) and used as the objective variable. The sugar content of each sample was measured using a digital concentration meter DBX-55 manufactured by Atago Co., Ltd. and used as the objective variable.

[0089] (1-4. Creating an estimation model) For the 13 samples of tomato fluids and tomato processed products, PLS regression analysis was performed with the spectral absorption spectrum as the explanatory variable and the quantified lycopene or sugar content value as the objective variable. In the PLS regression analysis, R was used to comprehensively examine combinations of the measurement wavelength range and data pretreatment of the sample's spectral absorption spectrum.

[0090] (1-4-1. Data preprocessing and evaluation of rapid quantitative methods) The combination of data preprocessing was as follows: smoothing window size (wsize, number of points, odd number) was 11, the number of dimensions of the smoothing filter (forder) was 2, and the Venetian blinds method was applied for cross-validation. The method was applied to 31 samples of crushed tomato fruit and processed tomato products (samples for evaluating the rapid quantitative method) and evaluated. The following 16 patterns (2 × 4 × 2 × 1) of spectral pretreatment methods were applied.

[0091] Step 1: Whether to apply standard normal variate (SNV) Step 2: Whether to apply the Savitzky-Golay smoothing method (SGSM), first derivative (SG1D), or second derivative (SG2D) Step 3: Whether to apply auto-scaling Step 4: Make sure to apply mean-centering Raw means either no SNV processing, no SNV and Savitzky-Golay processing, no SNV and Auto-scaling processing, or no SNV, Savitzky-Golay and Auto-scaling processing.

[0092] That is, the following 16 combinations are available: (1) SNV, (2) SNV+SGSM(wsize 11pt, forder 2), (3) SNV+SG1D(wsize 11pt, forder 2), (4) SNV+SG2D(wsize 11pt, forder 2), (5) SNV+Auto-scaling, (6) SNV+SGSM(wsize 11pt, forder 2)+Auto-scaling, (7) SNV+SG1D(wsize 11pt, forder 2)+Auto-scaling, (8) SNV+SG2D(wsize 11pt, forder 2)+Auto-scaling, (9) Raw, (10) Raw+SGSM(wsize 11pt, forder 2), (11)Raw+SG1D(wsize 11pt,forder 2),(12)Raw+SG2D(wsize 11pt,forder 2),(13)Raw+Auto-scaling,(14)Raw+SGSM(wsize 11pt,forder 2)+Auto-scaling,(15)Raw+SG1D(wsize 11pt,forder 2)+Auto-scaling, (16)Raw+SG2D(wsize 11pt,forder 2)+Auto-scaling

[0093] For lycopene, four wavelength ranges (500nm-750nm, 750nm-1000nm, 500nm-1000nm, 560nm-750nm) and 16 combinations of data pretreatments (total 4×16=64) were examined. For sugar content, six wavelength ranges (500nm-750nm, 750nm-1000nm, 500nm-1000nm, 800nm-1000nm, 800nm-960nm, 850nm-930nm) and 16 combinations of spectrum pretreatments (total 6×16=96) were examined. The results are shown in descending order of R(Test). R(Test) is the correlation coefficient when evaluating the rapid quantitative method.

[0094] (1-4-2. Visible and near infrared spectrum) Fig. 8 is a diagram showing the visible / near-infrared spectroscopic absorption spectra (500 nm to 1000 nm) of a tomato fluid and a tomato processed product. In Fig. 8, the horizontal axis represents wavelength (nm) and the vertical axis represents absorbance. In the figure, the thin line represents the tomato fluid and the thick line represents the tomato processed product.

[0095] As shown in Figure 8, the visible light region of 500nm to 600nm has strong absorption, including absorption due to the J-aggregation band of lycopene (Ishigaki et al., J.Phys.Chem.B,121,8046(2017)). In other words, when the visible spectroscopic absorption spectrum of tomato fluids and tomato processed products is measured using the diffuse reflectance measurement method, the absorption band due to the J-aggregation band of lycopene is considered to be dominant. The absorption band with a peak around 670nm is absorption due to chlorophyll, and there were samples that showed a clearer peak in tomato fluids than in tomato processed products. On the other hand, absorption was relatively weak in the short wavelength near infrared light region of 750nm to 1000nm. 970 The absorption band with a peak around nm is due to water absorption.

[0096] (1-4-3. PLS regression analysis) For lycopene, the top 10 R(Test) values ​​were in the wavelength range including visible or visible / near infrared light, while the bottom 10 were in the wavelength range of 750nm to 1000nm, which does not include the visible light range. These results suggest that it may be possible to develop a highly accurate and rapid quantitative method for lycopene by using not only visible light but also information from the near infrared short wavelength range. The best result (560nm to 750nm, Raw (no data preprocessing)) is shown in Figure 9.

[0097] Fig. 9 is a diagram showing the correlation (PLS regression analysis) between the measured and estimated values ​​of the lycopene content of tomato fluids and tomato processed products. In Fig. 9, the horizontal axis shows the measured value of the lycopene content [mg / 100g], and the vertical axis shows the estimated value of the lycopene content [mg / 100g] in the raw (no data preprocessing) spectral absorption spectrum from 560nm to 750nm.

[0098] In Figure 9, the R(Cal.) (correlation coefficient between the objective variable and the estimated value when developing the rapid quantitative method) value was 0.98, the R(Test) (correlation coefficient between the objective variable and the estimated value when evaluating the rapid quantitative method) value was 0.95, and the RPD(Test) (RPD value when evaluating the rapid quantitative method) value was 2.96, indicating good measurement accuracy.

[0099] In terms of sugar content, the near-infrared wavelength range dominated the top 24 rankings in order of highest R(Test) value. The best result (750nm-1000nm, Raw+SG2D (wsize 11pt, forder 2)) is shown in Figure 10. Figure 10 shows the correlation between the measured and estimated sugar content values ​​(PLS regression analysis) of tomato fluids and tomato processed products. In Figure 10, the horizontal axis shows the measured sugar content value [Brix%], and the vertical axis shows the estimated sugar content value [Brix%] from data pre-processing (Raw+SG2D (wsize 11pt, forder 2)) of the spectroscopic absorption spectrum from 750nm-1000nm.

[0100] In FIG. 10, the R(Cal.) value was 0.99, the R(Test) value was 0.99, and the RPD(Test) value was 7.53, showing very good measurement accuracy.

[0101] From the above results, it is possible to simultaneously measure the lycopene and sugar content in a sample quickly and accurately by filling the sample cell 14 with crushed tomato fruits (tomato liquids) or tomato processed foods (juice, paste, ketchup), measuring the spectroscopic absorption spectrum of the sample by visible / near-infrared spectroscopy using a non-contact diffuse reflectance measurement method with the light receiving unit, and creating a calibration curve by multivariate analysis after examining the data preprocessing method of the measured spectroscopic absorption spectrum.

[0102] 2. Citrus fruits and processed products (2-1. Preparation of the measurement object) The samples used were 16 (carotenoids) or 15 (sugar content) samples of citrus fruits (mandarin oranges, tangold oranges) and citrus processed products (juice). The edible parts of the citrus fruits were crushed (citrus fluid) using a Silent Miller manufactured by Iwatani Corporation, and 10 g of the citrus processed products (juice) were filled directly into the sample cell 14.

[0103] (2-2. Acquisition of spectroscopic absorption spectrum) As described above, each of the prepared measurement objects was filled into the sample cell 14, the sample cell 14 was placed on the sample stage 13, and the spectral absorption spectrum was measured in the visible / near-infrared wavelength range of 500 nm to 1000 nm (750 nm to 1000 nm is defined as the near-infrared wavelength range) by the spectral absorption spectrum acquisition device 10, and the spectral absorption spectrum of each sample (measurement object) was acquired by processing in the spectral absorption spectrum acquisition unit 24-1. (The above is the same as for the tomato fluid and processed products.)

[0104] (2-3. Actual measurements of carotenoid content and sugar content) The carotenoids in each measurement target were quantified according to the method of Ito and Horie (2009) (Reports of the Vegetable and Tea Research Institute (8), 165-173, 2009) and used as the objective variable when developing the rapid quantitative method. That is, carotenoids were extracted with a mixture of diethyl ether and methanol, and the quantitative value was calculated based on the absorbance at 450 nm and the absorption coefficient of 2400. The sugar content of each measurement target was measured using a digital concentration meter DBX-55 manufactured by Atago Co., Ltd. and used as the objective variable when developing the rapid quantitative method.

[0105] (2-4. Creating an estimation model) For the measured citrus fluids and citrus processed products samples (carotenoids 16, sugar content 15), PLS regression analysis was performed with the spectroscopic absorption spectrum as the explanatory variable and the quantified carotenoid or sugar content value as the objective variable. In the PLS regression analysis, R was used to comprehensively examine combinations of measurement wavelength ranges and data pretreatment of the sample spectroscopic absorption spectra.

[0106] (2-4-1. Data preprocessing and evaluation of rapid quantitative methods) The combination of data preprocessing was as follows: smoothing window size (wsize, number of points, odd number) was 11, the number of dimensions of the smoothing filter (forder) was 2, and the Venetian blinds method was applied for cross-validation. The following 16 patterns (2 × 4 × 2 × 1) of spectral preprocessing methods were applied.

[0107] Step 1: Whether to apply standard normal variate (SNV) Step 2: Whether to apply the Savitzky-Golay smoothing method (SGSM), first derivative (SG1D), or second derivative (SG2D) Step 3: Whether to apply auto-scaling Step 4: Make sure to apply mean-centering Raw means no SNV processing, no SNV and Savitzky-Golay processing, no SNV and Auto-scaling processing, or no SNV, Savitzky-Golay and Auto-scaling processing. (The above is the same as for tomato liquids and processed products.)

[0108] That is, the following 16 combinations are available: (1) SNV, (2) SNV+SGSM(wsize 11pt, forder 2), (3) SNV+SG1D(wsize 11pt, forder 2), (4) SNV+SG2D(wsize 11pt, forder 2), (5) SNV+Auto-scaling, (6) SNV+SGSM(wsize 11pt, forder 2)+Auto-scaling, (7) SNV+SG1D(wsize 11pt, forder 2)+Auto-scaling, (8) SNV+SG2D(wsize 11pt, forder 2)+Auto-scaling, (9) Raw, (10) Raw+SGSM(wsize 11pt, forder 2), (11) Raw+SG1D(wsize 11pt, forder 2), (12) Raw+SG2D(wsize 11pt, forder 2), (13) Raw+Auto-scaling, (14) Raw+SGSM(wsize 11pt, forder 2)+Auto-scaling, (15) Raw+SG1D(wsize 11pt, forder 2)+Auto-scaling, (16) Raw+SG2D(wsize 11pt, forder 2)+Auto-scaling (The above are the same as for tomato fruits and processed products.)

[0109] For carotenoids, three wavelength ranges (500 nm to 750 nm, 750 nm to 1000 nm, and 500 nm to 1000 nm) and 16 combinations of data pretreatment (total 3 × 16 For sugar content, six wavelength ranges (500nm-750nm, 750nm-1000nm, 500nm-1000nm, 800nm-1000nm, 800nm-960nm, 850nm-930nm) and 16 combinations of spectral pretreatments (total 6 x 16 = 96) were examined. The results are shown in descending order of R(Val.). R(Val.) is the correlation coefficient in cross-validation.

[0110] (2-4-2. Visible and near infrared spectrum) Fig. 11 is a diagram showing the visible / near-infrared spectroscopic absorption spectra (500 nm to 1000 nm) of citrus fluids and citrus processed products. In Fig. 11, the horizontal axis represents wavelength (nm) and the vertical axis represents absorbance. In the figure, thin lines represent citrus fluids and thick lines represent citrus processed products.

[0111] As shown in Figure 11, absorption is strong in the visible light region of 500nm to 580nm, and is thought to include absorption due to the J-aggregation band of carotenoids (Ishigaki et al., J. Phys. Chem. B, 121, 8046 (2017)). In other words, when the visible spectroscopic absorption spectrum of citrus fluids or citrus processed products is measured using the diffuse reflectance measurement method, absorption due to the J-aggregation band of carotenoids is thought to be dominant. On the other hand, absorption was relatively weak in the near-infrared short wavelength region of 750nm to 1000nm. The absorption band with a peak around 970nm is due to absorption by water.

[0112] (2-4-3. PLS regression analysis) For carotenoids, the top 10 results in terms of the highest R(Val.) values ​​were in the visible (500-750nm, 6 results) or visible / near infrared (500-1000nm, 4 results), while the lower results were in the 750nm-1000nm wavelength range, which does not include the visible light range. These results suggest that it may be possible to develop a highly accurate and rapid quantitative method for carotenoids by using not only visible light but also near infrared short wavelength information. The best result (500nm-750nm, Raw+SG2D (wsize 11pt, forder 2)) is shown in Figure 12.

[0113] Fig. 12 shows the correlation (PLS regression analysis) between the measured and estimated carotenoid contents of citrus fluids and citrus processed products. In Fig. 12, the horizontal axis shows the measured carotenoid contents [mg / 100g], and the vertical axis shows the estimated carotenoid contents [mg / 100g] in the Raw+SG2D (wsize 11pt, Forder 2) spectral absorption spectrum from 500nm to 750nm.

[0114] In Figure 12, the R(Cal.) (correlation coefficient between the objective variable and the estimated value when developing the rapid quantitative method) value was 0.99, and the RPD(Val.) (RPD value when evaluating the rapid quantitative method) value was 10.5, indicating very good measurement accuracy.

[0115] In terms of sugar content, the near-infrared wavelength range dominated the top 17 rankings in order of highest R(Val.) value. The best result (800nm-1000nm, Raw+SG1D (wsize 11pt, forder 2)) is shown in Figure 13. Figure 13 shows the correlation (PLS regression analysis) between the measured and estimated sugar content values ​​of citrus fluids and citrus processed products. In Figure 13, the horizontal axis shows the measured sugar content value [Brix%], and the vertical axis shows the estimated sugar content value [Brix%] based on data pre-processing (Raw+SG1D (wsize 11pt, forder 2)) of the spectroscopic absorption spectrum from 800nm-1000nm.

[0116] In FIG. 13, the R(Cal.) value was 0.99 and the RPD(Val.) value was 8.72, indicating very good measurement accuracy.

[0117] From the above results, it is possible to simultaneously measure the carotenoids and sugar content in a sample quickly and accurately by filling crushed citrus fruits (citrus fluids) or citrus processed foods (juice) into the sample cell 14, measuring the spectroscopic absorption spectrum of the sample by visible / near-infrared spectroscopy using a non-contact diffuse reflectance measurement method with the light receiving part, and creating a calibration curve by multivariate analysis after examining the data pre-processing method of the measured spectroscopic absorption spectrum.

[0118] 3. Citrus fruits (3-1. Preparation of the measurement object) Sixty samples of citrus fruits (mandarin, tangerine, and tangold) were used as samples.

[0119] (3-2. Acquisition of spectroscopic absorption spectrum) The citrus fruit blossom scar (near the center of the lower part of the fruit) was placed on the sample stage 13 facing the light collecting unit 13-1, and the spectral absorption spectrum in the visible / near-infrared wavelength range of 500 nm to 1000 nm (750 nm to 1000 nm is defined as the near-infrared wavelength range) was measured using the spectral absorption spectrum acquisition device 10, and the spectral absorption spectrum of each sample (measurement target) was acquired by processing in the spectral absorption spectrum acquisition unit 24-1.

[0120] (3-3. Actual measurements of carotenoid content and sugar content) The carotenoids in each measurement target were quantified according to the method of Ito and Horie (2009) (Reports of the Vegetable and Tea Industry Research Institute (8), 165-173, 2009) and used as the objective variable. That is, carotenoids were extracted using a mixture of diethyl ether and methanol, and the quantitative value was calculated based on the absorbance at 450 nm and the absorption coefficient of 2400. The sugar content of each measurement target was measured using a digital concentration meter DBX-55 manufactured by Atago Co., Ltd., and used as the objective variable when developing the non-destructive measurement method. (The above is the same as for citrus liquids and processed products.)

[0121] (3-4. Creating an estimation model) For the 32 citrus fruit samples measured, PLS regression analysis was performed with the spectral absorption spectrum as the explanatory variable and the quantified carotenoid or sugar content value as the objective variable. In the PLS regression analysis, R was used to comprehensively examine combinations of measurement wavelength ranges and data pretreatment of the sample's spectral absorption spectrum.

[0122] (3-4-1. Data pre-processing and evaluation of non-destructive measurement methods) The combination of data preprocessing was as follows: smoothing window size (wsize, number of points, odd number) was 11, the number of dimensions of the smoothing filter (forder) was 2, and the Venetian blinds method was applied for cross-validation. The developed nondestructive measurement method was applied to 28 samples of other citrus fruits (samples for evaluating the nondestructive measurement method) and evaluated. The following 16 patterns (2 x 4 x 2 x 1) of spectral preprocessing methods were applied.

[0123] Step 1: Whether to apply standard normal variate (SNV) Step 2: Whether to apply the Savitzky-Golay smoothing method (SGSM), first derivative (SG1D), or second derivative (SG2D) Step 3: Whether to apply auto-scaling Step 4: Make sure to apply mean-centering Raw means no SNV processing, no SNV and Savitzky-Golay processing, no SNV and Auto-scaling processing, or no SNV, Savitzky-Golay and Auto-scaling processing. (The above is the same as for tomato liquids and processed products.)

[0124] That is, the following 16 combinations are available: (1) SNV, (2) SNV+SGSM(wsize 11pt, forder 2), (3) SNV+SG1D(wsize 11pt, forder 2), (4) SNV+SG2D(wsize 11pt, forder 2), (5) SNV+Auto-scaling, (6) SNV+SGSM(wsize 11pt, forder 2)+Auto-scaling, (7) SNV+SG1D(wsize 11pt, forder 2)+Auto-scaling, (8) SNV+SG2D(wsize 11pt, forder 2)+Auto-scaling, (9) Raw, (10) Raw+SGSM(wsize 11pt, forder 2), (11) Raw+SG1D(wsize 11pt, forder 2), (12) Raw+SG2D(wsize 11pt, forder 2), (13) Raw+Auto-scaling, (14) Raw+SGSM(wsize 11pt, forder 2)+Auto-scaling, (15) Raw+SG1D(wsize 11pt, forder 2)+Auto-scaling, (16) Raw+SG2D(wsize 11pt, forder 2)+Auto-scaling (The above are the same as for tomato liquids and processed products.)

[0125] For carotenoids, three wavelength ranges (500nm-750nm, 750nm-1000nm, 500nm-1000nm) and 16 combinations of data pretreatments (total 3×16=48) were considered. For sugar content, six wavelength ranges (500nm-750nm, 750nm-1000nm, 500nm-1000nm, 800nm-1000nm, 800nm-960nm, 850nm-930nm) and 16 combinations of spectrum pretreatments (total 6×16=96) were considered. The results are listed in descending order of R(Test). R(Test) is the correlation coefficient of samples for evaluation of nondestructive measurement methods.

[0126] (3-4-2. Visible and near infrared spectrum) Fig. 14 is a diagram showing the visible / near-infrared spectroscopic absorption spectrum (500 nm to 1000 nm) of citrus fruits, in which the horizontal axis represents wavelength (nm) and the vertical axis represents absorbance.

[0127] As shown in Figure 14, the visible light region around 500nm to 590nm has strong absorption, which is thought to include absorption due to the J-aggregation band of carotenoids (Ishigaki et al., J.Phys.Chem.B,121,8046(2017)). In other words, when measuring the visible spectroscopic absorption spectrum of citrus fruits using the diffuse reflectance measurement method, absorption due to the J-aggregation band of carotenoids is thought to be dominant. The absorption band with a peak around 670nm is absorption due to chlorophyll, and the absorption band with a peak around 970nm is absorption due to water. The absorbance of citrus fruits is higher than that of citrus fluids and citrus processed products (juice), but it is possible to measure the visible range and the short wavelength near-infrared range in a balanced manner.

[0128] (3-4-3. PLS regression analysis) For carotenoids, the top ten results were from the wavelength range including visible and near-infrared light (500nm-1000nm, 9), while the bottom ten results were from the wavelength range of 750nm to 1000nm, which does not include the visible light range. These results suggest that it may be possible to develop a highly accurate non-destructive measurement method for carotenoids in citrus fruits by using not only visible light but also information from the short wavelength range of near-infrared light. An example of 500nm-1000nm, Raw+SG1D(wsize 11pt, forder 2)+Auto-scaling is shown in Figure 15.

[0129] Fig. 15 is a diagram showing the correlation (PLS regression analysis) between the measured values ​​and non-destructively measured values ​​(estimated values) of the carotenoid content in citrus fruits. In Fig. 15, the horizontal axis shows the measured values ​​of the carotenoid content [mg / 100 g], and the vertical axis shows the non-destructively measured values ​​of the carotenoid content [mg / 100 g] in the spectroscopic absorption spectrum from 500 nm to 1000 nm using Raw+SG1D (wsize 11pt, forder 2)+Auto-scaling.

[0130] As shown in Figure 15 (A), the R(Cal.) (correlation coefficient between the objective variable and the nondestructive measurement value when developing the nondestructive measurement method) value was 0.97, and as shown in (B), the R(Test) (correlation coefficient between the objective variable and the nondestructive measurement value when evaluating the nondestructive measurement method) value was 0.89, and the RPD(Test) (RPD value when evaluating the nondestructive measurement method) value was 2.04, indicating good measurement accuracy.

[0131] For sugar content, the top rankings in order of highest R(Test) value were almost exclusively in the near-infrared wavelength range. The best overall result (500nm-1000nm, SNV+SG1D (wsize 11pt, forder 2)+Auto-scaling) is shown in Figure 16. Figure 16 is a diagram showing the correlation (PLS regression analysis) between the actual sugar content value and the non-destructive measurement value of the sugar content of citrus fruit. In Figure 16, the horizontal axis shows the actual sugar content value [Brix%], and the vertical axis shows the non-destructive measurement value of sugar content [Brix%] using SNV+SG1D (wsize 11pt, forder 2)+Auto-scaling of the spectroscopic absorption spectrum from 500nm to 1000nm.

[0132] In FIG. 16, the R(Cal.) value was 0.97, the R(Test) value was 0.89, and the RPD(Test) value was 2.18, showing good measurement accuracy.

[0133] Based on the above results, it is possible to measure the spectroscopic absorption spectrum of the citrus fruit flower scar (near the center of the lower part of the fruit) using visible / near-infrared spectroscopy with a non-contact diffuse reflectance light measurement method with a light receiving part, and after examining the data preprocessing method of the measured spectroscopic absorption spectrum, create a calibration curve by multivariate analysis, and simultaneously and quickly measure the carotenoids and sugar content contained in the sample non-destructively and accurately.

[0134] 4. Melon fruit (red flesh) (4-1. Preparation of the measurement object) As samples, 43 samples of melon fruit (red flesh) were used.

[0135] (4-2. Acquisition of spectroscopic absorption spectrum) The flower scar of a melon fruit (red flesh) was placed on the sample stage 13 facing the light collecting unit 13-1, and the spectral absorption spectrum in the visible / near-infrared wavelength range of 500 nm to 1000 nm (750 nm to 1000 nm is defined as the near-infrared wavelength range) was measured using the spectral absorption spectrum acquisition device 10, and the spectral absorption spectrum of each sample (measurement target) was acquired by processing in the spectral absorption spectrum acquisition unit 24-1.

[0136] (4-3. Actual measured value of β-carotene content) The β-carotene content of each measurement target was quantified according to the method of Ito and Horie (2009) (Reports of the Vegetable and Tea Industry Research Institute (8), 165-173, 2009) and the method of Ito (JARQ, 48(2), 111-120, 2014), and was used as the objective variable in the development of the non-destructive measurement method. That is, carotenoids were extracted with a mixture of diethyl ether and methanol, and quantitative values ​​were obtained using a Waters UPLC device.

[0137] (4-4. Creating an estimation model) For the 43 samples of melon fruit (red flesh) measured, PLS regression analysis was performed with the spectroscopic absorption spectrum as the explanatory variable and the quantified β-carotene as the objective variable. In the PLS regression analysis, R was used to comprehensively examine combinations of the measurement wavelength range and data pretreatment of the sample's spectroscopic absorption spectrum.

[0138] (4-4-1. Data pre-processing and evaluation of non-destructive measurement methods) The combination of data preprocessing is as follows: smoothing window size (wsize, number of points, odd number) is 11, the number of dimensions of the smoothing filter (forder) is 2, and cross The Venetian blinds method was applied for validation. The following 16 patterns (2 × 4 × 2 × 1) of spectrum preprocessing methods were applied.

[0139] Step 1: Whether to apply standard normal variate (SNV) Step 2: Whether to apply the Savitzky-Golay smoothing method (SGSM), first derivative (SG1D), or second derivative (SG2D) Step 3: Whether to apply auto-scaling Step 4: Make sure to apply mean-centering Raw means either no SNV processing, no SNV and Savitzky-Golay processing, no SNV and Auto-scaling processing, or no SNV, Savitzky-Golay and Auto-scaling processing.

[0140] That is, the following 16 combinations are available: (1) SNV, (2) SNV+SGSM(wsize 11pt, forder 2), (3) SNV+SG1D(wsize 11pt, forder 2), (4) SNV+SG2D(wsize 11pt, forder 2), (5) SNV+Auto-scaling, (6) SNV+SGSM(wsize 11pt, forder 2)+Auto-scaling, (7) SNV+SG1D(wsize 11pt, forder 2)+Auto-scaling, (8) SNV+SG2D(wsize 11pt, forder 2)+Auto-scaling, (9) Raw, (10) Raw+SGSM(wsize 11pt, forder 2), (11) Raw+SG1D(wsize 11pt, forder 2), (12) Raw+SG2D(wsize 11pt, forder 2), (13) Raw+Auto-scaling, (14) Raw+SGSM(wsize 11pt, forder 2)+Auto-scaling, (15) Raw+SG1D(wsize 11pt, forder 2)+Auto-scaling, (16) Raw+SG2D(wsize 11pt, forder 2)+Auto-scaling (The above are the same as for tomato liquids and processed products.)

[0141] For β-carotene, four wavelength ranges (500 nm–750 nm, 750 nm–992 nm, 500 nm–992 nm, and 700 nm–992 nm) and 16 combinations of data pretreatment (total 4 × 16 = 64) were considered.

[0142] (4-4-2. Visible and near infrared spectrum) Fig. 17 is a diagram showing the visible / near-infrared spectroscopic absorption spectrum (500 nm to 992 nm) of melon fruit (red flesh), in which the horizontal axis represents wavelength (nm) and the vertical axis represents absorbance.

[0143] As shown in FIG. 17, the absorption by water was very strong, peaking in the visible light region of 500 nm to 700 nm and around 970 nm.

[0144] (4-4-3. PLS regression analysis) For β-carotene, the top six R(Val.) values ​​were from visible and near-infrared light (700nm-992nm), while the lower half was from the visible light range only (500nm-750nm). These results suggest that it may be possible to develop a highly accurate non-destructive measurement method for β-carotene by using not only visible light but also near-infrared short wavelength light information. The best result (700nm-992nm, SNV+SG1D(wsize 11pt,forder 2)+Auto-scaling) is shown in Figure 18.

[0145] Figure 18 shows the actual measured value and non-destructive measured value (estimated value) of β-carotene content in melon fruit (red flesh). 18 is a diagram showing the correlation (PLS regression analysis) between the β-carotene content (value) and the β-carotene content (value) of ...

[0146] In Figure 18, the R(Cal.) (correlation coefficient between the objective variable when developing the nondestructive measurement method and the nondestructive measurement value (estimated value)) value was 0.91, and the RPD(Val.) (RPD value when evaluating the nondestructive measurement method) value was 1.55, indicating good measurement accuracy.

[0147] From the above results, it is possible to measure the spectroscopic absorption spectrum of the flower scar of melon fruit (red flesh) using visible / near-infrared spectroscopy with a non-contact diffuse reflectance light measurement method with a light receiving part, and after examining the data preprocessing method of the measured spectroscopic absorption spectrum, create a calibration curve by multivariate analysis, and thereby accurately, quickly and non-destructively measure the β-carotene contained in the sample.

[0148] 5. Tomato fruit (5-1. Preparation of the measurement object) As samples, 21 and 72 tomato fruit samples were used for lycopene and sugar content, respectively.

[0149] (5-2. Acquisition of spectroscopic absorption spectrum) The tomato fruit flower scar was placed on the sample stage 13 facing the light collecting unit 13-1, and the spectral absorption spectrum in the visible / near-infrared wavelength range of 500 nm to 1000 nm (750 nm to 1000 nm is defined as the near-infrared wavelength range) was measured using the spectral absorption spectrum acquisition device 10, and the spectral absorption spectrum of each sample (measurement target) was acquired by processing in the spectral absorption spectrum acquisition unit 24-1.

[0150] (5-3. Lycopene content and sugar content measured) The lycopene content of each sample was quantified according to the method of Ito and Horie (2009) (Reports of the Vegetable and Tea Research Institute (8), 165-173, 2009) and used as the objective variable when developing the non-destructive measurement method. The sugar content of each sample was measured using a digital concentration meter DBX-55 manufactured by Atago Co., Ltd. and used as the objective variable when developing the non-destructive measurement method.

[0151] (5-4. Creating an estimation model) A PLS regression analysis was performed on the 21 measured tomato fruit samples, with the spectral absorption spectrum as the explanatory variable and the quantified lycopene as the response variable. A PLS regression analysis was performed on the 72 tomato fruit samples, with the spectral absorption spectrum as the explanatory variable and the sugar content as the response variable. In the PLS regression analysis, R was used to comprehensively examine combinations of measurement wavelength ranges and data pre-processing of the sample's spectral absorption spectrum.

[0152] (5-4-1. Data pre-processing and evaluation of non-destructive measurement methods) The combination of data preprocessing was as follows: smoothing window size (wsize, number of points, odd number) was 11, the number of dimensions of the smoothing filter (forder) was 2, and the Venetian blinds method was applied for cross-validation. The following 16 patterns (2 × 4 × 2 × 1) of spectral preprocessing methods were applied.

[0153] Step 1: Whether to apply standard normal variate (SNV) Step 2: Whether to apply the Savitzky-Golay smoothing method (SGSM), first derivative (SG1D), or second derivative (SG2D) Step 3: Whether to apply auto-scaling Step 4: Make sure to apply mean-centering Raw means either no SNV processing, no SNV and Savitzky-Golay processing, no SNV and Auto-scaling processing, or no SNV, Savitzky-Golay and Auto-scaling processing.

[0154] That is, the following 16 combinations are available: (1) SNV, (2) SNV+SGSM(wsize 11pt, forder 2), (3) SNV+SG1D(wsize 11pt, forder 2), (4) SNV+SG2D(wsize 11pt, forder 2), (5) SNV+Auto-scaling, (6) SNV+SGSM(wsize 11pt, forder 2)+Auto-scaling, (7) SNV+SG1D(wsize 11pt, forder 2)+Auto-scaling, (8) SNV+SG2D(wsize 11pt, forder 2)+Auto-scaling, (9) Raw, (10) Raw+SGSM(wsize 11pt, forder 2), (11) Raw+SG1D(wsize 11pt, forder 2), (12) Raw+SG2D(wsize 11pt, forder 2), (13) Raw+Auto-scaling, (14) Raw+SGSM(wsize 11pt, forder 2)+Auto-scaling, (15) Raw+SG1D(wsize 11pt, forder 2)+Auto-scaling, (16) Raw+SG2D(wsize 11pt, forder 2)+Auto-scaling (The above are the same as for tomato liquids and processed products.)

[0155] For lycopene, there are three wavelength ranges (500nm-750nm, 750nm-1000nm, For sugar content, six wavelength ranges (500nm-750nm, 750nm-1000nm, 500nm-1000nm, 800nm-1000nm, 800nm-960nm, 850nm-930nm) and 16 combinations of spectral pretreatment (total 6×16=96) were considered.

[0156] (5-4-2. Visible and near infrared spectrum) Fig. 19 is a diagram showing the visible / near-infrared light absorption spectrum (500 nm to 1000 nm) of a tomato fruit, in which the horizontal axis represents wavelength (nm) and the vertical axis represents absorbance.

[0157] As shown in FIG. 19, the absorption was strong around 600 nm.

[0158] (5-4-3. PLS regression analysis) For lycopene, the top 11 results for the highest R(Val.) values ​​were for visible light (500nm-750nm, 3) or visible / near-infrared light (500-1000nm, 8). These results suggest that it may be possible to develop a highly accurate non-destructive measurement method for lycopene by using not only visible light but also near-infrared short wavelength information. The best result (500nm-750nm, Raw+Auto-scaling) is shown in Figure 20.

[0159] Fig. 20 is a diagram showing the correlation (PLS regression analysis) between the actual measured value and the non-destructive measured value (estimated value) of the lycopene content in tomato fruits. In Fig. 20, the horizontal axis shows the actual measured value of the lycopene content [mg / 100g], and the vertical axis shows the non-destructive measured value of the lycopene content [mg / 100g] in the Raw+Auto-scaling of the spectroscopic absorption spectrum from 500nm to 750nm.

[0160] In Figure 20, the R(Cal.) (correlation coefficient between the objective variable and the nondestructive measurement value when developing the nondestructive measurement method) value was 0.999, and the RPD(Val.) (RPD value when evaluating the nondestructive measurement method) value was 8.44, indicating very good measurement accuracy.

[0161] In terms of sugar content, the near-infrared light range dominated the top 52 rankings in terms of R(Val.) value. The best results (800nm ​​to 960nm, SNV) are shown in Figure 21.

[0162] Fig. 21 is a diagram showing the correlation (PLS regression analysis) between the actual measured value and the non-destructive measured value (estimated value) of sugar content in tomato fruit. In Fig. 21, the horizontal axis shows the actual measured value of sugar content [Brix %], and the vertical axis shows the non-destructive measured value of sugar content [Brix %] in SNV of the spectroscopic absorption spectrum from 800 nm to 960 nm.

[0163] In Figure 21, the R(Cal.) (correlation coefficient between the objective variable and the nondestructive measurement value when developing the nondestructive measurement method) value was 0.99, and the RPD(Val.) (RPD value when evaluating the nondestructive measurement method) value was 4.90, indicating good measurement accuracy.

[0164] Based on the above results, it is possible to measure the spectroscopic absorption spectrum of the flower scar of tomato fruit using visible / near-infrared spectroscopy with a non-contact diffuse reflectance light measurement method with a light-receiving part, and after examining the data pre-processing method of the measured spectroscopic absorption spectrum, create a calibration curve by multivariate analysis, thereby enabling accurate, rapid, simultaneous and non-destructive measurement of lycopene and sugar content in a sample.

[0165] 6. Grape berries (6-1. Preparation of the measurement object) As samples, 10 grape berry samples were used for anthocyanin.

[0166] (6-2. Acquisition of spectroscopic absorption spectrum) The flower scar of one grape fruit (near the center of the lower part of the fruit) was placed on the sample stage 13 facing the light collecting unit 13-1, and the spectral absorption spectrum in the visible / near-infrared wavelength range of 500 nm to 1000 nm (750 nm to 1000 nm is defined as the near-infrared wavelength range) was measured using the spectral absorption spectrum acquiring device 10, and the spectral absorption spectrum of each sample (measurement target) was acquired by processing in the spectral absorption spectrum acquiring unit 24-1.

[0167] (6-3. Measured anthocyanin content) The anthocyanins of each measurement target were extracted from the skin using a 1% hydrochloric acid / methanol (v / v) mixture, and the absorbance of the extract at 531 nm was measured. The amount of cyanidin-3-glucoside chloride equivalent per 100 g of grape berries (hereafter referred to as "cyanidin content") was calculated from the fruit weight, the measured absorbance, and the molar extinction coefficient of 18,000, and used as the objective variable in the development of the non-destructive measurement method.

[0168] (6-4. Creating an estimation model) For the 10 grape berry samples measured, PLS regression analysis was performed with the spectroscopic absorption spectrum as the explanatory variable and the quantified cyanidin content as the objective variable. In the PLS regression analysis, R was used to comprehensively examine combinations of the measurement wavelength range and data pretreatment of the sample's spectroscopic absorption spectrum.

[0169] (6-4-1. Data pre-processing and evaluation of non-destructive measurement methods) The combination of data preprocessing was as follows: smoothing window size (wsize, number of points, odd number) was 11, the number of dimensions of the smoothing filter (forder) was 2, and the Venetian blinds method was applied for cross-validation. The following 16 patterns (2 × 4 × 2 × 1) of spectral preprocessing methods were applied.

[0170] Step 1: Whether to apply standard normal variate (SNV) Step 2: Whether to apply the Savitzky-Golay smoothing method (SGSM), first derivative (SG1D), or second derivative (SG2D) Step 3: Whether to apply auto-scaling Step 4: Make sure to apply mean-centering Raw means either no SNV processing, no SNV and Savitzky-Golay processing, no SNV and Auto-scaling processing, or no SNV, Savitzky-Golay and Auto-scaling processing.

[0171] That is, the following 16 combinations are available: (1) SNV, (2) SNV+SGSM(wsize 11pt, forder 2), (3) SNV+SG1D(wsize 11pt, forder 2), (4) SNV+SG2D(wsize 11pt, forder 2), (5) SNV+Auto-scaling, (6) SNV+SGSM(wsize 11pt, forder 2)+Auto-scaling, (7) SNV+SG1D(wsize 11pt, forder 2)+Auto-scaling, (8) SNV+SG2D(wsize 11pt, forder 2)+Auto-scaling, (9) Raw, (10) Raw+SGSM(wsize 11pt, forder 2), (11) Raw+SG1D(wsize 11pt, forder 2), (12) Raw+SG2D(wsize 11pt, forder 2), (13) Raw+Auto-scaling, (14) Raw+SGSM(wsize 11pt, forder 2)+Auto-scaling, (15) Raw+SG1D(wsize 11pt, forder 2)+Auto-scaling, (16) Raw+SG2D(wsize 11pt, forder 2)+Auto-scaling (The above are the same as for tomato liquids and processed products.)

[0172] For cyanidin content, three wavelength ranges (500 nm–750 nm, 750 nm–1000 nm, and 500 nm–1000 nm) and 16 combinations of data pretreatment (total 3 × 16 = 48) were considered.

[0173] (6-4-2. Visible and near infrared spectrum) Fig. 22 is a diagram showing the visible / near-infrared light absorption spectrum (500 nm to 1000 nm) of grape berries, in which the horizontal axis represents wavelength (nm) and the vertical axis represents absorbance.

[0174] As shown in FIG. 22, the absorption was strong around 560 nm.

[0175] (6-4-3. PLS regression analysis) For cyanidin content, the top five results in order of highest R(Val.) were for visible light (500nm-750nm, 2), near-infrared light (750nm-1000nm, 2) or visible-near-infrared light (500nm-1000nm, 1). These results suggest that it may be possible to develop a highly accurate non-destructive measurement method for cyanidin content by using not only visible light but also near-infrared short wavelength light information in combination. The best result (500nm-750nm, Raw+SGSM(wsize 11pt,forder 2)+Auto-scaling) is shown in Figure 23.

[0176] Fig. 23 is a diagram showing the correlation (PLS regression analysis) between the measured value and the non-destructive measurement value (estimated value) of the cyanidin content. In Fig. 23, the horizontal axis shows the measured value of the cyanidin content [mg / 100 g], and the vertical axis shows the non-destructive measurement value of the cyanidin content in Raw+SGSM (wsize 11pt, forder 2)+Auto-scaling of the spectroscopic absorption spectrum from 500 nm to 750 nm [mg / 100 g].

[0177] In Figure 23, R(Cal.) (the correlation between the objective variable and the nondestructive measurement value when developing the nondestructive measurement method) The value of RPD(Val.) (RPD value when evaluated by non-destructive measurement method) was 18.2, which showed very good measurement accuracy.

[0178] Based on the above results, it is possible to measure the spectroscopic absorption spectrum of the grape berry flower scar (near the center of the lower part of the fruit) using visible / near-infrared spectroscopy with a non-contact diffuse reflectance light measurement method with a light receiving part, and after examining the data preprocessing method of the measured spectroscopic absorption spectrum, create a calibration curve by multivariate analysis, and thereby accurately, quickly and non-destructively measure the cyanidin content in a sample.

[0179] [Overall summary] Crushed fruit or processed food was filled into the sample cell 14, and the spectroscopic absorption spectrum was measured by the visible-near infrared spectroscopy method using the diffuse reflection light measurement method without contact with the light receiving unit. As a result, the absorbance was lower than that in the case of non-destructive measurement of the fruit. This makes it easier to ensure high-precision quantification. A non-destructive measurement method for lycopene in tomato fruit was developed by placing the fruit equator on the sample stage facing the light collecting unit 13-1 and measuring the spectroscopic absorption spectrum (Ito, Morimoto, Journal of the Illuminating Engineering Society of Japan, 93, 510-513, 2009). In the non-destructive measurement of the flower scar of citrus fruit (near the center of the lower part of the fruit), the absorbance is generally high, but it is possible to measure the visible range and the short wavelength near-infrared range in a well-balanced manner. If the non-destructive measurement part of the fruit is the equator, many measurement parts will be generated, but in this embodiment, the flower scar (near the center of the lower part of the fruit) is used, so the measurement part is one place, and the trouble of targeting multiple measurement parts as in the equator part does not occur. In melon (red flesh), the absorbance is generally high, especially in the visible wavelength range, so that the nondestructive measurement accuracy of carotenoids (β-carotene) tends to be lower in the 500-750 nm wavelength range compared to 700-992 nm. In this way, when the absorption in the visible light range is generally strong and quantitativeness is not easily reflected, it is possible to obtain a certain degree of nondestructive measurement accuracy by applying the long visible wavelength range and the short near infrared wavelength range (700-992 nm) as explanatory variables. In citrus fruits, an example was shown as carotenoids, but if the content of β-cryptoxanthin in carotenoids is almost constant, it can be used as a rapid quantitative or nondestructive measurement method for β-cryptoxanthin. As described above, it is possible to accurately and simultaneously measure functional components and sugar content contained in a sample quickly or nondestructively without changing the measurement probe.

[0180] [Other embodiments] While the embodiments of the present invention have been described above, the present invention may be embodied in various different embodiments other than those described above within the scope of the technical ideas set forth in the claims.

[0181] Furthermore, among the various processes described in the embodiments, all or part of the processes described as being performed automatically may be performed manually, or all or part of the processes described as being performed manually may be performed automatically using known methods.

[0182] In addition, the processing procedures, control procedures, specific names, registered data for each process, information including parameters such as search conditions, screen examples, and database configurations shown in the above documents and drawings can be changed as desired unless otherwise specified.

[0183] Moreover, with regard to the functional component measuring device 1, each of the illustrated components is a functional concept, and does not necessarily have to be physically configured as shown in the drawings.

[0184] For example, the processing functions of each device of the functional component measuring device 1, particularly the processing functions performed by the calculation processing unit 24, are realized in whole or in part by a CPU (Central Processing Unit) and a program interpreted and executed by the CPU. The program may be implemented as a program for measuring a functional component by a computer or may be implemented as hardware using wired logic. The program is recorded on a recording medium described later, and is mechanically read by the functional component measuring device 1 as necessary. That is, a memory 21 such as a ROM or HD has recorded therein a computer program that cooperates as an OS (Operating System) to give instructions to the CPU and perform various processes. This computer program is executed by being loaded into a RAM, and cooperates with the CPU to constitute a control unit.

[0185] In addition, this computer program may be stored in an application program server connected to the functional component measuring device 1 via an arbitrary network, and it is also possible to download all or a part of it as necessary.

[0186] The program according to the present invention may be stored in a computer-readable recording medium, or may be configured as a program product. Here, the term "recording medium" includes any "portable physical medium" such as a memory card, a USB memory, an SD card, a flexible disk, a magneto-optical disk, a ROM, an EPROM, an EEPROM, a CD-ROM, an MO, a DVD, and a Blu-ray (registered trademark) Disc.

[0187] Moreover, a "program" is a data processing method written in any language or description method, and may be in any format, such as source code or binary code. Note that a "program" is not necessarily limited to a single configuration, but also includes a distributed configuration as multiple modules or libraries, and a program that achieves its function by working with a separate program, such as an OS (Operating System). Note that the specific configuration and reading procedure for reading a recording medium in each device shown in the embodiments, or the installation procedure after reading, may use well-known configurations and procedures.

[0188] The various databases etc. stored in memory 21 are storage means such as memory devices such as RAM, ROM, fixed disk devices such as hard disks, flexible disks, optical disks, etc., and store various programs, tables, databases, web page files, etc. used for various processes and providing websites.

[0189] The functional component measuring device 1 may be configured as an information processing device such as a known personal computer or a workstation, or may be configured by connecting any peripheral device to the information processing device. The functional component measuring device 1 may be realized by installing software (including programs, data, etc.) that realizes the method of the present invention in the information processing device.

[0190] Furthermore, the specific form of distribution and integration of the devices is not limited to that shown in the drawings, and all or part of them can be functionally or physically distributed and integrated in any unit according to various additions, etc., or according to functional loads. In other words, the above-mentioned embodiments can be implemented in any combination, or the embodiments can be implemented selectively. [Explanation of symbols]

[0191] 1. Functional ingredient measuring device 10. Spectrophotometric absorption spectrum acquisition device 11 Light source 12 Spectroscopic detection device 13 Sample stage 14 Sample cell 15. Fiber Optics 20 Data Processing Device 21 Memory 22 Keyboard and Mouse 23 Control Unit 24 Calculation Processing Unit 24-1 Spectroscopic absorption spectrum acquisition unit 24-2 Estimation model creation section 24-3 Estimation section 26 I / O ports 27 Estimation model 30 Display

Claims

[Claim 1] A functional component measurement method for measuring the content and sugar content of functional components in fruits and vegetables and / or processed fruits and vegetables, comprising: a spectral absorption spectrum acquisition step of irradiating light from a light source onto the fruit or vegetable and / or processed fruit or vegetable product to be measured and detecting the diffuse reflected light by an interaction measurement method to acquire a spectral absorption spectrum in the visible and near-infrared wavelength range; an estimation model creation step of creating an estimation model for estimating the content of a functional component and an estimation model for estimating a sugar content by performing a multivariate analysis on the spectral absorption spectrum in the visible / near-infrared wavelength range acquired in the spectral absorption spectrum acquisition step; A functional component measuring method comprising:

Images