JP2013040832A - Food discrimination apparatus and food discrimination method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique capable of discriminating whether an object to be inspected is food or not.SOLUTION: Spectroscopic information of the object to be inspected is acquired through a spectroscopic measuring instrument (step S210). Then, percentage contents of respective components occupied in the object to be inspected are calculated by analyzing the acquired spectroscopic information by using food feature spectroscopic information (step S220). Then, whether the object to be inspected is food or not is discriminated based on the calculated percentage contents of respective components (step S230). Concretely, when the total percentage content of respective components is 50% and more, the object to be inspected is discriminated as food, and when the total percentage content is less than 50%, discriminated as non-food. Finally, a discriminated result is outputted (displayed) by an output device (step S240).

Description

  The present invention relates to a technique for making a determination regarding food.

  Based on the result of measuring reflected light obtained by irradiating food with laser light including the first wavelength and the second wavelength, and comparing the intensity of the reflected light between the first wavelength and the second wavelength. In addition, a technique for detecting foreign matters mixed in the food has been proposed (Patent Document 1).

JP 2007-278846 A

  The prior art has a problem in that it is difficult to determine whether or not the object to be inspected is a food because the wavelength of the laser and the comparison method are determined for a specific food. The present invention has been made to solve this problem, and an object of the present invention is to provide a technique capable of determining whether or not an object to be inspected is food.

  SUMMARY An advantage of some aspects of the invention is to solve at least a part of the problems described above, and the invention can be implemented as the following forms or application examples.

[Application Example 1]
A food discrimination device for discriminating whether or not an object to be inspected is food,
An acquisition unit for acquiring spectral information for the inspection object;
The inspected object is analyzed by analyzing spectroscopic information for the inspected object by using constituent component spectroscopic information that is spectroscopic information corresponding to the constituent components of each of a plurality of constituent elements constituting the food. A food discrimination device comprising: a discrimination unit that discriminates whether or not the product is food.
According to this application example, the spectroscopic information for the inspected object acquired by the acquiring unit can be analyzed by the determining unit using the constituent component spectroscopic information about the constituent components constituting the food. Therefore, it can be determined whether or not the object to be inspected is food based on the component spectral information of the food. Therefore, it is possible to determine whether or not a more general inspection object is a food. The food includes food and drink. Further, “spectral information” is information indicating the absorbance of irradiated light with respect to each wavelength.

[Application Example 2]
A food discrimination apparatus according to Application Example 1,
The discrimination unit is
A calculation unit that calculates the content of each of the plurality of components included in the inspection object by analyzing the spectral information targeted for the inspection object using the component spectral information;
A food discrimination device that discriminates whether or not an object to be inspected is food based on the content rate.
According to this application example, the content of each of the plurality of components included in the inspection object is calculated by the calculation unit, and it can be determined whether the inspection object is a food based on the content ratio. Note that the “content ratio” of the component included in the inspection object means the component ratio of the spectrum of each component included in the spectral information for the inspection object.

[Application Example 3]
A food discrimination device according to Application Example 2,
The said discrimination | determination part discriminate | determines that the said to-be-inspected object is a foodstuff when the value which totaled the said content rate about the said some component exceeds a reference value.

[Application Example 4]
The food discrimination apparatus according to Application Example 2 or 3,
The said discrimination | determination part discriminate | determines that the said to-be-inspected object is a foodstuff, when all the said content rates about the said some component calculated by the said calculation part exceed a reference value. Food discrimination apparatus.

[Application Example 5]
The food discrimination apparatus according to any one of Application Examples 1 to 4,
The said discrimination | determination part analyzes the spectral information which made the said test object the object on the assumption that the vector which shows the said component spectral information is mutually orthogonal. The food discrimination apparatus.
According to this application example, the analysis can be performed simply. Note that “the component spectral information is orthogonal to each other” means that there is no correlation between certain component spectral information and other component spectral information.

[Application Example 6]
The food discrimination apparatus according to any one of Application Examples 1 to 5,
The component spectroscopic information is normalized food identification device.
According to this application example, the analysis can be performed simply. The normalization means that the average of the distribution is zero and the variance is 1.

[Application Example 7]
The food discrimination apparatus according to any one of Application Examples 1 to 6,
The plurality of constituent components are at least any two of carbohydrates, lipids, proteins, and water.

[Application Example 8]
A food discrimination method for determining whether an object to be inspected is food,
By analyzing spectral information for the object to be inspected using food characteristic spectral information that is a set of spectral information corresponding to the component for each of a plurality of components constituting the food, the object to be inspected A method for determining whether or not food is food.

The block diagram which showed schematic structure of the food discrimination apparatus. The graph which showed food characteristic spectroscopic information typically. The flowchart which shows a food discrimination process.

  FIG. 1 is a block diagram showing a schematic configuration of the food discrimination device 1. As shown in the figure, the food discrimination apparatus 1 includes a well-known desktop computer 100 and a spectroscopic instrument 200 that acquires spectroscopic information of an object to be inspected. Spectral information is information indicating the absorbance of each wavelength of irradiated light.

  The computer 100 includes a CPU 10 that performs processing described later, a storage device 20 using a RAM and a hard disk, an output device 50 using a display, an input device 60 using a keyboard and a mouse, and a hardware I / F 70 using a USB standard. Prepare. The storage device 20 stores food feature spectral information described later.

  On the other hand, the spectroscopic instrument 200 includes a light source 210 using an LED that emits light in the near infrared region to the outside, an autofocus function that adjusts the amount of light from the outside according to the aperture of the diaphragm, and forms an image of light. The optical system 220 to be refracted, the etalon 230 (Fabry-Perot filter) that selects the range of wavelengths to be transmitted with respect to the light from the optical system 220, and the light of the wavelength component transmitted through the etalon 230 is received and received. A light receiving element 240 using a CCD image sensor that outputs a signal indicating received light intensity for each position, and a process for performing a process for spectroscopic analysis based on a signal from the light receiving element 240 while controlling the other components described above. Part 250 is provided. The spectroscopic measuring instrument 200 inputs the spectroscopic information of the inspection object acquired using such a configuration to the computer 100 via the hardware I / F 70.

  As the etalon 230, for example, the configuration of an optical device disclosed in JP2009-134028A can be used. Specifically, in the spectroscopic measuring instrument 200, the processing unit 250 outputs a control signal to the etalon 230, so that the wavelength range of light passing through the etalon 230 is the near-infrared that is the target of spectral information in the present embodiment. It can be arbitrarily set with the wavelength of the region (800 nm to 2500 nm in this embodiment).

  FIG. 2 is a graph schematically showing food characteristic spectroscopic information. The food characteristic spectroscopic information is constituted by a set of constituent component spectroscopic information that is spectroscopic information corresponding to each constituent constituent of the food. FIG. 2 (A) is a protein, FIG. 2 (B) is a carbohydrate, FIG. (C) shows lipid, and FIG. 2 (D) shows water component spectral information. In the present embodiment, these four substances are referred to as “components constituting food”.

The food characteristic spectroscopic information is obtained by acquiring spectroscopic information for various foods whose content of each constituent component is known, and performing independent component analysis on the acquired spectroscopic information. That is, the food characteristic spectral information is a vector and can be expressed as food characteristic spectral information S = {s _protein , s _lipid , s _sugar , s _water } ^T. s _i (i = protein, carbohydrate, lipid, water) is each component spectroscopic information, and is a vector represented as s _i = {s _iλ1 , s _iλ2 ,..., s _iλn }. s _iλm (m = 1, 2,..., n) indicates the absorbance of the food corresponding to the wavelength λm used for spectroscopy (n is the number of wavelengths used for measurement). Each component component spectral information is normalized in advance in the wavelength region to be spectrally analyzed. Normalization means that the average of the distribution is zero and the variance is one.

  FIG. 3 is a flowchart showing the food discrimination process. This process is executed mainly by the CPU 10 when a start instruction is input from the user via the input device 60.

First, spectral information x = {x _λ1 , x _λ2 ,..., X _λn } is acquired through the spectroscopic measuring instrument 200 (step S210). x _λm (m = 1, 2,..., n) indicates the absorbance of the test object corresponding to the wavelength λm. In the present embodiment, the spectral information x is input to the computer 100 after being normalized by the spectroscopic instrument 200. However, the spectral information x may be normalized after being input to the computer 100.

  The spectral information acquired here is an average value of the spectral information acquired for a plurality of parts of the inspection object. Which part is to be subjected to spectral information acquisition is determined according to the shape and homogeneity of the object to be inspected. In the present embodiment, for example, in the case of a rectangular object to be inspected, the spectroscopic measuring instrument 200 acquires the spectral information for each of the rectangular areas obtained by equally dividing the object to be inspected by 4 × 4 and divided into 16 parts. The average value is input to the computer 100. In addition, since analysis is performed by independent component analysis, the number of parts for acquiring spectral information may be one.

Next, by using the food characteristic spectroscopic information stored in the storage device 20 and analyzing the spectroscopic information acquired above, the content ratio A = {a _protein , a _lipid , a _carbohydrate , a _water } is calculated (step S220). Specifically, it determined by A = xS ^T. This calculation projects x to S and assumes that the component spectral information s _i are orthogonal to each other. Furthermore, by normalizing x and s _i, each calculated a _i becomes a value indicating the content rate. The “content ratio” mentioned here means the composition ratio of the spectrum of each component included in the spectral information for the inspection object.

Subsequently, based on the calculated content of each component, it is determined whether or not the object to be inspected is food (whether food or non-food) (step S230). Specifically, if the total content _ai of each constituent component is 50% or more, it is determined that it is a food, and if it is less than 50%, it is not a food, that is, a non-food. Finally, the discrimination result is output (displayed) by the output device 50 (step S240), and the food discrimination process is finished.

  The reference value of 50% in step S230 is obtained by executing steps S210 and S220 for various foods and non-foods to obtain the total value of each component, and based on the acquisition results, It is a value determined as appropriate for discriminating between food and non-food.

  According to the food discrimination apparatus 1 described above, it is possible to easily discriminate whether or not the inspection object is food. In the present embodiment, since the relationship determined based on the result of spectral analysis of actual food and non-food is used as the discrimination criterion, it is possible to accurately discriminate between actual food and non-food. Therefore, for example, in a process of mass-producing a plurality of types of food, it can be used for inspection of whether foreign matters are mixed in the food. Furthermore, it can be calculated in the determination how much each component is contained in the inspection object.

  In addition, the food discrimination device 1 can be used when estimating the amount of heat of food. In other words, when spectral information is used to estimate the amount of heat of food, spectral information corresponding to non-food (such as tableware) that may be included in the acquired spectral information is identified and excluded, and the food is handled. The amount of heat of food can be accurately estimated by estimating the amount of heat using spectral information.

  The correspondence between the embodiment and the application example will be described. Step S210 corresponds to the acquisition unit, steps S220 and S230 correspond to the determination unit, and step S220 corresponds to software for realizing the calculation unit. On the other hand, the spectroscopic instrument 200 corresponds to hardware for realizing the acquisition unit and the computer 100 for realizing the determination unit and the calculation unit, respectively.

  The present invention is not limited to the above-described embodiments, and can be implemented in various forms within the technical scope of the invention. For example, additional components in the embodiment can be omitted from the embodiment. The additional components referred to here are elements corresponding to matters not specified in the substantially independent application example. Further, for example, the following embodiments can be considered.

-You may calculate the content rate of each component on the assumption that the vector which shows component component spectral information is not necessarily orthogonal to each other. In this case, the content is calculated using the A = xS ^T instead factor analysis.

-The discrimination | determination reference | standard of whether it is a foodstuff may not be the sum total of the content rate of each component. For example, a standard value is set for each component, and if all component components exceed the reference value (eg, zero%), food is used. If one or more components are less than the reference value (eg, zero%), non-food items are used. It may be distinguished. In practice, the content rate does not become a negative value, but a negative value may be calculated in calculation.
-Conversely, if the content of a certain component exceeds the reference value, it may be determined as non-food. For example, in the process of producing a dry food, if the water content exceeds a reference value, it may be determined that a foreign substance has been mixed and determined as non-food.

-Even if it is not based on content rate itself, you may discriminate | determine whether it is a food based on the value calculated from content rate. For example, a technique such as “whether the value obtained by squaring the content rate is greater than or equal to a reference value” can be considered.

-The component which comprises a foodstuff does not necessarily need to be protein, saccharide | sugar, lipid, and water. For example, two or three of these may be used, and other substances (such as minerals and vitamins) may be added.
-The wavelength region of the spectral information may be visible light. In this case, accurate color measurement can be performed.
・ When the spectral information wavelength region is set to the mid-infrared region, the content rate is calculated for more subdivided substances (such as starch, which is a type of carbohydrate) for proteins, carbohydrates, and lipids. can do. You may determine whether it is a food based on the content rate of such a subdivided substance.

-Various hardware configurations are possible. The computer and the spectroscopic instrument may be connected by wireless or LAN, or the computer and spectroscopic instrument may be configured integrally.
-You may comprise apparatuses, such as a smart phone and a digital camera, so that a food discrimination process can be performed. In this case, for example, a configuration in which a spectroscopic accessory (etalon or the like) is attached to these devices is conceivable.
In the embodiment, the light to be received is transmitted through the etalon for the purpose of spectroscopy, but the spectrum may be realized by sequentially switching the wavelength of the light emitted from the light source. For this purpose, light emitted from the light source may be transmitted through a wavelength selection filter such as an etalon, or a plurality of light sources having different wavelength characteristics may be prepared.
-In spectroscopic analysis, you may remove the influence by disturbance light. Specifically, it can be realized by measuring a spectrum generated by disturbance light without irradiating light from a light source, and subtracting the measurement result from a measurement result obtained by irradiating light from the light source. .
In the embodiment, the spectral information is acquired for a plurality of locations, but the spectral information is acquired only for one location as a representative point of the object to be inspected to determine whether it is food. You can go.

1 ... Food discrimination device 10 ... CPU
20 ... Storage device 50 ... Output device 60 ... Input device 70 ... Hardware I / F
DESCRIPTION OF SYMBOLS 100 ... Personal computer 200 ... Spectrometer 210 ... Light source 220 ... Optical system 230 ... Etalon 240 ... Light receiving element 250 ... Processing part

Claims (8)

A food discrimination device for discriminating whether or not an object to be inspected is food,
An acquisition unit for acquiring spectral information for the inspection object;
The inspected object is analyzed by analyzing spectroscopic information for the inspected object by using constituent component spectroscopic information that is spectroscopic information corresponding to the constituent components of each of a plurality of constituent elements constituting the food. A food discrimination device comprising: a discrimination unit that discriminates whether or not the product is food. The food discrimination device according to claim 1,
The discrimination unit is
A calculation unit that calculates the content of each of the plurality of components included in the inspection object by analyzing the spectral information targeted for the inspection object using the component spectral information;
A food discrimination device that discriminates whether or not an object to be inspected is food based on the content rate. The food discrimination device according to claim 2,
The said discrimination | determination part discriminate | determines that the said to-be-inspected object is a foodstuff when the value which totaled the said content rate about the said some component exceeds a reference value. A food discrimination device according to claim 2 or 3,
The said discrimination | determination part discriminate | determines that a to-be-inspected object is a foodstuff, when all the said content rates about the said some component calculated by the said calculation part exceed a reference value. A food discrimination device according to any one of claims 1 to 4,
The said discrimination | determination part analyzes the spectral information which made the said test object the object on the assumption that the vector which shows the said component spectral information is mutually orthogonal. The food discrimination apparatus. A food discrimination device according to any one of claims 1 to 5,
The component spectroscopic information is normalized food identification device. The food discrimination device according to any one of claims 1 to 6,
The plurality of constituent components are at least any two of carbohydrates, lipids, proteins, and water. A food discrimination method for determining whether an object to be inspected is food,
By analyzing spectral information for the object to be inspected using food characteristic spectral information that is a set of spectral information corresponding to the component for each of a plurality of components constituting the food, the object to be inspected A method for determining whether or not food is food.

Images